provider,endpoint,parameter,location,type,required,default,minimum,maximum,enum,compatible_models,beta_header,status,description,source
anthropic,GET /v1/organizations/analytics/apps/chat/projects,date,query,string,false,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get project activity for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive); only valid with `starting_date`. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day), so this can be at most today — which is also the default when omitted, resolved once when the first page is served and reused for the rest of the pagination sequence. At most 366 days after `starting_date`.",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `project_id`, `rbac_group_id`, `user_id`. Value forms: `project_id` takes a tagged project id (`claude_proj_...`); `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,group_by,query,"array of ""rbac_group_id"" or ""user_id""",false,,,,rbac_group_id|user_id,,,DOCUMENTED,"Dimensions to break results out by (e.g. `group_by[]=user_id`). Supported on this endpoint: `rbac_group_id`, `user_id`. Grouped rows carry the requested dimension values as additional fields and paginate like ungrouped responses via `next_page`; an unsupported dimension returns 400. `rbac_group_id` attributes a user to every group they held at any point during each covered UTC day, so grouped rows are not an exclusive partition and can sum above org-level totals. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Number of results per page (1-1000, default 100).",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,,DOCUMENTED,"Sort direction: `asc` or `desc`. Defaults to `asc` for the endpoint's sort column and to `desc` when `order_by` names a metric (a top-N ranking). Applies to `order_by`, or to the endpoint's default sort field when `order_by` is omitted.",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,order_by,query,string,false,,,,,,,DOCUMENTED,"Sort field. Restricted to the endpoint's sort column plus its rankable metrics (metrics default to descending; a few metrics rank in date-range mode only, per the endpoint's documented orderable set).",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/apps/chat/projects,starting_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. Start of a date range (inclusive). Enables rollup mode: one row per entity aggregated over the whole range — addable counters are summed across days, and a distinct count is never summed where summing could double-count (a field's range value is recomputed exactly over the window, approximate via HLL with typical error under 2%, null, or — for the creation-event counts, whose per-day values cannot overlap — a per-day sum that is itself exact; each field's own description says which). Use either `date` or `starting_date`, not both. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.",https://platform.claude.com/docs/en/api/beta/organization/analytics/chat_projects/list
anthropic,GET /v1/organizations/analytics/artifacts,date,query,string,true,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get artifact activity for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/artifacts/list
anthropic,GET /v1/organizations/analytics/artifacts,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `artifact_type`, `is_shared`, `product`, `rbac_group_id`, `user_id`. Value forms: `artifact_type` is a canonical artifact MIME type (e.g. `text/markdown`) or `other`; `is_shared` is `true` or `false`; `product` is `chat`, `claude_code`, or `cowork` (the surfaces that create artifacts); `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/artifacts/list
anthropic,GET /v1/organizations/analytics/artifacts,group_by,query,"array of ""product"" or ""rbac_group_id"" or ""user_id""",false,,,,product|rbac_group_id|user_id,,,DOCUMENTED,"Dimensions to break results out by: `product`, `user_id` and/or `rbac_group_id`. The ungrouped artifact-type cube is finite and returned in full; grouped queries multiply the cube and paginate via `next_page`. `product` takes the values `chat`, `claude_code`, or `cowork` (the surfaces that create artifacts). `rbac_group_id` attributes a user to every group they held at any point during the requested UTC day, so grouped rows are not an exclusive partition. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/artifacts/list
anthropic,GET /v1/organizations/analytics/artifacts,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Maximum rows to return (1-1000, default 100). The ungrouped artifact-type cube is finite and returned in full; `limit` is the page size only when `group_by[]` multiplies the cube.",https://platform.claude.com/docs/en/api/beta/organization/analytics/artifacts/list
anthropic,GET /v1/organizations/analytics/artifacts,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field. Only valid with `group_by[]` — the ungrouped cube is never paginated.,https://platform.claude.com/docs/en/api/beta/organization/analytics/artifacts/list
anthropic,GET /v1/organizations/analytics/connectors,date,query,string,false,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get connector usage for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive); only valid with `starting_date`. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day), so this can be at most today — which is also the default when omitted, resolved once when the first page is served and reused for the rest of the pagination sequence. At most 366 days after `starting_date`.",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `connector_name`, `product`, `rbac_group_id`, `user_id`. Value forms: `connector_name` matches case-insensitively, a display name such as 'GitHub MCP' also matches its normalized stored form ('github'), and for rows whose `connector_name` is an opaque connector id the connector's display name (`connector_display_name`) also matches; `product` is one of `chat`, `claude_code`, `cowork`, or `office_agent`; `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,group_by,query,"array of ""product"" or ""rbac_group_id"" or ""user_id""",false,,,,product|rbac_group_id|user_id,,,DOCUMENTED,"Dimensions to break results out by (e.g. `group_by[]=user_id`). Supported on this endpoint: `product`, `rbac_group_id`, `user_id`. Grouped rows carry the requested dimension values as additional fields and paginate like ungrouped responses via `next_page`; an unsupported dimension returns 400. `rbac_group_id` attributes a user to every group they held at any point during each covered UTC day, so grouped rows are not an exclusive partition and can sum above org-level totals. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Number of results per page (1-1000, default 100).",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,,DOCUMENTED,"Sort direction: `asc` or `desc`. Defaults to `asc` for the endpoint's sort column and to `desc` when `order_by` names a metric (a top-N ranking). Applies to `order_by`, or to the endpoint's default sort field when `order_by` is omitted.",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,order_by,query,string,false,,,,,,,DOCUMENTED,"Sort field. Restricted to the endpoint's sort column plus its rankable metrics (metrics default to descending; a few metrics rank in date-range mode only, per the endpoint's documented orderable set).",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/connectors,starting_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. Start of a date range (inclusive). Enables rollup mode: one row per entity aggregated over the whole range — addable counters are summed across days, and a distinct count is never summed where summing could double-count (a field's range value is recomputed exactly over the window, approximate via HLL with typical error under 2%, null, or — for the creation-event counts, whose per-day values cannot overlap — a per-day sum that is itself exact; each field's own description says which). Use either `date` or `starting_date`, not both. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.",https://platform.claude.com/docs/en/api/beta/organization/analytics/connectors/list
anthropic,GET /v1/organizations/analytics/cost_report,starting_at,query,string,true,,,,,,,DOCUMENTED,"Start of range, inclusive. RFC 3339 tz-aware. Must be within the last 365 days and no earlier than 2026-01-01T00:00:00Z.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,bucket_width,query,"""1d"" or ""1h"" or ""1m""",false,1d,,,1d|1h|1m,,,DOCUMENTED,Time bucket granularity.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,claude_tag_categories,query,"array of ""dm"" or ""engaged"" or ""monitoring"" or 2 more",false,,,,dm|engaged|monitoring|proactive|scheduled,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage in specific spend categories. Usage with no category never matches. `dm` usage is reported under the user's product rather than `claude-tag`, so combining this filter with `products[]=claude-tag` excludes it. Use `group_by[]=claude_tag_category` to break out per-category values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,claude_tag_user_ids,query,array of string,false,,,,,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage attributed to specific Slack users, by Slack user ID (for example `U0123ABCDEF`), not claude.ai user ID. Usage that is not Claude Tag, and Claude Tag usage not attributed to a single user, never matches. Use `group_by[]=claude_tag_user_id` to break out per-user values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,context_windows,query,"array of ""0-200k"" or ""200k-1M""",false,,,,0-200k|200k-1M,,,DOCUMENTED,Filter to specific context-window pricing tiers. Use `group_by[]=context_window` to break out per-tier values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,ending_at,query,string,false,,,,,,,DOCUMENTED,"End of range, exclusive. When omitted, defaults to the earlier of now and `starting_at` + 31 days. The range may span at most 31 days.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,group_by,query,"array of ""claude_tag_category"" or ""claude_tag_user_id"" or ""context_window"" or 8 more",false,,,,claude_tag_category|claude_tag_user_id|context_window|cost_type|inference_geo|model|product|rbac_group_id|slack_channel_id|speed|token_type,,,DOCUMENTED,"Dimensions to break each time bucket out by. Defaults to no grouping (one total per bucket). Each bucket reports at most its top 100 groups; a group beyond that cap has no row in that bucket (there is no remainder row), so grouped buckets are not exhaustive when a dimension has more than 100 distinct values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,inference_geos,query,"array of ""global"" or ""not_available"" or ""us""",false,,,,global|not_available|us,,,DOCUMENTED,Filter to specific inference regions. `not_available` matches rows where the region is unset. Use `group_by[]=inference_geo` to break out per-region values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,limit,query,number,false,,1,,,,,DOCUMENTED,"Maximum number of time buckets per page. Defaults and caps vary by `bucket_width` (`1d`: default 7, max 31; `1h`: default 24, max 168; `1m`: default 60, max 256).",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,models,query,array of string,false,,,,,,,DOCUMENTED,Models to include. Defaults to all models. Use `group_by[]=model` to break out per-model values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,products,query,"array of ""chat"" or ""claude-tag"" or ""claude_code"" or 4 more",false,,,,chat|claude-tag|claude_code|claude_design|claude_in_chrome|cowork|office_agent,,,DOCUMENTED,Product surfaces to include. Defaults to all products. Use `group_by[]=product` to break out per-product values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,rbac_group_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage attributed to specific RBAC groups. Accepts tagged RBAC group IDs (`rbac_group_...`) or bare group UUIDs. A row matches when the user belonged to any of the listed groups on the (UTC) day the usage occurred; usage with no group attribution never matches.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,slack_channel_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage originating from specific Slack channels. Use `group_by[]=slack_channel_id` to break out per-channel values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,speeds,query,"array of ""fast"" or ""standard""",false,,,,fast|standard,,,DOCUMENTED,Filter to fast or standard inference mode. Use `group_by[]=speed` to break out per-mode values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/cost_report,user_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to specific users by tagged user ID.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list
anthropic,GET /v1/organizations/analytics/plugins,date,query,string,false,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get plugin usage for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive); only valid with `starting_date`. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day), so this can be at most today — which is also the default when omitted, resolved once when the first page is served and reused for the rest of the pagination sequence. At most 366 days after `starting_date`.",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `plugin_name`, `product`, `rbac_group_id`, `user_id`. Value forms: `plugin_name` matches case-insensitively; `product` is `claude_code` or `cowork` (the only surfaces with plugin attribution); `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,group_by,query,"array of ""product"" or ""rbac_group_id"" or ""user_id""",false,,,,product|rbac_group_id|user_id,,,DOCUMENTED,"Dimensions to break results out by (e.g. `group_by[]=user_id`). Supported on this endpoint: `product`, `rbac_group_id`, `user_id`. On this endpoint `product` takes the values `claude_code` or `cowork` only (the surfaces with plugin attribution). Grouped rows carry the requested dimension values as additional fields and paginate like ungrouped responses via `next_page`; an unsupported dimension returns 400. `rbac_group_id` attributes a user to every group they held at any point during each covered UTC day, so grouped rows are not an exclusive partition and can sum above org-level totals. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Number of results per page (1-1000, default 100).",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,,DOCUMENTED,"Sort direction: `asc` or `desc`. Defaults to `asc` for the endpoint's sort column and to `desc` when `order_by` names a metric (a top-N ranking). Applies to `order_by`, or to the endpoint's default sort field when `order_by` is omitted.",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,order_by,query,string,false,,,,,,,DOCUMENTED,"Sort field. Restricted to the endpoint's sort column plus its rankable metrics (metrics default to descending; a few metrics rank in date-range mode only, per the endpoint's documented orderable set).",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/plugins,starting_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. Start of a date range (inclusive). Enables rollup mode: one row per entity aggregated over the whole range — addable counters are summed across days, and a distinct count is never summed where summing could double-count (a field's range value is recomputed exactly over the window, approximate via HLL with typical error under 2%, null, or — for the creation-event counts, whose per-day values cannot overlap — a per-day sum that is itself exact; each field's own description says which). Use either `date` or `starting_date`, not both. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.",https://platform.claude.com/docs/en/api/beta/organization/analytics/plugins/list
anthropic,GET /v1/organizations/analytics/skills,date,query,string,false,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get skill usage for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive); only valid with `starting_date`. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day), so this can be at most today — which is also the default when omitted, resolved once when the first page is served and reused for the rest of the pagination sequence. At most 366 days after `starting_date`.",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `product`, `rbac_group_id`, `share_status`, `skill_name`, `user_id`. Value forms: `product` is one of `chat`, `claude_code`, `cowork`, or `office_agent`; `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `share_status` is one of `organization`, `private`, or `public`; `skill_name` matches case-insensitively; `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,group_by,query,"array of ""product"" or ""rbac_group_id"" or ""user_id""",false,,,,product|rbac_group_id|user_id,,,DOCUMENTED,"Dimensions to break results out by (e.g. `group_by[]=user_id`). Supported on this endpoint: `product`, `rbac_group_id`, `user_id`. Grouped rows carry the requested dimension values as additional fields and paginate like ungrouped responses via `next_page`; an unsupported dimension returns 400. `rbac_group_id` attributes a user to every group they held at any point during each covered UTC day, so grouped rows are not an exclusive partition and can sum above org-level totals. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Number of results per page (1-1000, default 100).",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,,DOCUMENTED,"Sort direction: `asc` or `desc`. Defaults to `asc` for the endpoint's sort column and to `desc` when `order_by` names a metric (a top-N ranking). Applies to `order_by`, or to the endpoint's default sort field when `order_by` is omitted.",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,order_by,query,string,false,,,,,,,DOCUMENTED,"Sort field. Restricted to the endpoint's sort column plus its rankable metrics (metrics default to descending; a few metrics rank in date-range mode only, per the endpoint's documented orderable set).",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/skills,starting_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. Start of a date range (inclusive). Enables rollup mode: one row per entity aggregated over the whole range — addable counters are summed across days, and a distinct count is never summed where summing could double-count (a field's range value is recomputed exactly over the window, approximate via HLL with typical error under 2%, null, or — for the creation-event counts, whose per-day values cannot overlap — a per-day sum that is itself exact; each field's own description says which). Use either `date` or `starting_date`, not both. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.",https://platform.claude.com/docs/en/api/beta/organization/analytics/skills/list
anthropic,GET /v1/organizations/analytics/summaries,starting_date,query,string,true,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. Start of the date range (inclusive). Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/retrieve_summaries
anthropic,GET /v1/organizations/analytics/summaries,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive). Data is typically available with a 1-day lag, so this can be at most today — which is also the default when omitted, making the last entry cover the most recent available day. Data may be revised by a few percent over the following days. The range may span at most 366 days.",https://platform.claude.com/docs/en/api/beta/organization/analytics/retrieve_summaries
anthropic,GET /v1/organizations/analytics/summaries,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`. Only `rbac_group_id` is supported (e.g. `filter[]=rbac_group_id:{id}`); repeat the param to OR across groups. Scopes the whole day series to members of the matching group(s), re-aggregated from member-level activity — org-wide seat/invite fields and the adoption rates derived from them are null on scoped rows. `rbac_group_id` accepts the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each UTC day (time-of-usage attribution). At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/retrieve_summaries
anthropic,GET /v1/organizations/analytics/usage_report,starting_at,query,string,true,,,,,,,DOCUMENTED,"Start of range, inclusive. RFC 3339 tz-aware. Must be within the last 365 days and no earlier than 2026-01-01T00:00:00Z.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,bucket_width,query,"""1d"" or ""1h"" or ""1m""",false,1d,,,1d|1h|1m,,,DOCUMENTED,Time bucket granularity.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,claude_tag_categories,query,"array of ""dm"" or ""engaged"" or ""monitoring"" or 2 more",false,,,,dm|engaged|monitoring|proactive|scheduled,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage in specific spend categories. Usage with no category never matches. `dm` usage is reported under the user's product rather than `claude-tag`, so combining this filter with `products[]=claude-tag` excludes it. Use `group_by[]=claude_tag_category` to break out per-category values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,claude_tag_user_ids,query,array of string,false,,,,,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage attributed to specific Slack users, by Slack user ID (for example `U0123ABCDEF`), not claude.ai user ID. Usage that is not Claude Tag, and Claude Tag usage not attributed to a single user, never matches. Use `group_by[]=claude_tag_user_id` to break out per-user values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,context_windows,query,"array of ""0-200k"" or ""200k-1M""",false,,,,0-200k|200k-1M,,,DOCUMENTED,Filter to specific context-window pricing tiers. Use `group_by[]=context_window` to break out per-tier values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,ending_at,query,string,false,,,,,,,DOCUMENTED,"End of range, exclusive. When omitted, defaults to the earlier of now and `starting_at` + 31 days. The range may span at most 31 days.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,group_by,query,"array of ""claude_tag_category"" or ""claude_tag_user_id"" or ""context_window"" or 6 more",false,,,,claude_tag_category|claude_tag_user_id|context_window|inference_geo|model|product|rbac_group_id|slack_channel_id|speed,,,DOCUMENTED,"Dimensions to break each time bucket out by. Defaults to no grouping (one total per bucket). Each bucket reports at most its top 100 groups; a group beyond that cap has no row in that bucket (there is no remainder row), so grouped buckets are not exhaustive when a dimension has more than 100 distinct values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,inference_geos,query,"array of ""global"" or ""not_available"" or ""us""",false,,,,global|not_available|us,,,DOCUMENTED,Filter to specific inference regions. `not_available` matches rows where the region is unset. Use `group_by[]=inference_geo` to break out per-region values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,limit,query,number,false,,1,,,,,DOCUMENTED,"Maximum number of time buckets per page. Defaults and caps vary by `bucket_width` (`1d`: default 7, max 31; `1h`: default 24, max 168; `1m`: default 60, max 256).",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,models,query,array of string,false,,,,,,,DOCUMENTED,Models to include. Defaults to all models. Use `group_by[]=model` to break out per-model values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,products,query,"array of ""chat"" or ""claude-tag"" or ""claude_code"" or 4 more",false,,,,chat|claude-tag|claude_code|claude_design|claude_in_chrome|cowork|office_agent,,,DOCUMENTED,Product surfaces to include. Defaults to all products. Use `group_by[]=product` to break out per-product values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,rbac_group_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage attributed to specific RBAC groups. Accepts tagged RBAC group IDs (`rbac_group_...`) or bare group UUIDs. A row matches when the user belonged to any of the listed groups on the (UTC) day the usage occurred; usage with no group attribution never matches.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,slack_channel_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage originating from specific Slack channels. Use `group_by[]=slack_channel_id` to break out per-channel values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,speeds,query,"array of ""fast"" or ""standard""",false,,,,fast|standard,,,DOCUMENTED,Filter to fast or standard inference mode. Use `group_by[]=speed` to break out per-mode values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/usage_report,user_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to specific users by tagged user ID.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list
anthropic,GET /v1/organizations/analytics/user_cost_report,starting_at,query,string,true,,,,,,,DOCUMENTED,"Start of range, inclusive. RFC 3339 tz-aware. Must be within the last 365 days and no earlier than 2026-01-01T00:00:00Z.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,bucket_width,query,"""1d"" or ""1h"" or ""1m""",false,,,,1d|1h|1m,,,DOCUMENTED,"Time-bucket granularity. When set, each row's `starting_at` and `ending_at` are populated and one actor may span several rows (one per time bucket with usage). The time bucket counts toward `limit`, so one page can return multiple rows for the same actor. `ending_at` is required when `bucket_width` is set, and with `bucket_width=""1m""` the range may span at most 24 hours. When omitted, each row aggregates the full `[starting_at, ending_at)` range.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,claude_tag_categories,query,"array of ""dm"" or ""engaged"" or ""monitoring"" or 2 more",false,,,,dm|engaged|monitoring|proactive|scheduled,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage in specific spend categories. Usage with no category never matches. `dm` usage is reported under the user's product rather than `claude-tag`, so combining this filter with `products[]=claude-tag` excludes it. Use `group_by[]=claude_tag_category` to break out per-category values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,claude_tag_user_ids,query,array of string,false,,,,,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage attributed to specific Slack users, by Slack user ID (for example `U0123ABCDEF`), not claude.ai user ID. Usage that is not Claude Tag, and Claude Tag usage not attributed to a single user, never matches. Use `group_by[]=claude_tag_user_id` to break out per-user values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,context_windows,query,"array of ""0-200k"" or ""200k-1M""",false,,,,0-200k|200k-1M,,,DOCUMENTED,Filter to specific context-window pricing tiers. Use `group_by[]=context_window` to break out per-tier values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,ending_at,query,string,false,,,,,,,DOCUMENTED,"End of range, exclusive. When omitted, defaults to the earlier of now and `starting_at` + 31 days. The range may span at most 31 days.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,exclude_deleted_users,query,boolean,false,false,,,,,,DOCUMENTED,"If true, omit rows for users who are deleted (`deleted: true`). A page may contain fewer than `limit` rows; use `has_more` and `next_page` to paginate as usual.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,group_by,query,"array of ""claude_tag_category"" or ""claude_tag_user_id"" or ""context_window"" or 8 more",false,,,,claude_tag_category|claude_tag_user_id|context_window|cost_type|inference_geo|model|product|rbac_group_id|slack_channel_id|speed|token_type,,,DOCUMENTED,"Break each actor's row out by the given dimensions. Accepts the same values as the bucketed `/cost_report` endpoint. The `product`, `model`, `context_window`, `inference_geo`, and `speed` dimensions — and the time bucket, when `bucket_width` is set — count toward `limit`. `cost_type` and `token_type` do not: `cost_type` returns one row per cost component (tokens, web search, code execution); `token_type` returns one row per token type, each with `cost_type: ""tokens""`; combining both returns the per-token-type rows plus the web-search and code-execution rows. A page can therefore contain more rows than `limit` when `cost_type` or `token_type` is requested.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,inference_geos,query,"array of ""global"" or ""not_available"" or ""us""",false,,,,global|not_available|us,,,DOCUMENTED,Filter to specific inference regions. `not_available` matches rows where the region is unset. Use `group_by[]=inference_geo` to break out per-region values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,limit,query,number,false,20,1,1000,,,,DOCUMENTED,"Number of rows per page (1-1000, default 20). One row per actor unless `group_by[]` or `bucket_width` splits an actor across rows; `cost_type`/`token_type` fan-out rows (cost endpoint only) are the exception — they do not count toward this limit, so `data` can exceed it.",https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,models,query,array of string,false,,,,,,,DOCUMENTED,Models to include. Defaults to all models. Use `group_by[]=model` to break out per-model values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,order,query,"""asc"" or ""desc""",false,desc,,,asc|desc,,,DOCUMENTED,Sort direction. Defaults to `desc`.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,order_by,query,"""amount"" or ""list_amount""",false,amount,,,amount|list_amount,,,DOCUMENTED,Metric to rank actors by. Defaults to `amount`.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,products,query,"array of ""chat"" or ""claude-tag"" or ""claude_code"" or 4 more",false,,,,chat|claude-tag|claude_code|claude_design|claude_in_chrome|cowork|office_agent,,,DOCUMENTED,Product surfaces to include. Defaults to all products.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,rbac_group_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage attributed to specific RBAC groups. Accepts tagged RBAC group IDs (`rbac_group_...`) or bare group UUIDs. A row matches when the user belonged to any of the listed groups on the (UTC) day the usage occurred; usage with no group attribution never matches.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,slack_channel_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage originating from specific Slack channels. Use `group_by[]=slack_channel_id` to break out per-channel values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,speeds,query,"array of ""fast"" or ""standard""",false,,,,fast|standard,,,DOCUMENTED,Filter to fast or standard inference mode. Use `group_by[]=speed` to break out per-mode values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_cost_report,user_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to specific users by tagged user ID.,https://platform.claude.com/docs/en/api/beta/organization/analytics/cost/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,starting_at,query,string,true,,,,,,,DOCUMENTED,"Start of range, inclusive. RFC 3339 tz-aware. Must be within the last 365 days and no earlier than 2026-01-01T00:00:00Z.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,bucket_width,query,"""1d"" or ""1h"" or ""1m""",false,,,,1d|1h|1m,,,DOCUMENTED,"Time-bucket granularity. When set, each row's `starting_at` and `ending_at` are populated and one actor may span several rows (one per time bucket with usage). The time bucket counts toward `limit`, so one page can return multiple rows for the same actor. `ending_at` is required when `bucket_width` is set, and with `bucket_width=""1m""` the range may span at most 24 hours. When omitted, each row aggregates the full `[starting_at, ending_at)` range.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,claude_tag_categories,query,"array of ""dm"" or ""engaged"" or ""monitoring"" or 2 more",false,,,,dm|engaged|monitoring|proactive|scheduled,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage in specific spend categories. Usage with no category never matches. `dm` usage is reported under the user's product rather than `claude-tag`, so combining this filter with `products[]=claude-tag` excludes it. Use `group_by[]=claude_tag_category` to break out per-category values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,claude_tag_user_ids,query,array of string,false,,,,,,,DOCUMENTED,"Filter to Claude Tag (Claude in Slack) usage attributed to specific Slack users, by Slack user ID (for example `U0123ABCDEF`), not claude.ai user ID. Usage that is not Claude Tag, and Claude Tag usage not attributed to a single user, never matches. Use `group_by[]=claude_tag_user_id` to break out per-user values.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,context_windows,query,"array of ""0-200k"" or ""200k-1M""",false,,,,0-200k|200k-1M,,,DOCUMENTED,Filter to specific context-window pricing tiers. Use `group_by[]=context_window` to break out per-tier values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,ending_at,query,string,false,,,,,,,DOCUMENTED,"End of range, exclusive. When omitted, defaults to the earlier of now and `starting_at` + 31 days. The range may span at most 31 days.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,exclude_deleted_users,query,boolean,false,false,,,,,,DOCUMENTED,"If true, omit rows for users who are deleted (`deleted: true`). A page may contain fewer than `limit` rows; use `has_more` and `next_page` to paginate as usual.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,group_by,query,"array of ""claude_tag_category"" or ""claude_tag_user_id"" or ""context_window"" or 6 more",false,,,,claude_tag_category|claude_tag_user_id|context_window|inference_geo|model|product|rbac_group_id|slack_channel_id|speed,,,DOCUMENTED,"Break each actor's row out by the given dimensions. Accepts the same values as the bucketed `/usage_report` endpoint. `limit` bounds (actor × time bucket × dimension) rows — with dimensions or `bucket_width` present, one actor may span several rows.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,inference_geos,query,"array of ""global"" or ""not_available"" or ""us""",false,,,,global|not_available|us,,,DOCUMENTED,Filter to specific inference regions. `not_available` matches rows where the region is unset. Use `group_by[]=inference_geo` to break out per-region values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,limit,query,number,false,20,1,1000,,,,DOCUMENTED,"Number of rows per page (1-1000, default 20). One row per actor unless `group_by[]` or `bucket_width` splits an actor across rows; `cost_type`/`token_type` fan-out rows (cost endpoint only) are the exception — they do not count toward this limit, so `data` can exceed it.",https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,models,query,array of string,false,,,,,,,DOCUMENTED,Models to include. Defaults to all models. Use `group_by[]=model` to break out per-model values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,order,query,"""asc"" or ""desc""",false,desc,,,asc|desc,,,DOCUMENTED,Sort direction. Defaults to `desc`.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,order_by,query,"""output_tokens"" or ""requests"" or ""total_tokens"" or ""uncached_input_tokens""",false,total_tokens,,,output_tokens|requests|total_tokens|uncached_input_tokens,,,DOCUMENTED,Metric to rank actors by. Defaults to `total_tokens`.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,products,query,"array of ""chat"" or ""claude-tag"" or ""claude_code"" or 4 more",false,,,,chat|claude-tag|claude_code|claude_design|claude_in_chrome|cowork|office_agent,,,DOCUMENTED,Product surfaces to include. Defaults to all products.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,rbac_group_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage attributed to specific RBAC groups. Accepts tagged RBAC group IDs (`rbac_group_...`) or bare group UUIDs. A row matches when the user belonged to any of the listed groups on the (UTC) day the usage occurred; usage with no group attribution never matches.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,slack_channel_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to usage originating from specific Slack channels. Use `group_by[]=slack_channel_id` to break out per-channel values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,speeds,query,"array of ""fast"" or ""standard""",false,,,,fast|standard,,,DOCUMENTED,Filter to fast or standard inference mode. Use `group_by[]=speed` to break out per-mode values.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/user_usage_report,user_ids,query,array of string,false,,,,,,,DOCUMENTED,Filter to specific users by tagged user ID.,https://platform.claude.com/docs/en/api/beta/organization/analytics/usage/list_by_user
anthropic,GET /v1/organizations/analytics/users,date,query,string,false,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. The day to get user activity for. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.,https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,ending_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. End of the date range (exclusive); only valid with `starting_date`. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day), so this can be at most today — which is also the default when omitted, resolved once when the first page is served and reused for the rest of the pagination sequence. At most 366 days after `starting_date`.",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,filter,query,array of string,false,,,,,,,DOCUMENTED,"Filters as `dimension:value`, e.g. `filter[]=rbac_group_id:{id}`. Repeat the param for OR within a dimension and across dimensions for AND. Supported dimensions on this endpoint: `project_id`, `rbac_group_id`, `user_id`. Value forms: `project_id` takes a tagged project id (`claude_proj_...`) and scopes each member's row to their claude.ai chat activity within that project (it cannot be combined with `group_by[]` or an `rbac_group_id` filter); `rbac_group_id` takes the tagged id (`rbac_group_...`, as emitted in responses and by the spend-limits API) or a bare group UUID, and matches users who held the group at any point during each covered UTC day (time-of-usage attribution); `user_id` takes a tagged user id (`user_...`), as emitted in responses. An unsupported dimension returns 400. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,group_by,query,"array of ""rbac_group_id""",false,,,,,,,DOCUMENTED,"Dimensions to break results out by (e.g. `group_by[]=rbac_group_id`). Supported on this endpoint: `rbac_group_id`. Rows are already per-member, so the one supported grouping aggregates them per RBAC group instead. Grouped rows carry the requested dimension values as additional fields and paginate like ungrouped responses via `next_page`; an unsupported dimension returns 400. `rbac_group_id` attributes a user to every group they held at any point during each covered UTC day, so grouped rows are not an exclusive partition and can sum above org-level totals. At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Number of results per page (1-1000, default 100).",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,,DOCUMENTED,"Sort direction: `asc` or `desc`. Defaults to `asc` for the endpoint's sort column and to `desc` when `order_by` names a metric (a top-N ranking). Applies to `order_by`, or to the endpoint's default sort field when `order_by` is omitted.",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,order_by,query,string,false,,,,,,,DOCUMENTED,"Sort field. Restricted to the endpoint's sort column plus its rankable metrics (metrics default to descending; a few metrics rank in date-range mode only, per the endpoint's documented orderable set).",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/analytics/users,starting_date,query,string,false,,,,,,,DOCUMENTED,"UTC date in YYYY-MM-DD format. Start of a date range (inclusive). Enables rollup mode: one row per entity aggregated over the whole range — addable counters are summed across days, and a distinct count is never summed where summing could double-count (a field's range value is recomputed exactly over the window, approximate via HLL with typical error under 2%, null, or — for the creation-event counts, whose per-day values cannot overlap — a per-day sum that is itself exact; each field's own description says which). Use either `date` or `starting_date`, not both. Data is typically available with a 1-day lag (varies by query; the error for a too-recent date names the latest available day) and may be revised by a few percent over the following days. No earlier than 2026-01-01.",https://platform.claude.com/docs/en/api/beta/organization/analytics/users/list
anthropic,GET /v1/organizations/api_keys,after_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys,before_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys,created_by_user_id,query,string,false,,,,,,,DOCUMENTED,Filter by the ID of the User who created the object.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys,status,query,"""active"" or ""archived"" or ""expired"" or ""inactive""",false,,,,active|archived|expired|inactive,,,DOCUMENTED,Filter by API key status.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys,workspace_id,query,string,false,,,,,,,DOCUMENTED,Filter by Workspace ID.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/list
anthropic,GET /v1/organizations/api_keys/{api_key_id},api_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the API key.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/retrieve
anthropic,POST /v1/organizations/api_keys/{api_key_id},api_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the API key.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/update
anthropic,POST /v1/organizations/api_keys/{api_key_id},name,body,string or null,false,,,,,,,DOCUMENTED,Name of the API key.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/update
anthropic,POST /v1/organizations/api_keys/{api_key_id},status,body,"""active"" or ""archived"" or ""inactive"" or null",false,,,,active|archived|inactive,,,DOCUMENTED,Status of the API key.,https://platform.claude.com/docs/en/api/beta/organization/api_keys/update
anthropic,POST /v1/organizations/compliance_settings,state,body,BetaComplianceSettingsStateParam,true,,,,,,,DOCUMENTED,"Desired state. Accepts the string shorthand ""enabled"" or ""disabled"" in place of the object form; the response always returns the canonical object form.",https://platform.claude.com/docs/en/api/beta/organization/compliance_settings/update
anthropic,POST /v1/organizations/compliance_settings,state.type,body,"""enabled""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/compliance_settings/update
anthropic,POST /v1/organizations/compliance_settings,state.type,body,"""disabled""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/compliance_settings/update
anthropic,GET /v1/organizations/cost_report,starting_at,query,string,true,,,,,,,DOCUMENTED,Time buckets that start on or after this RFC 3339 timestamp will be returned. Each time bucket will be snapped to the start of the minute/hour/day in UTC.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/cost_report,bucket_width,query,"""1d""",false,1d,,,,,,DOCUMENTED,Time granularity of the response data.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/cost_report,ending_at,query,string,false,,,,,,,DOCUMENTED,Time buckets that end before this RFC 3339 timestamp will be returned.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/cost_report,group_by,query,"array of ""description"" or ""workspace_id""",false,,,,description|workspace_id,,,DOCUMENTED,Group by any subset of the available options.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/cost_report,limit,query,number,false,7,1,31,,,,DOCUMENTED,Maximum number of time buckets to return in the response.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/cost_report,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/cost_report/retrieve
anthropic,GET /v1/organizations/external_keys,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/list
anthropic,GET /v1/organizations/external_keys,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/list
anthropic,POST /v1/organizations/external_keys,provider_config,body,BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfigParam,true,,,,,,,DOCUMENTED,KMS provider identity and auth coordinates.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.type,body,"""aws""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.kms_arn,body,string,true,,,,,,,DOCUMENTED,"Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.",https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.region,body,string or null,false,,,,,,,DOCUMENTED,AWS region. Derived from `kms_arn` if omitted.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.role_arn,body,string or null,false,,,,,,,DOCUMENTED,"**Deprecated** IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.",https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.type,body,"""gcp""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.key_name,body,string,true,,,,,,,DOCUMENTED,Name of the key within the vault.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.type,body,"""azure""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.tenant_id,body,string,true,,,,,,,DOCUMENTED,Azure AD tenant ID.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.vault_uri,body,string,true,,,,,,,DOCUMENTED,Key Vault data-plane URI — `https://{vault-name}.vault.azure.net` or `https://{hsm-name}.managedhsm.azure.net`.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,provider_config.client_id,body,string or null,false,,,,,,,DOCUMENTED,Azure AD application (client) ID. Omit to use Anthropic's multitenant app. Provide only if using a single-tenant app registration in the customer's directory.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,display_name,body,string or null,false,,,,,,,DOCUMENTED,Human-friendly display name.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,POST /v1/organizations/external_keys,geo,body,"""us""",false,,,,,,,DOCUMENTED,Data residency geo. Only `us` is supported.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/create
anthropic,DELETE /v1/organizations/external_keys/{external_key_id},external_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the External Key.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/delete
anthropic,GET /v1/organizations/external_keys/{external_key_id},external_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the External Key.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/retrieve
anthropic,POST /v1/organizations/external_keys/{external_key_id},external_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the External Key.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},display_name,body,string or null,false,,,,,,,DOCUMENTED,Human-friendly display name.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},geo,body,"""us"" or null",false,,,,,,,DOCUMENTED,Data residency geo. Only `us` is supported.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config,body,BetaAWSExternalKeyConfig or BetaGCPExternalKeyConfig or BetaAzureExternalKeyConfigParam or null,false,,,,,,,DOCUMENTED,KMS provider identity and auth coordinates.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.type,body,"""aws""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.kms_arn,body,string,true,,,,,,,DOCUMENTED,"Full ARN of the AWS KMS key. On Claude Platform on AWS the key must be a single-Region key in your organization's own AWS account; cross-account keys, multi-Region keys, and alias ARNs are rejected.",https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.region,body,string or null,false,,,,,,,DOCUMENTED,AWS region. Derived from `kms_arn` if omitted.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.role_arn,body,string or null,false,,,,,,,DOCUMENTED,"**Deprecated** IAM role ARN. Deprecated — Anthropic reaches the KMS key through its own intermediate role (or, on Claude Platform on AWS, with credentials AWS issues for the Workspace); this field is ignored.",https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.type,body,"""gcp""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.key_name,body,string,true,,,,,,,DOCUMENTED,Name of the key within the vault.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.type,body,"""azure""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.tenant_id,body,string,true,,,,,,,DOCUMENTED,Azure AD tenant ID.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.vault_uri,body,string,true,,,,,,,DOCUMENTED,Key Vault data-plane URI — `https://{vault-name}.vault.azure.net` or `https://{hsm-name}.managedhsm.azure.net`.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id},provider_config.client_id,body,string or null,false,,,,,,,DOCUMENTED,Azure AD application (client) ID. Omit to use Anthropic's multitenant app. Provide only if using a single-tenant app registration in the customer's directory.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/update
anthropic,POST /v1/organizations/external_keys/{external_key_id}/validate,external_key_id,path,string,true,,,,,,,DOCUMENTED,ID of the External Key.,https://platform.claude.com/docs/en/api/beta/organization/external_keys/validate
anthropic,GET /v1/organizations/federation_issuers,include_archived,query,boolean,false,false,,,,,,DOCUMENTED,Include archived resources. Defaults to false.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/list
anthropic,GET /v1/organizations/federation_issuers,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/list
anthropic,GET /v1/organizations/federation_issuers,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/list
anthropic,POST /v1/organizations/federation_issuers,issuer_url,body,string,true,,,,,,,DOCUMENTED,The `iss` claim value to match against.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,name,body,string,true,,,,,,,DOCUMENTED,"Slug identifier (lowercase, digits, hyphens). Unique within the organization; a duplicate name returns 409.",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,check_jti,body,boolean or null,false,,,,,,,DOCUMENTED,Whether the jwt-bearer exchange enforces JTI single-use (replay protection) for tokens from this issuer. Defaults to true. Applies only to assertions carrying a `jti` claim; tokens without one are accepted without single-use enforcement.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks,body,BetaJWKSDiscovery or BetaJWKSExplicitURL or BetaJWKSInline,false,,,,,,,DOCUMENTED,How signing keys are obtained. Defaults to OIDC discovery.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.type,body,"""discovery""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.ca_cert_pem,body,string or null,false,,,,,,,DOCUMENTED,Optional custom CA (PEM) for TLS verification of the JWKS fetch.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.discovery_base,body,string or null,false,,,,,,,DOCUMENTED,Set when the discovery URL differs from `issuer_url`.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.type,body,"""explicit_url""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.url,body,string,true,,,,,,,DOCUMENTED,JWKS endpoint.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.type,body,"""inline""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,jwks.keys,body,array of map[unknown],true,,,,,,,DOCUMENTED,Inline JWK objects.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,POST /v1/organizations/federation_issuers,max_jwt_lifetime_seconds,body,number or null,false,,,,,,,DOCUMENTED,"Maximum allowed iat→exp spread for assertions from this issuer (1-176400 seconds, i.e. up to 49h). Defaults to 3600 (1h). Assertions must carry both `iat` and `exp`; a missing `iat` is rejected. maximum: 176400, exclusiveMinimum: 0",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/create
anthropic,GET /v1/organizations/federation_issuers/{federation_issuer_id},federation_issuer_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation issuer.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/retrieve
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},federation_issuer_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation issuer to update.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},check_jti,body,boolean or null,false,,,,,,,DOCUMENTED,Whether the jwt-bearer exchange enforces JTI single-use (replay protection) for tokens from this issuer. Applies only to assertions carrying a `jti` claim; tokens without one are accepted without single-use enforcement.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},issuer_url,body,string or null,false,,,,,,,DOCUMENTED,"Replaces the `iss` claim value to match against. For discovery-mode issuers without a `discovery_base`, this is also the URL Anthropic fetches the OIDC discovery document and signing keys from, so changing it repoints the JWKS source. Changing the issuer URL to a well-known shared platform is rejected while any live rule under this issuer would not constrain tenant identity.",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks,body,BetaJWKSDiscovery or BetaJWKSExplicitURL or BetaJWKSInline or null,false,,,,,,,DOCUMENTED,Replaces the entire JWKS configuration.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.type,body,"""discovery""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.ca_cert_pem,body,string or null,false,,,,,,,DOCUMENTED,Optional custom CA (PEM) for TLS verification of the JWKS fetch.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.discovery_base,body,string or null,false,,,,,,,DOCUMENTED,Set when the discovery URL differs from `issuer_url`.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.type,body,"""explicit_url""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.url,body,string,true,,,,,,,DOCUMENTED,JWKS endpoint.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.type,body,"""inline""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks.keys,body,array of map[unknown],true,,,,,,,DOCUMENTED,Inline JWK objects.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},jwks_polling_disabled,body,boolean or null,false,,,,,,,DOCUMENTED,"Only `false` is accepted, to re-enable polling after the system pauses it. Polling is paused automatically; sending `true` is rejected.",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},max_jwt_lifetime_seconds,body,number or null,false,,,,,,,DOCUMENTED,"Maximum allowed iat→exp spread for assertions from this issuer (1-176400 seconds, i.e. up to 49h). Assertions must carry both `iat` and `exp`; a missing `iat` is rejected. maximum: 176400, exclusiveMinimum: 0",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id},name,body,string or null,false,,,,,,,DOCUMENTED,"Replaces the slug identifier (lowercase, digits, hyphens). Unique within the organization; a duplicate name returns 409.",https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/update
anthropic,POST /v1/organizations/federation_issuers/{federation_issuer_id}/archive,federation_issuer_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation issuer to archive.,https://platform.claude.com/docs/en/api/beta/organization/federation/issuers/archive
anthropic,GET /v1/organizations/federation_rules,include_archived,query,boolean,false,false,,,,,,DOCUMENTED,Include archived resources. Defaults to false.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/list
anthropic,GET /v1/organizations/federation_rules,issuer_id,query,string,false,,,,,,,DOCUMENTED,Filter to rules referencing this federation issuer.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/list
anthropic,GET /v1/organizations/federation_rules,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/list
anthropic,GET /v1/organizations/federation_rules,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/list
anthropic,POST /v1/organizations/federation_rules,issuer_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the federation issuer.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,match,body,BetaFederationRuleMatch,true,,,,,,,DOCUMENTED,"Conditions the verified JWT must satisfy for this rule to apply. At least one of `subject_prefix` (other than a wildcard-only value like `*`), `claims`, or `condition` is required; `audience` alone is not sufficient.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,match.audience,body,string or null,false,,,,,,,DOCUMENTED,"Exact match against the `aud` claim (any element if array). When omitted, the JWT's `aud` must still equal Anthropic's expected audience for the issuer; setting this field overrides that default.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,match.claims,body,map[string] or null,false,,,,,,,DOCUMENTED,Exact-match `{claim: value}` pairs against top-level claims. Only string-valued claims can be matched; use `condition` for non-string claims.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,match.condition,body,string or null,false,,,,,,,DOCUMENTED,CEL expression over claims for logic the structural fields can't express. Must evaluate to a boolean and may reference only the `claims` variable; a constant-true expression (such as `true`) is rejected with 400.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,match.subject_prefix,body,string or null,false,,,,,,,DOCUMENTED,"Match the verified JWT `sub` claim. Exact match unless the value ends with `*`, in which case it is a prefix match. Example: `repo:my-org/my-repo:ref:refs/heads/main`.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,name,body,string,true,,,,,,,DOCUMENTED,"Slug identifier (lowercase, digits, hyphens). Unique within the organization; a duplicate name returns 409.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,oauth_scope,body,string,true,,,,,,,DOCUMENTED,Space-separated OAuth scopes. OAuth callers may only set `workspace:developer` or `workspace:inference`; other scopes (such as `org:admin`) require a Console session.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,target,body,BetaServiceAccountTarget,true,,,,,,,DOCUMENTED,Identity that tokens minted via this rule act as. Currently always a `service_account` target.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,target.type,body,"""service_account""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,target.service_account_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the service account to mint tokens for.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,target.service_account_name,body,string or null,false,,,,,,,DOCUMENTED,Service account's display name at read time. Ignored on writes.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,applies_to_all_workspaces,body,boolean,false,,,,,,,DOCUMENTED,"When true, enable this rule for every workspace in the org (including workspaces created later).",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,attributes,body,map[string] or null,false,,,,,,,DOCUMENTED,CEL expressions `{name: expr}` extracting named values from claims. Not yet supported; any non-empty value is rejected with 400.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,description,body,string or null,false,,,,,,,DOCUMENTED,Optional free-text description.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,token_lifetime_seconds,body,number,false,,60,86400,,,,DOCUMENTED,"Lifetime in seconds for access tokens minted via this rule (60-86400). Defaults to 3600 (1h). Minted tokens are capped at `max(60, min(this value, 2 × remaining assertion validity))` seconds.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,POST /v1/organizations/federation_rules,workspace_id,body,string or null,false,,,,,,,DOCUMENTED,Tagged ID of the workspace to enable this rule for. Required unless `applies_to_all_workspaces` is true. Additional workspaces can be added via the `/federation_rules/{federation_rule_id}/workspaces` sub-resource.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/create
anthropic,GET /v1/organizations/federation_rules/{federation_rule_id},federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/retrieve
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule to update.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},applies_to_all_workspaces,body,boolean or null,false,,,,,,,DOCUMENTED,"When true, enables this rule for every workspace in the org (including workspaces created later). Setting `false` is rejected with 400 if no workspace would remain enabled; a rule with only a legacy `workspace_id` binding continues to mint.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},attributes,body,map[string] or null,false,,,,,,,DOCUMENTED,Replaces the CEL expressions `{name: expr}` extracting named values from claims. Send null to clear them. Not yet supported; any non-empty value is rejected with 400.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},description,body,string or null,false,,,,,,,DOCUMENTED,Replaces the description. Omit to leave unchanged; send `null` to clear (the field is stored as an empty string).,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},match,body,BetaFederationRuleMatch or null,false,,,,,,,DOCUMENTED,"Does the incoming JWT qualify? All populated fields must pass; omitted fields are skipped. At least one of `subject_prefix` (other than a wildcard-only value like `*`), `claims`, or `condition` is required; `audience` alone is not sufficient.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},match.audience,body,string or null,false,,,,,,,DOCUMENTED,"Exact match against the `aud` claim (any element if array). When omitted, the JWT's `aud` must still equal Anthropic's expected audience for the issuer; setting this field overrides that default.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},match.claims,body,map[string] or null,false,,,,,,,DOCUMENTED,Exact-match `{claim: value}` pairs against top-level claims. Only string-valued claims can be matched; use `condition` for non-string claims.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},match.condition,body,string or null,false,,,,,,,DOCUMENTED,CEL expression over claims for logic the structural fields can't express. Must evaluate to a boolean and may reference only the `claims` variable; a constant-true expression (such as `true`) is rejected with 400.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},match.subject_prefix,body,string or null,false,,,,,,,DOCUMENTED,"Match the verified JWT `sub` claim. Exact match unless the value ends with `*`, in which case it is a prefix match. Example: `repo:my-org/my-repo:ref:refs/heads/main`.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},name,body,string or null,false,,,,,,,DOCUMENTED,"Replaces the slug identifier (lowercase, digits, hyphens). Unique within the organization; a duplicate name returns 409.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},oauth_scope,body,string or null,false,,,,,,,DOCUMENTED,Replaces the space-separated OAuth scopes granted on minted tokens. OAuth callers may only set `workspace:developer` or `workspace:inference`; other scopes (such as `org:admin`) require a Console session.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},target,body,BetaServiceAccountTarget or null,false,,,,,,,DOCUMENTED,Bind to a fixed service account by ID.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},target.type,body,"""service_account""",true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},target.service_account_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the service account to mint tokens for.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},target.service_account_name,body,string or null,false,,,,,,,DOCUMENTED,Service account's display name at read time. Ignored on writes.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},token_lifetime_seconds,body,number or null,false,,60,86400,,,,DOCUMENTED,"Replaces the lifetime in seconds for access tokens minted via this rule (60-86400). Minted tokens are capped at `max(60, min(this value, 2 × remaining assertion validity))` seconds.",https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id},workspace_id,body,string or null,false,,,,,,,DOCUMENTED,Replaces the existing single workspace enablement (the previous one is removed). Rejected with 400 if the rule is enabled for more than one workspace; use the `/federation_rules/{federation_rule_id}/workspaces` sub-resource instead.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/update
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id}/archive,federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule to archive.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/archive
anthropic,GET /v1/organizations/federation_rules/{federation_rule_id}/workspaces,federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/list
anthropic,GET /v1/organizations/federation_rules/{federation_rule_id}/workspaces,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/list
anthropic,GET /v1/organizations/federation_rules/{federation_rule_id}/workspaces,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/list
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id}/workspaces,federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/add
anthropic,POST /v1/organizations/federation_rules/{federation_rule_id}/workspaces,workspace_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the workspace to enable this rule for.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/add
anthropic,DELETE /v1/organizations/federation_rules/{federation_rule_id}/workspaces/{workspace_id},federation_rule_id,path,string,true,,,,,,,DOCUMENTED,ID of the federation rule.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/remove
anthropic,DELETE /v1/organizations/federation_rules/{federation_rule_id}/workspaces/{workspace_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace to disable for.,https://platform.claude.com/docs/en/api/beta/organization/federation/rules/workspaces/remove
anthropic,GET /v1/organizations/invites,after_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,GET /v1/organizations/invites,before_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,GET /v1/organizations/invites,email,query,string,false,,,,,,,DOCUMENTED,"Filter by the email address the Invite was sent to. Matches the same way as the Users list's `email` filter (normalized, case-insensitive).",https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,GET /v1/organizations/invites,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,GET /v1/organizations/invites,roles,query,array of string,false,,,,,,,DOCUMENTED,"Filter to items whose `role` equals one of the supplied values. Repeatable; values are OR'ed together. Accepted values depend on the organization type: Console and API organizations accept `user`, `developer`, `billing`, `admin`, and `claude_code_user`; Claude Enterprise organizations accept `user`, `owner`, `primary_owner`, `membership_admin`, and `managed`.",https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,GET /v1/organizations/invites,statuses,query,"array of ""accepted"" or ""expired"" or ""pending""",false,,,,accepted|expired|pending,,,DOCUMENTED,"Filter by Invite status. Repeatable; values are OR'ed together. Omit to return `pending`, `accepted`, and `expired` Invites alike.",https://platform.claude.com/docs/en/api/beta/organization/invites/list
anthropic,POST /v1/organizations/invites,email,body,string,true,,,,,,,DOCUMENTED,Email of the User.,https://platform.claude.com/docs/en/api/beta/organization/invites/create
anthropic,POST /v1/organizations/invites,role,body,"""billing"" or ""claude_code_user"" or ""developer"" or 2 more",true,,,,billing|claude_code_user|developer|managed|user,,,DOCUMENTED,"Role for the invited User. The accepted values depend on the organization type. Console and API organizations accept `user`, `developer`, `billing`, and `claude_code_user`; `admin` cannot be assigned through the API. Claude Enterprise organizations accept `user` and `managed`.",https://platform.claude.com/docs/en/api/beta/organization/invites/create
anthropic,POST /v1/organizations/invites,rbac_group_ids,body,array of string,false,,,,,,,DOCUMENTED,"RBAC group IDs to assign to the User when the Invite is accepted. A non-empty array is accepted only for a Claude Enterprise organization with RBAC groups, and requires the key to carry the `write:rbac_groups` scope.",https://platform.claude.com/docs/en/api/beta/organization/invites/create
anthropic,DELETE /v1/organizations/invites/{invite_id},invite_id,path,string,true,,,,,,,DOCUMENTED,ID of the Invite.,https://platform.claude.com/docs/en/api/beta/organization/invites/delete
anthropic,GET /v1/organizations/invites/{invite_id},invite_id,path,string,true,,,,,,,DOCUMENTED,ID of the Invite.,https://platform.claude.com/docs/en/api/beta/organization/invites/retrieve
anthropic,GET /v1/organizations/rate_limits,group_type,query,"""batch"" or ""files"" or ""model_group"" or 3 more",false,,,,batch|files|model_group|skills|token_count|web_search,,,DOCUMENTED,Filter by group type.,https://platform.claude.com/docs/en/api/beta/organization/rate_limits/list
anthropic,GET /v1/organizations/rate_limits,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Maximum number of items to return per page. Ranges from `1` to `1000`. When omitted, every remaining entry is returned in a single page and `next_page` is `null`.",https://platform.claude.com/docs/en/api/beta/organization/rate_limits/list
anthropic,GET /v1/organizations/rate_limits,model,query,string,false,,,,,,,DOCUMENTED,Filter to the single entry containing this model. Accepts full model names and aliases. Returns 404 if the model is not found or has no rate limits for this organization.,https://platform.claude.com/docs/en/api/beta/organization/rate_limits/list
anthropic,GET /v1/organizations/rate_limits,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/rate_limits/list
anthropic,GET /v1/organizations/rbac_groups,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/list
anthropic,GET /v1/organizations/rbac_groups,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/list
anthropic,POST /v1/organizations/rbac_groups,name,body,string,true,,,,,,,DOCUMENTED,Name of the RBAC Group. Not uniqueness-enforced.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/create
anthropic,DELETE /v1/organizations/rbac_groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/delete
anthropic,GET /v1/organizations/rbac_groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/retrieve
anthropic,POST /v1/organizations/rbac_groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/update
anthropic,POST /v1/organizations/rbac_groups/{group_id},name,body,string or null,false,,,,,,,DOCUMENTED,Name of the RBAC Group. Not uniqueness-enforced.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/update
anthropic,GET /v1/organizations/rbac_groups/{group_id}/members,group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/list
anthropic,GET /v1/organizations/rbac_groups/{group_id}/members,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/list
anthropic,GET /v1/organizations/rbac_groups/{group_id}/members,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/list
anthropic,POST /v1/organizations/rbac_groups/{group_id}/members,group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/create
anthropic,POST /v1/organizations/rbac_groups/{group_id}/members,user_id,body,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/create
anthropic,DELETE /v1/organizations/rbac_groups/{group_id}/members/{user_id},group_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Group.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/delete
anthropic,DELETE /v1/organizations/rbac_groups/{group_id}/members/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/rbac_groups/members/delete
anthropic,GET /v1/organizations/rbac_roles,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/list
anthropic,GET /v1/organizations/rbac_roles,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/list
anthropic,GET /v1/organizations/rbac_roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Role.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/retrieve
anthropic,GET /v1/organizations/rbac_roles/{role_id}/permissions,role_id,path,string,true,,,,,,,DOCUMENTED,ID of the RBAC Role.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/permissions/list
anthropic,GET /v1/organizations/rbac_roles/{role_id}/permissions,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/permissions/list
anthropic,GET /v1/organizations/rbac_roles/{role_id}/permissions,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/rbac_roles/permissions/list
anthropic,GET /v1/organizations/service_accounts,include_archived,query,boolean,false,false,,,,,,DOCUMENTED,Include archived resources. Defaults to false.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/list
anthropic,GET /v1/organizations/service_accounts,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/list
anthropic,GET /v1/organizations/service_accounts,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/list
anthropic,POST /v1/organizations/service_accounts,name,body,string,true,,,,,,,DOCUMENTED,"Slug identifier (lowercase, digits, hyphens). Unique within the organization; a duplicate name returns 409.",https://platform.claude.com/docs/en/api/beta/organization/service_accounts/create
anthropic,POST /v1/organizations/service_accounts,description,body,string or null,false,,,,,,,DOCUMENTED,Optional free-text description.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/create
anthropic,POST /v1/organizations/service_accounts,organization_role,body,"""admin"" or ""developer""",false,,,,admin|developer,,,DOCUMENTED,Org-level role. Defaults to `developer`.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/create
anthropic,GET /v1/organizations/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/retrieve
anthropic,POST /v1/organizations/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account to update.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/update
anthropic,POST /v1/organizations/service_accounts/{service_account_id},description,body,string or null,false,,,,,,,DOCUMENTED,Replaces the description. Omit to leave unchanged; send `null` to clear (the field is stored as an empty string).,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/update
anthropic,POST /v1/organizations/service_accounts/{service_account_id},organization_role,body,"""admin"" or ""developer"" or null",false,,,,admin|developer,,,DOCUMENTED,Replaces the org-level role. Omit or send `null` to leave unchanged.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/update
anthropic,POST /v1/organizations/service_accounts/{service_account_id}/archive,service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account to archive.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/archive
anthropic,GET /v1/organizations/service_accounts/{service_account_id}/workspaces,service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/list
anthropic,GET /v1/organizations/service_accounts/{service_account_id}/workspaces,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/list
anthropic,GET /v1/organizations/service_accounts/{service_account_id}/workspaces,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/list
anthropic,POST /v1/organizations/service_accounts/{service_account_id}/workspaces,service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/add
anthropic,POST /v1/organizations/service_accounts/{service_account_id}/workspaces,workspace_id,body,string,true,,,,,,,DOCUMENTED,Tagged workspace ID to add the service account to.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/add
anthropic,POST /v1/organizations/service_accounts/{service_account_id}/workspaces,workspace_role,body,BetaNoBillingWorkspaceRole,true,,,,workspace_admin|workspace_developer|workspace_restricted_developer|workspace_user,,,DOCUMENTED,Role to assign to the service account in this workspace.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/add
anthropic,DELETE /v1/organizations/service_accounts/{service_account_id}/workspaces/{workspace_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/remove
anthropic,DELETE /v1/organizations/service_accounts/{service_account_id}/workspaces/{workspace_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/service_accounts/workspaces/remove
anthropic,GET /v1/organizations/spend_limit_increase_requests,actor_ids,query,array of string,false,,,,,,,DOCUMENTED,"Filter by requester, as `user_...` tagged IDs.",https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/list
anthropic,GET /v1/organizations/spend_limit_increase_requests,limit,query,number,false,20,1,1000,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/list
anthropic,GET /v1/organizations/spend_limit_increase_requests,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/list
anthropic,GET /v1/organizations/spend_limit_increase_requests,status,query,"array of ""approved"" or ""denied"" or ""pending""",false,,,,approved|denied|pending,,,DOCUMENTED,Filter by status. Omit to return all.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/list
anthropic,GET /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id},spend_limit_increase_request_id,path,string,true,,,,,,,DOCUMENTED,ID of the spend limit increase request.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/retrieve
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/approve,spend_limit_increase_request_id,path,string,true,,,,,,,DOCUMENTED,ID of the spend limit increase request.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/approve
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/approve,amount,body,string,true,,,,,,,DOCUMENTED,New per-user spend limit as a non-negative integer decimal string (minor units).,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/approve
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/approve,period,body,"""daily"" or ""monthly"" or ""weekly"" or null",false,,,,daily|monthly|weekly,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/approve
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/approve,suppress_notification,body,boolean,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/approve
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/deny,spend_limit_increase_request_id,path,string,true,,,,,,,DOCUMENTED,ID of the spend limit increase request.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/deny
anthropic,POST /v1/organizations/spend_limit_increase_requests/{spend_limit_increase_request_id}/deny,suppress_notification,body,boolean,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/increase_requests/deny
anthropic,POST /v1/organizations/spend_limits,amount,body,string or null,true,,,,,,,DOCUMENTED,"Limit amount as a non-negative integer decimal string in the minor unit of the organization's billing currency (cents for USD): ""50000"" is $500.00. `null` sets an explicit no-limit override for this scope and `period` only — each period resolves independently, so caps for other periods still apply.",https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope,body,User or Organization or Workspace,true,,,,,,,DOCUMENTED,"What the limit applies to. Claude Enterprise organizations set `user` limits. Claude Console organizations set `organization` and `workspace` limits. Any other combination returns 400. Setting `organization` and `workspace` limits through the API is in an early access preview. To request access, contact your Anthropic account team.",https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope.type,body,"""user""",true,user,,,,,,DOCUMENTED,Scope type. Always `user` for this scope.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope.user_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the member the spend limit applies to.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope.type,body,"""organization""",true,organization,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope.type,body,"""workspace""",true,workspace,,,,,,DOCUMENTED,Scope type. Always `workspace` for this scope.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,scope.workspace_id,body,string,true,,,,,,,DOCUMENTED,Tagged ID of the workspace the spend limit applies to.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,POST /v1/organizations/spend_limits,period,body,"""daily"" or ""monthly"" or ""weekly""",false,,,,daily|monthly|weekly,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/create
anthropic,GET /v1/organizations/spend_limits/effective,limit,query,number,false,20,1,1000,,,,DOCUMENTED,"Maximum number of members per page. A member's period rows never split across pages, so a page may carry more rows than this. Defaults to `20`.",https://platform.claude.com/docs/en/api/beta/organization/spend_limits/list_effective
anthropic,GET /v1/organizations/spend_limits/effective,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/list_effective
anthropic,GET /v1/organizations/spend_limits/effective,period,query,"array of ""daily"" or ""monthly"" or ""weekly""",false,,,,daily|monthly|weekly,,,DOCUMENTED,Restrict the report to these limit periods. Omit to return one row per period each member resolves a spend limit for.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/list_effective
anthropic,GET /v1/organizations/spend_limits/effective,user_ids,query,array of string,false,,,,,,,DOCUMENTED,"Restrict the report to these members, by tagged user ID (`user_...`). At most 100 entries.",https://platform.claude.com/docs/en/api/beta/organization/spend_limits/list_effective
anthropic,DELETE /v1/organizations/spend_limits/{spend_limit_id},spend_limit_id,path,string,true,,,,,,,DOCUMENTED,ID of the Spend Limit.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/delete
anthropic,GET /v1/organizations/spend_limits/{spend_limit_id},spend_limit_id,path,string,true,,,,,,,DOCUMENTED,ID of the Spend Limit.,https://platform.claude.com/docs/en/api/beta/organization/spend_limits/retrieve
anthropic,GET /v1/organizations/tunnels,include_archived,query,boolean,false,false,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Include archived tunnels in the results. Archived tunnels are excluded by default.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/list
anthropic,GET /v1/organizations/tunnels,limit,query,number,false,20,1,1000,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Maximum number of tunnels to return in a single page.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/list
anthropic,GET /v1/organizations/tunnels,page,query,string,false,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Opaque pagination cursor from a previous response's `next_page`. Omit to fetch the first page.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/list
anthropic,GET /v1/organizations/tunnels,workspace_id,query,string,false,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Return only tunnels in this Workspace. Accepts a `wrkspc_`-prefixed Workspace ID; omit to list tunnels across all Workspaces.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/list
anthropic,GET /v1/organizations/tunnels/{tunnel_id},tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/retrieve
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/archive,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/archive
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/list
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates,include_archived,query,boolean,false,false,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Include archived certificates in the results. Archived certificates are excluded by default.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/list
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates,limit,query,number,false,20,1,1000,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,Maximum number of certificates to return.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/list
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates,page,query,string,false,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,"A tunnel has at most two active certificates, so this list is not paginated.",https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/list
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/certificates,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/create
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/certificates,ca_certificate_pem,body,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,PEM-encoded X.509 CA certificate. Must contain exactly one certificate and no private-key material.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/create
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates/{certificate_id},tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/retrieve
anthropic,GET /v1/organizations/tunnels/{tunnel_id}/certificates/{certificate_id},certificate_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel Certificate.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/retrieve
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/certificates/{certificate_id}/archive,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/archive
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/certificates/{certificate_id}/archive,certificate_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel Certificate.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/tunnel_certificates/archive
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/reveal_token,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/reveal_token
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/rotate_token,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,ID of the Tunnel.,https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/rotate_token
anthropic,POST /v1/organizations/tunnels/{tunnel_id}/rotate_token,reason,body,string or null,false,,,,,,mcp-tunnels-2026-05-19,DOCUMENTED|BETA,"Optional free-text reason for the rotation, recorded for audit.",https://platform.claude.com/docs/en/api/beta/organization/mcp_tunnels/rotate_token
anthropic,GET /v1/organizations/usage_report/claude_code,starting_at,query,string,true,,,,,,,DOCUMENTED,UTC date in YYYY-MM-DD format. Returns metrics for this single day only.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_claude_code
anthropic,GET /v1/organizations/usage_report/claude_code,limit,query,number,false,20,1,1000,,,,DOCUMENTED,"Number of records per page (default: 20, max: 1000).",https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_claude_code
anthropic,GET /v1/organizations/usage_report/claude_code,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor token from previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_claude_code
anthropic,GET /v1/organizations/usage_report/messages,starting_at,query,string,true,,,,,,,DOCUMENTED,Time buckets that start on or after this RFC 3339 timestamp will be returned. Each time bucket will be snapped to the start of the minute/hour/day in UTC.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,account_ids,query,array of string,false,,,,,,,DOCUMENTED,Restrict usage returned to the specified user account ID(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,api_key_ids,query,array of string,false,,,,,,,DOCUMENTED,Restrict usage returned to the specified API key ID(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,bucket_width,query,"""1d"" or ""1h"" or ""1m""",false,1d,,,1d|1h|1m,,,DOCUMENTED,Time granularity of the response data.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,context_window,query,"array of ""0-200k"" or ""200k-1M""",false,,,,0-200k|200k-1M,,,DOCUMENTED,Restrict usage returned to the specified context window(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,ending_at,query,string,false,,,,,,,DOCUMENTED,Time buckets that end before this RFC 3339 timestamp will be returned.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,group_by,query,"array of ""account_id"" or ""api_key_id"" or ""context_window"" or 6 more",false,,,,account_id|api_key_id|context_window|inference_geo|model|service_account_id|service_tier|speed|workspace_id,,,DOCUMENTED,Group by any subset of the available options. Grouping by `speed` requires the `fast-mode-2026-02-01` beta header.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,inference_geos,query,"array of ""global"" or ""not_available"" or ""us""",false,,,,global|not_available|us,,,DOCUMENTED,Restrict usage returned to the specified inference geo(s). Use `not_available` for models that do not support specifying `inference_geo`.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,limit,query,number,false,,,,,,,DOCUMENTED,"Maximum number of time buckets to return in the response. The default and max limits depend on `bucket_width`: • `""1d""`: Default of 7 days, maximum of 31 days • `""1h""`: Default of 24 hours, maximum of 168 hours • `""1m""`: Default of 60 minutes, maximum of 1440 minutes",https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,models,query,array of string,false,,,,,,,DOCUMENTED,Restrict usage returned to the specified model(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,service_account_ids,query,array of string,false,,,,,,,DOCUMENTED,Restrict usage returned to the specified service account ID(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,service_tiers,query,"array of ""batch"" or ""flex"" or ""flex_discount"" or 3 more",false,,,,batch|flex|flex_discount|priority|priority_on_demand|standard,,,DOCUMENTED,Restrict usage returned to the specified service tier(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,speeds,query,"array of ""standard"" or ""fast""",false,,,,standard|fast,,,DOCUMENTED,Restrict usage returned to the specified speed(s) (Claude Code research preview). Requires the `fast-mode-2026-02-01` beta header.,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/usage_report/messages,workspace_ids,query,array of string,false,,,,,,,DOCUMENTED,Restrict usage returned to the specified workspace ID(s).,https://platform.claude.com/docs/en/api/beta/organization/usage_report/retrieve_messages
anthropic,GET /v1/organizations/users,after_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/organization/users/list
anthropic,GET /v1/organizations/users,before_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/organization/users/list
anthropic,GET /v1/organizations/users,email,query,string,false,,,,,,,DOCUMENTED,Filter by user email.,https://platform.claude.com/docs/en/api/beta/organization/users/list
anthropic,GET /v1/organizations/users,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/users/list
anthropic,GET /v1/organizations/users,roles,query,array of string,false,,,,,,,DOCUMENTED,"Filter to items whose `role` equals one of the supplied values. Repeatable; values are OR'ed together. Accepted values depend on the organization type: Console and API organizations accept `user`, `developer`, `billing`, `admin`, and `claude_code_user`; Claude Enterprise organizations accept `user`, `owner`, `primary_owner`, `membership_admin`, and `managed`.",https://platform.claude.com/docs/en/api/beta/organization/users/list
anthropic,DELETE /v1/organizations/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/users/remove
anthropic,GET /v1/organizations/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/users/retrieve
anthropic,POST /v1/organizations/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/users/update
anthropic,POST /v1/organizations/users/{user_id},role,body,"""billing"" or ""claude_code_user"" or ""developer"" or 2 more",true,,,,billing|claude_code_user|developer|managed|user,,,DOCUMENTED,"New role for the User. The accepted values depend on the organization type. Console and API organizations accept `user`, `developer`, `billing`, and `claude_code_user`; `admin` cannot be assigned through the API. Claude Enterprise organizations accept `user` and `managed`.",https://platform.claude.com/docs/en/api/beta/organization/users/update
anthropic,GET /v1/organizations/workspaces,after_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/list
anthropic,GET /v1/organizations/workspaces,before_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/list
anthropic,GET /v1/organizations/workspaces,include_archived,query,boolean,false,false,,,,,,DOCUMENTED,Whether to include Workspaces that have been archived in the response,https://platform.claude.com/docs/en/api/beta/organization/workspaces/list
anthropic,GET /v1/organizations/workspaces,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/list
anthropic,POST /v1/organizations/workspaces,name,body,string,true,,,,,,,DOCUMENTED,Name of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,data_residency,body,BetaDataResidencyCreateConfig or null,false,,,,,,,DOCUMENTED,"Data residency configuration for the workspace. If omitted, defaults to `workspace_geo: ""us""`, `allowed_inference_geos: ""unrestricted""`, and `default_inference_geo: ""global""`.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,data_residency.allowed_inference_geos,body,"array of BetaAllowedInferenceGeo or ""unrestricted"" or null",false,,,,,,,DOCUMENTED,"Permitted inference geo values. Defaults to 'unrestricted' if omitted, which allows all geos. Use the string 'unrestricted' to allow all geos, or a list of specific geos.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,data_residency.default_inference_geo,body,"""global"" or ""us"" or null",false,,,,global|us,,,DOCUMENTED,"Default inference geo applied when requests omit the parameter. Defaults to 'global' if omitted. Must be a member of `allowed_inference_geos` unless `allowed_inference_geos` is `""unrestricted""`.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,data_residency.workspace_geo,body,"""us"" or null",false,,,,,,,DOCUMENTED,Geographic region for workspace data storage. Immutable after creation. Defaults to 'us' if omitted.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,display_color,body,string or null,false,,,,,,,DOCUMENTED,Hex color code representing the Workspace in the Anthropic Console.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,external_key_id,body,string or null,false,,,,,,,DOCUMENTED,"ID of the customer-managed encryption key (CMEK) configuration to use for this Workspace. Setting this field requires CMEK to be enabled for your organization. When set, data stored for this Workspace is encrypted with the referenced key. Create key configurations with the External Keys API. On Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a single-Region key in the same AWS account and Region as the Workspace. On that platform the key is validated against this Workspace when it is attached, so a key-policy problem is reported as an error on this request. This field is write-once: once a key is attached to a Workspace it cannot be detached or replaced. To rotate key material, rotate the underlying key on your cloud KMS; the `external_key_id` stays the same.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,POST /v1/organizations/workspaces,tags,body,map[string] or null,false,,,,,,,DOCUMENTED,User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/create
anthropic,GET /v1/organizations/workspaces/{workspace_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/retrieve
anthropic,POST /v1/organizations/workspaces/{workspace_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},data_residency,body,BetaDataResidencyUpdateConfig or null,false,,,,,,,DOCUMENTED,Data residency configuration for the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},data_residency.allowed_inference_geos,body,"array of BetaAllowedInferenceGeo or ""unrestricted"" or null",false,,,,,,,DOCUMENTED,"Permitted inference geo values. Use 'unrestricted' to allow all geos, or a list of specific geos.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},data_residency.default_inference_geo,body,"""global"" or ""us"" or null",false,,,,global|us,,,DOCUMENTED,"Default inference geo applied when requests omit the parameter. Must be a member of `allowed_inference_geos` unless `allowed_inference_geos` is `""unrestricted""`.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},display_color,body,string,false,,,,,,,DOCUMENTED,Hex color code representing the Workspace in the Anthropic Console.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},external_key_id,body,string,false,,,,,,,DOCUMENTED,"ID of the customer-managed encryption key (CMEK) configuration to use for this Workspace. Setting this field requires CMEK to be enabled for your organization. When set, data stored for this Workspace is encrypted with the referenced key. Create key configurations with the External Keys API. On Claude Platform on AWS the value is the AWS KMS key ARN, and the key must be a single-Region key in the same AWS account and Region as the Workspace. On that platform the key is validated against this Workspace when it is attached, so a key-policy problem is reported as an error on this request. This field is write-once: once a key is attached to a Workspace it cannot be detached or replaced. To rotate key material, rotate the underlying key on your cloud KMS; the `external_key_id` stays the same.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},name,body,string,false,,,,,,,DOCUMENTED,Name of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id},tags,body,map[string] or null,false,,,,,,,DOCUMENTED,User-defined tags as string key-value pairs. Keys may not begin with `anthropic`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/update
anthropic,POST /v1/organizations/workspaces/{workspace_id}/archive,workspace_id,path,string,true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/workspaces/archive
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members,workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members,after_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members,before_id,query,string,false,,,,,,,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/list
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members,workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/add
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members,user_id,body,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/add
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members,workspace_role,body,BetaNoBillingWorkspaceRole,true,,,,workspace_admin|workspace_developer|workspace_restricted_developer|workspace_user,,,DOCUMENTED,Role of the new Workspace Member. Cannot be `workspace_billing`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/add
anthropic,DELETE /v1/organizations/workspaces/{workspace_id}/members/{user_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/remove
anthropic,DELETE /v1/organizations/workspaces/{workspace_id}/members/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/remove
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members/{user_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/retrieve
anthropic,GET /v1/organizations/workspaces/{workspace_id}/members/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/retrieve
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members/{user_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the Workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/update
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,ID of the User.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/update
anthropic,POST /v1/organizations/workspaces/{workspace_id}/members/{user_id},workspace_role,body,BetaWorkspaceRole,true,,,,workspace_admin|workspace_billing|workspace_developer|workspace_restricted_developer|workspace_user,,,DOCUMENTED,New workspace role for the User.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/members/update
anthropic,GET /v1/organizations/workspaces/{workspace_id}/rate_limits,workspace_id,path,string,true,,,,,,,DOCUMENTED,The ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/rate_limits/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/rate_limits,group_type,query,"""batch"" or ""files"" or ""model_group"" or 3 more",false,,,,batch|files|model_group|skills|token_count|web_search,,,DOCUMENTED,Filter by group type.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/rate_limits/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/rate_limits,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Maximum number of items to return per page. Ranges from `1` to `1000`. When omitted, every remaining entry is returned in a single page and `next_page` is `null`.",https://platform.claude.com/docs/en/api/beta/organization/workspaces/rate_limits/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/rate_limits,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/rate_limits/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/service_accounts,workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/service_accounts,limit,query,number,false,20,1,100,,,,DOCUMENTED,Number of results per page.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/list
anthropic,GET /v1/organizations/workspaces/{workspace_id}/service_accounts,page,query,string,false,,,,,,,DOCUMENTED,Opaque cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/list
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts,workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/add
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts,service_account_id,body,string,true,,,,,,,DOCUMENTED,Tagged service account ID to add.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/add
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts,workspace_role,body,BetaNoBillingWorkspaceRole,true,,,,workspace_admin|workspace_developer|workspace_restricted_developer|workspace_user,,,DOCUMENTED,Role to assign to the service account in this workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/add
anthropic,DELETE /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/remove
anthropic,DELETE /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/remove
anthropic,GET /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/retrieve
anthropic,GET /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/retrieve
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},workspace_id,path,string,true,,,,,,,DOCUMENTED,ID of the workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/update
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,ID of the service account.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/update
anthropic,POST /v1/organizations/workspaces/{workspace_id}/service_accounts/{service_account_id},workspace_role,body,BetaNoBillingWorkspaceRole,true,,,,workspace_admin|workspace_developer|workspace_restricted_developer|workspace_user,,,DOCUMENTED,New role for the service account in this workspace.,https://platform.claude.com/docs/en/api/beta/organization/workspaces/service_accounts/update
anthropic,POST /v1/messages,cache_control,body,object (CacheControlEphemeral) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Top-level automatic caching: the API places one cache breakpoint on the last cacheable block and moves it forward as the conversation grows. Uses one of the 4 breakpoint slots. 400 if the last block already carries an explicit cache_control with a different TTL, or if 4 explicit breakpoints exist. Not accepted on legacy Bedrock (Opus 4.6 and earlier).",https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,cache_control.type,body,string,false,,,,ephemeral,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Cache type. Only 'ephemeral' exists.,https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,cache_control.ttl,body,string,false,5m,,,5m|1h,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Cache lifetime. 5m (default, 1.25x input price on write) or 1h (2x on write). Longer-TTL breakpoints must precede shorter ones in the prompt.",https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,system[].cache_control,body,object (CacheControlEphemeral) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Explicit breakpoint on a system text block. Caches tools + system up to and including this block.,https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,tools[].cache_control,body,object (CacheControlEphemeral) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Explicit breakpoint on a tool definition; caches all tools up to and including it.,https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,messages[].content[].cache_control,body,object (CacheControlEphemeral) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Explicit breakpoint on a content block (text, image, document, tool_use, tool_result, search_result, compaction). Max 4 breakpoints per request (400 'A maximum of 4 blocks with cache_control may be provided'). Not allowed on thinking blocks, sub-blocks (citations) or empty text blocks. NOT accepted at message level (messages[].cache_control -> 400 'Extra inputs are not permitted').",https://platform.claude.com/docs/en/build-with-claude/prompt-caching
anthropic,POST /v1/messages,max_tokens,body,integer,true,,0,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Max output tokens. Required (missing → 400 'max_tokens: Field required'). Above the model max → 400 (Haiku 4.5: '1000000 > 64000'). `0` pre-warms the prompt cache without sampling (live: stop_reason max_tokens, empty content, output_tokens 0).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,diagnostics,body,object (BetaDiagnosticsParam) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,cache-diagnosis-2026-04-07,DOCUMENTED|BETA|LIVE_VERIFIED,Cache diagnostics (beta): pass the previous response id so the API reports where the prompt prefix diverged in response.diagnostics.,https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics
anthropic,POST /v1/messages,diagnostics.previous_message_id,body,string | null,false,,,256,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,cache-diagnosis-2026-04-07,DOCUMENTED|BETA,msg_… id of this client's previous response; null on the first turn to opt in.,https://platform.claude.com/docs/en/build-with-claude/cache-diagnostics
anthropic,POST /v1/messages,thinking,body,object (ThinkingConfigParam),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Thinking configuration. Variants: {type:'enabled', budget_tokens, display}, {type:'adaptive', display}, {type:'disabled'}. Per-model acceptance differs (see compatibility matrix). Omitting it means: on/always-on for 5.x & Fable, off for Opus 4.6-4.8/Sonnet 4.6 and all 4.5 models.",https://platform.claude.com/docs/en/build-with-claude/thinking
anthropic,POST /v1/messages,thinking.type,body,string,true,,,,enabled|disabled|adaptive,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"`enabled` (extended thinking with budget; Claude ≤ 4.6 and Mythos Preview), `adaptive` (Claude 4.6+; 4.5 models reject it: 'adaptive thinking is not supported on this model'), `disabled` (rejected on Fable 5.x / Mythos 5.x / Mythos Preview where thinking is always on).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,thinking.budget_tokens,body,integer,false,,1024,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Extended thinking budget (type=enabled). ≥1024 and < max_tokens; counts toward max_tokens.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,thinking.display,body,string,false,"model-specific (summarized <=4.6, omitted 4.7+)",,,summarized|omitted|updates,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|BETA|LIVE_VERIFIED,"How thinking text is returned. summarized = summary text (default on <=4.6); omitted = empty thinking field, signature kept (default on 4.7+, 5.x, Fable/Mythos); updates (beta, header thinking-display-updates-2026-08-18) = progress-update blocks carry text, reasoning blocks empty. Invalid with type disabled. Billing identical for all values.",https://platform.claude.com/docs/en/build-with-claude/thinking
anthropic,POST /v1/messages,thinking.block_binding,body,object (BetaThinkingBlockBinding) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,thinking-binding-controls-2026-08-01,DOCUMENTED|BETA|LIVE_VERIFIED,Preserved-thinking controls (beta). Accepted with type adaptive or enabled. Without the beta header -> 400 'block_binding: Extra inputs are not permitted'.,https://platform.claude.com/docs/en/build-with-claude/preserved-thinking
anthropic,POST /v1/messages,thinking.block_binding.prefix_mismatch_behavior,body,string,false,error,,,error|drop_block,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,thinking-binding-controls-2026-08-01,DOCUMENTED|BETA,"What happens when a replayed thinking block fails the conversation (prefix) check: error (default, 400 naming the block) or drop_block (drop it and later thinking blocks, report in input_transformations). Enforced by default only for accounts created on/after 2026-08-31 on Fable 5.1 / Mythos 5.1.",https://platform.claude.com/docs/en/build-with-claude/preserved-thinking
anthropic,POST /v1/messages,temperature,body,number,false,1.0,0,1,,all current models unless noted,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Sampling randomness 0..1 (default 1.0). DEPRECATED: models after Opus 4.6 accept only 1.0. Live on Haiku 4.5: `temperature` and `top_p` together → 400 '`temperature` and `top_p` cannot both be specified for this model'; 5 → 400 'temperature: range: 0..1'.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,top_k,body,integer,false,,0,,,all current models unless noted,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Top-K sampling (≥0). DEPRECATED: rejected (400) on models after Opus 4.6. Wrong type → 400 'top_k: Input should be a valid integer'.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,top_p,body,number,false,,0,1,,all current models unless noted,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Nucleus sampling 0..1. DEPRECATED: models after Opus 4.6 accept only ≥0.99. Cannot be combined with temperature on current models (live 400).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tool_choice,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"How Claude uses tools: {type: auto|any|tool|none, name?, disable_parallel_tool_use?}. Default auto when tools present, none otherwise. any/tool unsupported with manual thinking and on Fable 5.1/Mythos 5.1 (400).",https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools
anthropic,POST /v1/messages,messages[].content[] (assistant prefill),body,string,false,,,,,claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_DISCOVERED,"Assistant prefill is documented as unsupported while thinking is on (400 on Opus 4.6+, Fable). OBSERVED on Haiku 4.5: prefill + thinking enabled returned 200 with thinking silently disabled (no thinking block, no output_tokens_details).",https://platform.claude.com/docs/en/build-with-claude/thinking
anthropic,POST /v1/messages,output_config,body,object (OutputConfig),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Output configuration container: effort, format (structured outputs), task_budget (beta).",https://platform.claude.com/docs/en/build-with-claude/effort
anthropic,POST /v1/messages,output_config.effort,body,string,false,high,,,low|medium|high|xhigh|max,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101,,DOCUMENTED|LIVE_VERIFIED,"How much work Claude puts into the whole response (text, tool calls, thinking). Default high (explicit high == omitted, cache-neutral). xhigh/max availability per model. Rendered into the prompt: changing the top-level value invalidates message cache breakpoints (and tool/system on some models). 400 on unsupported models: 'This model does not support the effort parameter.' Invalid value -> 400 ""Input should be 'low', 'medium', 'high', 'xhigh' or 'max'"".",https://platform.claude.com/docs/en/build-with-claude/effort
anthropic,POST /v1/messages,messages[].output_config.effort,body,string,false,,,,low|medium|high|xhigh|max,claude-fable-5-1|claude-opus-5,mid-conversation-output-config-2026-07-01,DOCUMENTED|BETA|LIVE_VERIFIED,"Per-message effort (beta): a {role:'system', content:[], output_config:{effort}} entry inside messages changes effort from the next user turn on without invalidating the cache. Only Fable 5.1, Mythos 5.1, Opus 5; others -> 400 'output_config.effort requires a model that supports per-turn effort; this model does not'.",https://platform.claude.com/docs/en/build-with-claude/effort
anthropic,POST /v1/messages,output_config.task_budget,body,object (BetaTokenTaskBudget) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7,task-budgets-2026-03-13,DOCUMENTED|BETA|LIVE_DISCOVERED,"Advisory token budget for a whole agentic loop (thinking + tool calls + tool results + output); the model sees a server-injected countdown. Soft hint, not enforced. Doc minimum 20,000 (reference schema says minimum 1024 — discrepancy). Doc says unsupported on Sonnet 5, but a Sonnet 5 request with the header and total 20000 returned 200 (silently accepted?).",https://platform.claude.com/docs/en/build-with-claude/task-budgets
anthropic,POST /v1/messages,output_config.task_budget.type,body,string,false,,,,tokens,claude-fable-5-1|claude-opus-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7,task-budgets-2026-03-13,DOCUMENTED|BETA,Always 'tokens'.,https://platform.claude.com/docs/en/build-with-claude/task-budgets
anthropic,POST /v1/messages,output_config.task_budget.total,body,integer,false,,20000,,,claude-fable-5-1|claude-opus-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7,task-budgets-2026-03-13,DOCUMENTED|BETA,"Total tokens for the agentic loop. Documented minimum 20,000; schema minimum 1024.",https://platform.claude.com/docs/en/build-with-claude/task-budgets
anthropic,POST /v1/messages,output_config.task_budget.remaining,body,integer | null,false,,0,,,claude-fable-5-1|claude-opus-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7,task-budgets-2026-03-13,DOCUMENTED|BETA,Budget remainder carried over after client-side compaction. Defaults to total. Not allowed with the compaction parameter or a signed compaction block (400). Changing it invalidates the cache.,https://platform.claude.com/docs/en/build-with-claude/task-budgets
anthropic,POST /v1/messages,speed,body,string | null,false,standard,,,standard|fast,claude-opus-5|claude-opus-4-8,fast-mode-2026-02-01,DOCUMENTED|PREVIEW|BETA|ACCOUNT_RESTRICTED,"Inference speed. 'fast' = up to 2.5x output tokens/s at 2x pricing ($10/$50 Opus 5 & 4.8), research preview, dedicated rate limit. Requires beta header; without it -> 400 'speed: Extra inputs are not permitted'. Unsupported model -> 400 ""'<model>' does not support the `speed` parameter"". Opus 4.6: silently runs standard. Switching speed invalidates system+messages cache. Not on Batch API, Priority Tier, Bedrock/Vertex/Foundry/Claude Platform on AWS.",https://platform.claude.com/docs/en/build-with-claude/fast-mode
anthropic,POST /v1/messages,anthropic-beta: context-1m-2025-08-07,header,header,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-1m-2025-08-07,DOCUMENTED|LEGACY|LIVE_VERIFIED,"Legacy 1M-context beta header. 1M is now the default (no header) on Opus 4.6+/Sonnet 4.6+/Sonnet 5/Opus 5/Fable; retired for Sonnet 4.5/Sonnet 4 (no effect, >200k errors). Observed: accepted and ignored on Sonnet 5 and Haiku 4.5.",https://platform.claude.com/docs/en/build-with-claude/context-windows
anthropic,POST /v1/messages,anthropic-beta: model-context-window-exceeded-2025-08-26,header,header,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,model-context-window-exceeded-2025-08-26,DOCUMENTED|BETA,Opt older models into stop_reason model_context_window_exceeded instead of a validation error when input + max_tokens exceeds the window (default behaviour on 4.5+).,https://platform.claude.com/docs/en/build-with-claude/context-windows
anthropic,POST /v1/messages,context_management,body,object (BetaContextManagementConfig) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,Server-side context editing. Requires beta header (without: 400 'context_management: Extra inputs are not permitted'). Also accepted by count_tokens (returns context_management.original_input_tokens).,https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[],body,array of edit objects,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27 (compact_20260112: compact-2026-01-12),DOCUMENTED|BETA|LIVE_VERIFIED,"Ordered strategies. clear_thinking_20251015 must be listed first when combined. Types: clear_tool_uses_20250919, clear_thinking_20251015, compact_20260112.",https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[].type,body,string,true,,,,clear_tool_uses_20250919|clear_thinking_20251015|compact_20260112,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,Edit strategy.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].trigger,body,object,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA,"clear_tool_uses/compact: {type: input_tokens, value} or {type: tool_uses, value}.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].keep,body,object | 'all',false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,"clear_tool_uses: {type:'tool_uses', value} default 3. clear_thinking: {type:'thinking_turns', value>=1} | {type:'all'} | 'all'; default model-specific (all turns on Opus 4.5+/Sonnet 4.6+/Fable, last turn on Haiku & earlier).",https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[].clear_at_least,body,object | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,"{type:'input_tokens', value}: minimum tokens that must be clearable or the strategy is skipped (protects the prompt cache). clear_tool_uses only.",https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[].exclude_tools,body,array of string | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27,DOCUMENTED|BETA,Tool names whose uses/results are never cleared (clear_tool_uses).,https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[].clear_tool_inputs,body,boolean | array of string | null,false,false,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,context-management-2025-06-27,DOCUMENTED|BETA,"Also clear tool_use inputs (all, or listed tool names). Default false (results only).",https://platform.claude.com/docs/en/build-with-claude/context-editing
anthropic,POST /v1/messages,context_management.edits[].pause_after_compaction,body,boolean,false,,,,,all current models unless noted,compact-2026-01-12,DOCUMENTED|BETA,compact_20260112: stop after compaction (stop_reason `compaction`).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].instructions,body,string | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6,compact-2026-01-12,DOCUMENTED|BETA,compact_20260112: custom summarization prompt; completely replaces the default. On Fable 5.1/Mythos 5.1 custom instructions summarize the visible conversation only (no earlier thinking).,https://platform.claude.com/docs/en/build-with-claude/compaction
anthropic,POST /v1/messages,compaction,body,object (BetaCompactionConfig) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6,compact-2026-09-04,DOCUMENTED|BETA|LIVE_VERIFIED,"On-demand compaction (beta compact-2026-09-04): {type:'summarize', instructions?}. Summarizes every message sent, returns a single signed compaction block, no reply, stop_reason 'compaction', top-level usage 0 with usage.iterations[type=compaction]. Cannot combine with context_management, stop_sequences, output_config.format, tool_choice any/tool, task_budget.remaining. Send the block first in later requests (else 400 compaction_block_misplaced).",https://platform.claude.com/docs/en/build-with-claude/compaction
anthropic,POST /v1/messages,compaction.type,body,string,true,,,,summarize,all current models unless noted,compact-2026-09-04,DOCUMENTED|BETA,Always `summarize`.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,compaction.instructions,body,string | null,false,,,16384,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6,compact-2026-09-04,DOCUMENTED|BETA,"Replaces the server summarization prompt (max 16,384 chars; blank counts as absent). The summarizer reads earlier thinking too in this mode.",https://platform.claude.com/docs/en/build-with-claude/compaction
anthropic,POST /v1/messages,output_config.format,body,object (JSONOutputFormat) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"JSON outputs: constrain the text response to a JSON Schema via constrained decoding. GA, no beta header. Legacy top-level output_format -> 400 ""output_format: This field is deprecated. Use 'output_config.format' instead"". 400 with citations enabled ('Citations cannot be enabled when output format is set'). Incompatible with prefill and max_tokens:0. Adds a system prompt (more input tokens) and invalidates the cache when changed.",https://platform.claude.com/docs/en/build-with-claude/structured-outputs
anthropic,POST /v1/messages,output_config.format.type,body,string,true,,,,json_schema,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Always `json_schema`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config.format.schema,body,object (JSON Schema subset),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Supported: basic types, enum (primitives), const, anyOf/allOf (no allOf+$ref), $ref/$defs/definitions (internal), default, required, additionalProperties:false, formats date-time/time/date/duration/email/hostname/uri/ipv4/ipv6/uuid, minItems 0|1, simple regex pattern. Unsupported -> 400 e.g. ""For 'object' type, property 'patternProperties' is not supported"", ""For 'integer' type, property 'minimum' is not supported"". Limits: 20 strict tools, 24 optional params, 16 union params, 180 s compile timeout.",https://platform.claude.com/docs/en/build-with-claude/structured-outputs
anthropic,POST /v1/messages/count_tokens,output_config.format,body,object (JSONOutputFormat) | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,count_tokens accepts output_config.format (no grammar compilation); observed same input_tokens (205) as the real call.,https://platform.claude.com/docs/en/build-with-claude/structured-outputs
anthropic,POST /v1/messages,tools[].strict,body,boolean,false,false,,,,,,DOCUMENTED|LIVE_VERIFIED,Grammar-constrained validation of tool name+input. Not on mcp_toolset/computer_toolset/browser_toolset; not with code_execution callers (400).,https://platform.claude.com/docs/en/agents-and-tools/tool-use/strict-tool-use
anthropic,POST /v1/messages,messages[].content[].citations.enabled,body,boolean,false,,,,,all current models unless noted,,DOCUMENTED,Document-level citations config: enable citations for this document.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[] {type:'document'},body,object (DocumentBlockParam),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Document block. source variants: base64 (application/pdf), text (text/plain), content (custom content: array of text/image blocks), url, file (file_id). Optional title, context, citations, cache_control.",https://platform.claude.com/docs/en/build-with-claude/pdf-support
anthropic,POST /v1/messages,messages[].content[].source.type (document),body,string,false,,,,base64|text|content|url|file,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Document source discriminator.,https://platform.claude.com/docs/en/build-with-claude/citations
anthropic,POST /v1/messages,messages[].content[].source.media_type (document),body,string,false,,,,application/pdf|text/plain,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,application/pdf for base64; text/plain for text sources. Files API: PDF or text/plain files only in document blocks.,https://platform.claude.com/docs/en/build-with-claude/pdf-support
anthropic,POST /v1/messages,messages[].content[].title (document),body,string | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Passed to the model, echoed as document_title in citations, not citable. Limited length (use context for metadata).",https://platform.claude.com/docs/en/build-with-claude/citations
anthropic,POST /v1/messages,messages[].content[].context (document),body,string | null,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED,"Free-text metadata shown to the model, not citable.",https://platform.claude.com/docs/en/build-with-claude/citations
anthropic,POST /v1/messages,messages[].content[] {type:'search_result'},body,object (SearchResultBlockParam),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"RAG result block: source (string), title (string), content (array of non-empty text blocks), citations, cache_control. User messages and tool_result only (all-or-nothing inside a tool_result). Cited as search_result_location with block indices.",https://platform.claude.com/docs/en/build-with-claude/search-results
anthropic,POST /v1/messages,messages[].content[] {type:'image'},body,object (ImageBlockParam),false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Image block. source: base64 (media_type image/jpeg|png|gif|webp), url, file (file_id). Limits: 8000x8000 px, 10 MB base64 (5 MB Bedrock/Vertex), 100 images/request on 200k models, 600 on 1M models; >20 images => per-image ~2000 px limit. Cost = ceil(w/28)*ceil(h/28) visual tokens, capped by tier (1568 std, 4784 on 4.7+). Invalid bytes -> 400 'Could not process image'; unreachable URL -> 400 'Unable to download the file'.",https://platform.claude.com/docs/en/build-with-claude/vision
anthropic,POST /v1/messages,messages[].content[].source.type (image),body,string,false,,,,base64|url|file,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Image source discriminator.,https://platform.claude.com/docs/en/build-with-claude/vision
anthropic,POST /v1/messages,messages[].content[].source.media_type (image),body,string,false,,,,image/jpeg|image/png|image/gif|image/webp,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Image MIME type (base64 sources). Animations: first frame only.,https://platform.claude.com/docs/en/build-with-claude/vision
anthropic,POST /v1/messages,messages[].content[].transformations.oversized_image,body,string,false,downsize,,,downsize|error,all current models unless noted,,DOCUMENTED,"What to do when the image exceeds the model's max size: `downsize` (default, silent) or `error` (400).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[] {type:'container_upload'},body,object,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED,"Reference a Files API file inside the code execution container: {type:'container_upload', file_id}.",https://platform.claude.com/docs/en/build-with-claude/files
anthropic,POST /v1/files,file,body,binary (multipart part),true,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,File to upload. filename final path component kept; Content-Type on the part optional (detected: an application/octet-stream .txt part was stored as text/plain). Max 500 MB; 1 TB per org (guide) — reference says 100 GB in older text.,https://platform.claude.com/docs/en/api/files
anthropic,POST /v1/files,expires_in_seconds,body,integer,false,,3600,7776000,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Seconds until the file expires (3600..7776000). 400 'expires_in_seconds: must be between 3600 and 7776000' otherwise. expires_at = upload time + value.,https://platform.claude.com/docs/en/api/files
anthropic,GET /v1/files,limit,query,integer,false,20,1,1000,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,"Page size, default 20, 1..1000.",https://platform.claude.com/docs/en/api/files
anthropic,GET /v1/files,page,query,string,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED,Opaque cursor (prefixed page_) from next_page.,https://platform.claude.com/docs/en/api/files
anthropic,GET /v1/files,ids[],query,array of string,false,,,100,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Up to 100 ids; single page; unknown ids silently omitted; exclusive with page/limit.,https://platform.claude.com/docs/en/api/files
anthropic,GET /v1/files,before_id / after_id,query,string,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,files-api-2025-04-14,DOCUMENTED|LEGACY|LIVE_VERIFIED,"Legacy cursors, only with anthropic-beta: files-api-2025-04-14 (response then has has_more/first_id/last_id instead of next_page). 400 without the header.",https://platform.claude.com/docs/en/build-with-claude/files
anthropic,POST /v1/files,anthropic-workspace-id,header,string,false,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED,Optional header selecting the workspace (files are workspace-scoped).,https://platform.claude.com/docs/en/api/files
anthropic,GET /v1/files/{file_id},file_id,path,string,true,,,,,claude-fable-5-1|claude-opus-5|claude-sonnet-5|claude-fable-5|claude-opus-4-8|claude-opus-4-7|claude-sonnet-4-6|claude-opus-4-6|claude-opus-4-5-20251101|claude-haiku-4-5-20251001|claude-sonnet-4-5-20250929,,DOCUMENTED|LIVE_VERIFIED,Path parameter for GET/DELETE /v1/files/{file_id} and GET /v1/files/{file_id}/content.,https://platform.claude.com/docs/en/api/files
anthropic,POST /v1/messages,messages[].content[].source.file_id,body,string,false,,,,,all current models unless noted,,DOCUMENTED,Files API id (`file` source). Not accepted by /v1/messages/count_tokens.,https://platform.claude.com/docs/en/api/messages
anthropic,GET /v1/messages/batches,after_id,query,string,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately after this object.",https://platform.claude.com/docs/en/api/beta/messages/batches/list
anthropic,GET /v1/messages/batches,before_id,query,string,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"ID of the object to use as a cursor for pagination. When provided, returns the page of results immediately before this object.",https://platform.claude.com/docs/en/api/beta/messages/batches/list
anthropic,GET /v1/messages/batches,limit,query,number,false,20,1,1000,,,message-batches-2024-09-24,DOCUMENTED,Number of items to return per page. Defaults to `20`. Ranges from `1` to `1000`.,https://platform.claude.com/docs/en/api/beta/messages/batches/list
anthropic,GET /v1/messages/batches,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/list
anthropic,POST /v1/messages/batches,requests,body,array of object,true,,,,,,message-batches-2024-09-24,DOCUMENTED,List of requests for prompt completion. Each is an individual request to create a Message.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].custom_id,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,"Developer-provided ID created for each request in a Message Batch. Useful for matching results to requests, as results may be given out of request order. Must be unique for each request within the Message Batch. maxLength: 64, minLength: 1, pattern: ^[a-zA-Z0-9_-]{1,64}$",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params,body,object,true,,,,,,message-batches-2024-09-24,DOCUMENTED,Messages API creation parameters for the individual request. See the [Messages API reference](https://platform.claude.com/docs/en/api/messages) for full documentation on available parameters.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.max_tokens,body,number,true,,0,,,,message-batches-2024-09-24,DOCUMENTED,The maximum number of tokens to generate before stopping. Note that our models may stop _before_ reaching this maximum. This parameter only specifies the absolute maximum number of tokens to generate. Set to `0` to populate the [prompt cache](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#pre-warming-the-cache) without generating a response. Different models have different maximum values for this parameter.  See [models](https://platform.claude.com/docs/en/about-claude/models/overview) for details.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages,body,array of BetaMessageParam,true,,,,,,message-batches-2024-09-24,DOCUMENTED,"Input messages. Our models are trained to operate on alternating `user` and `assistant` conversational turns. When creating a new `Message`, you specify the prior conversational turns with the `messages` parameter, and the model then generates the next `Message` in the conversation. Consecutive `user` or `assistant` turns in your request will be combined into a single turn. Each input message must be an object with a `role` and `content`. You can specify a single `user`-role message, or you can include multiple `user` and `assistant` messages. If the final message uses the `assistant` role, the response content will continue immediately from the content in that message. This can be used to constrain part of the model's response. Example with a single `user` message: ```json [{""role"": ""user"", ""content"": ""Hello, Claude""}] ``` Example with multiple conversational turns: ```json [ {""role"": ""user"", ""content"": ""Hello there.""}, {""role"": ""assistant"", ""content"": ""Hi, I'm Claude. How can I help you?""}, {""role"": ""user"", ""content"": ""Can you explain LLMs in plain English?""}, ] ``` Example with a partially-filled response from Claude: ```json [ {""role"": ""user"", ""content"": ""What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun""}, {""role"": ""assistant"", ""content"": ""The best answer is (""}, ] ``` Each input message `content` may be either a single `string` or an array of content blocks, where each block has a specific `type`. Using a `string` for `content` is shorthand for an array of one content block of type `""text""`. The following input messages are equivalent: ```json {""role"": ""user"", ""content"": ""Hello, Claude""} ``` ```json {""role"": ""user"", ""content"": [{""type"": ""text"", ""text"": ""Hello, Claude""}]} ``` See [input examples](https://platform.claude.com/docs/en/build-with-claude/working-with-messages). Note that if you want to include a [system prompt](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices#give-claude-a-role), you can use the top-level `system` parameter — there is no `""system""` role for input messages in the Messages API. There is a limit of 100,000 messages in a single request.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages[].content,body,string or array of BetaContentBlockParam,true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages[].role,body,"""user"" or ""assistant"" or ""system""",true,,,,user|assistant|system,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages[].clear_at,body,"""next_user_message"" or ""never"" or null",false,,,,next_user_message|never,,message-batches-2024-09-24,DOCUMENTED,"How long this system message's text stays in front of the model. `""never""` (the default) renders it on every request that includes it. `""next_user_message""` renders it only for the user turn it follows: once a later `role: ""user""` message exists in `messages` the message stays in the array (send it unchanged) but is no longer shown to the model. Only permitted on `role: ""system""` messages.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages[].output_config,body,BetaSystemMessageOutputConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Per-message output configuration on a role:""system"" input message. Fields here apply per-turn; `format` remains top-level only. An empty `{}` is accepted on a message that carries content; a message with neither content nor output_config fields is rejected.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.messages[].output_config.effort,body,"""low"" or ""medium"" or ""high"" or 2 more or null",false,,,,low|medium|high|xhigh|max,,message-batches-2024-09-24,DOCUMENTED,All possible effort levels.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.model,body,Model,true,,,,,,message-batches-2024-09-24,DOCUMENTED,The model that will complete your prompt. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.cache_control,body,BetaCacheControlEphemeral or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Top-level cache control automatically applies a cache_control marker to the last cacheable block in the request.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.compaction,body,BetaCompactionConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Compact the whole conversation and return a signed `compaction` block, alone, that a later request sends back first in `messages`, in place of the messages it summarizes. There is no trigger and no pause flag: sending the parameter compacts, and nothing is sampled after the block. The summarization prompt is the server's own unless `instructions` are given, which then replace it for this request; a value that is empty or only whitespace counts as absent.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.compaction.type,body,"""summarize""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.compaction.instructions,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Replaces the server's default summarization prompt for this request. An empty or whitespace-only value counts as absent.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container,body,BetaContainerParams or string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Container identifier for reuse across requests.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container.id,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Container id,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container.skills,body,array of BetaSkillParams or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,List of skills to load in the container,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container.skills[].type,body,"""anthropic"" or ""custom""",true,,,,anthropic|custom,,message-batches-2024-09-24,DOCUMENTED,Type of skill - either 'anthropic' (built-in) or 'custom' (user-defined),https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container.skills[].skill_id,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,Skill ID,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.container.skills[].version,body,string,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Skill version or 'latest' for most recent version,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.context_management,body,BetaContextManagementConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Context management configuration. This allows you to control how Claude manages context across multiple requests, such as whether to clear function results or not.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.context_management.edits,body,array of BetaClearToolUses20250919Edit or BetaClearThinking20251015Edit or BetaCompact20260112Edit,false,,,,,,message-batches-2024-09-24,DOCUMENTED,List of context management edits to apply,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.diagnostics,body,BetaDiagnosticsParam or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Request-level diagnostics. Currently carries the previous response id for prompt-cache divergence reporting.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.diagnostics.previous_message_id,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,The `id` (`msg_...`) from this client's previous /v1/messages response. The server compares that request's prompt fingerprint against this one and returns `diagnostics.cache_miss_reason` when the prompt-cache prefix could not be reused. Pass `null` on the first turn to opt in without a prior message to compare.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallback_credit_token,body,string or BetaFallbackCreditTokenParam or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"The `fallback_credit_token` from a prior refusal's `stop_details`. When a preceding request was refused and returned a `fallback_credit_token`, pass that code here on the retry to have the retry's cache-creation tokens for the prefix that was warm on the refused model billed at the cache-read rate. Must be redeemed by the same organization and workspace, with the same request body (optionally extended by one appended `assistant` message whose content is the partial text — with any trailing whitespace stripped from the final text block — and paired server-tool blocks streamed before the refusal; the appended-assistant form is not available for requests with `output_format` set or forced `tool_choice`), on an eligible fallback model, on the same platform, and within 5 minutes of the refusal; a mismatch is a 400. A token minted mid-server-tool-loop whose partial content was continuable may only be redeemed with the appended-assistant form — if an exact-body retry is rejected with a 400 saying the token must be redeemed by continuing the partial response, retry with the appended-assistant form instead. When the appended-assistant form is used on a model that otherwise disallows assistant-turn prefill, this token also authorizes that one prefill.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallback_credit_token.token,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,The opaque `fallback_credit_token` from a prior refusal's `stop_details` — the same string the bare-string form carries.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallback_credit_token.mode,body,"""strict"" or ""best_effort""",false,,,,strict|best_effort,,message-batches-2024-09-24,DOCUMENTED,"How a failing token affects the retry. `strict` (the default, and the bare-string behavior): a failing redemption is a 400 and the retry is not served. `best_effort`: the retry is served either way — a token-layer failure no longer rejects the request; the retry proceeds at normal price and the outcome is reported on the response's `usage.fallback_credit`. Two failures stay hard in both modes: a malformed token, and combining `fallback_credit_token` with `fallbacks`.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks,body,BetaFallbacksParam or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Opt-in server-side retry on one or more substitute models when the requested model declines for policy reasons. Tried in order: if the first entry also declines, the second is tried, and so on. The string ""default"" requests the requested model's server-defined default fallback configuration.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.model,body,Model,true,,,,,,message-batches-2024-09-24,DOCUMENTED,The model that will complete your prompt. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.max_tokens,body,number or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.output_config,body,BetaOutputConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.output_config.effort,body,"""low"" or ""medium"" or ""high"" or 2 more or null",false,,,,low|medium|high|xhigh|max,,message-batches-2024-09-24,DOCUMENTED,All possible effort levels.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.output_config.format,body,BetaJSONOutputFormat or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,A schema to specify Claude's output format in responses. See [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs),https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.output_config.task_budget,body,BetaTokenTaskBudget or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,User-configurable total token budget across contexts.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,message-batches-2024-09-24,DOCUMENTED,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.fallbacks.thinking,body,BetaThinkingConfigEnabled or BetaThinkingConfigDisabled or BetaThinkingConfigAdaptive or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.inference_geo,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Specifies the geographic region for inference processing. If not specified, the workspace's `default_inference_geo` is used.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers,body,array of BetaRequestMCPServerURLDefinition,false,,,,,,message-batches-2024-09-24,DOCUMENTED,MCP servers to be utilized in this request,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].type,body,"""url""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].name,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].url,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].authorization_token,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].tool_configuration,body,BetaRequestMCPServerToolConfiguration or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].tool_configuration.allowed_tools,body,array of string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.mcp_servers[].tool_configuration.enabled,body,boolean or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.metadata,body,BetaMetadata,false,,,,,,message-batches-2024-09-24,DOCUMENTED,An object describing metadata about the request.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.metadata.user_id,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"An external identifier for the user who is associated with the request. This should be a uuid, hash value, or other opaque identifier. Anthropic may use this id to help detect abuse. Do not include any identifying information such as name, email address, or phone number.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.output_config,body,BetaOutputConfig,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Configuration options for the model's output, such as the output format.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.service_tier,body,"""auto"" or ""standard_only""",false,,,,auto|standard_only,,message-batches-2024-09-24,DOCUMENTED,Determines whether to use priority capacity (if available) or standard capacity for this request. Anthropic offers different levels of service for your API requests. See [service-tiers](https://platform.claude.com/docs/en/api/service-tiers) for details.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,message-batches-2024-09-24,DOCUMENTED,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.stop_sequences,body,array of string,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Custom text sequences that will cause the model to stop generating. Our models will normally stop when they have naturally completed their turn, which will result in a response `stop_reason` of `""end_turn""`. If you want the model to stop generating when it encounters custom strings of text, you can use the `stop_sequences` parameter. If the model encounters one of the custom sequences, the response `stop_reason` value will be `""stop_sequence""` and the response `stop_sequence` value will contain the matched stop sequence.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.stream,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Whether to incrementally stream the response using server-sent events. See [streaming](https://platform.claude.com/docs/en/build-with-claude/streaming) for details.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.system,body,string or array of BetaTextBlockParam,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"System prompt. A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role. See our [guide to system prompts](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices#give-claude-a-role).",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.system.type,body,"""text""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.system.text,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.system.cache_control,body,BetaCacheControlEphemeral or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Create a cache control breakpoint at this content block.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.system.citations,body,array of BetaTextCitationParam or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.thinking,body,BetaThinkingConfigParam,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Configuration for enabling Claude's extended thinking. When enabled, responses include `thinking` content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your `max_tokens` limit. See [extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) for details.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice,body,BetaToolChoice,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"How the model should use the provided tools. The model can use a specific tool, any available tool, decide by itself, or not use tools at all.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.type,body,"""auto""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.disable_parallel_tool_use,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Whether to disable parallel tool use. Defaults to `false`. If set to `true`, the model will output exactly one tool use.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.type,body,"""any""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.type,body,"""tool""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.name,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,The name of the tool to use.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tool_choice.type,body,"""none""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools,body,array of BetaToolUnion,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Definitions of tools that the model may use. If you include `tools` in your API request, the model may return `tool_use` content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using `tool_result` content blocks. There are two types of tools: **client tools** and **server tools**. The behavior described below applies to client tools. For [server tools](https://platform.claude.com/docs/en/agents-and-tools/tool-use/server-tools), see their individual documentation as each has its own behavior (e.g., the [web search tool](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool)). Each tool definition includes: * `name`: Name of the tool. * `description`: Optional, but strongly-recommended description of the tool. * `input_schema`: [JSON schema](https://json-schema.org/draft/2020-12) for the tool `input` shape that the model will produce in `tool_use` output content blocks. For example, if you defined `tools` as: ```json [ { ""name"": ""get_stock_price"", ""description"": ""Get the current stock price for a given ticker symbol."", ""input_schema"": { ""type"": ""object"", ""properties"": { ""ticker"": { ""type"": ""string"", ""description"": ""The stock ticker symbol, e.g. AAPL for Apple Inc."" } }, ""required"": [""ticker""] } } ] ``` And then asked the model ""What's the S&P 500 at today?"", the model might produce `tool_use` content blocks in the response like this: ```json [ { ""type"": ""tool_use"", ""id"": ""toolu_01D7FLrfh4GYq7yT1ULFeyMV"", ""name"": ""get_stock_price"", ""input"": { ""ticker"": ""^GSPC"" } } ] ``` You might then run your `get_stock_price` tool with `{""ticker"": ""^GSPC""}` as an input, and return the following back to the model in a subsequent `user` message: ```json [ { ""type"": ""tool_result"", ""tool_use_id"": ""toolu_01D7FLrfh4GYq7yT1ULFeyMV"", ""content"": ""259.75 USD"" } ] ``` Tools can be used for workflows that include running client-side tools and functions, or more generally whenever you want the model to produce a particular JSON structure of output. See our [guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) for more details.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""custom"" or null",false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].input_schema,body,object,true,,,,,,message-batches-2024-09-24,DOCUMENTED,[JSON schema](https://json-schema.org/draft/2020-12) for this tool's input. This defines the shape of the `input` that your tool accepts and that the model will produce.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].input_schema.type,body,"""object""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].input_schema.properties,body,map[unknown] or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].input_schema.required,body,array of string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,"Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks. maxLength: 128, minLength: 1, pattern: ^[a-zA-Z0-9_-]{1,128}$",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].allowed_callers,body,"array of ""direct"" or ""code_execution_20250825"" or ""code_execution_20260120"" or ""code_execution_20260521""",false,,,,direct|code_execution_20250825|code_execution_20260120|code_execution_20260521,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].cache_control,body,BetaCacheControlEphemeral or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Create a cache control breakpoint at this content block.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].defer_loading,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"If true, tool will not be included in initial system prompt. Only loaded when returned via tool_reference from tool search.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].description,body,string,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Description of what this tool does. Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].eager_input_streaming,body,boolean or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Enable eager input streaming for this tool. When true, tool input parameters will be streamed incrementally as they are generated, and types will be inferred on-the-fly rather than buffering the full JSON output. When false, streaming is disabled for this tool even if the fine-grained-tool-streaming beta is active. When null (default), uses the default behavior based on beta headers.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].input_examples,body,array of map[unknown],false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].strict,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"When true, guarantees schema validation on tool names and inputs",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""bash_20241022""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""bash""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""bash_20250124""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""code_execution_20250522""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""code_execution""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""code_execution_20250825""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""code_execution_20260120""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""code_execution_20260521""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""browser_toolset_20260801""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs,body,BetaBrowserToolsetConfigs or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Per-member configuration for `browser_toolset_20260801`: one optional field per member tool, keyed by the member name — the same name the member's `tool_use` blocks carry. Every member is an accepted key, and a member's defaults apply wherever its key is absent. Unknown keys are rejected: the field set is this toolset version's complete member set.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.type,body,BetaBrowserTypeConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`type`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.close_tab,body,BetaBrowserCloseTabConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`close_tab`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.double_click,body,BetaBrowserDoubleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`double_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.file_upload,body,BetaBrowserFileUploadConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`file_upload`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.find,body,BetaBrowserFindConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`find`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.form_input,body,BetaBrowserFormInputConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`form_input`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.get_page_text,body,BetaBrowserGetPageTextConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`get_page_text`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.hold_key,body,BetaBrowserHoldKeyConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`hold_key`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.hover,body,BetaBrowserHoverConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`hover`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.javascript_exec,body,BetaBrowserJavascriptExecConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`javascript_exec`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.key,body,BetaBrowserKeyConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`key`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_click,body,BetaBrowserLeftClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_click_drag,body,BetaBrowserLeftClickDragConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_click_drag`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_mouse_down,body,BetaBrowserLeftMouseDownConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_mouse_down`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_mouse_up,body,BetaBrowserLeftMouseUpConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_mouse_up`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.list_tabs,body,BetaBrowserListTabsConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`list_tabs`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.middle_click,body,BetaBrowserMiddleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`middle_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.mouse_move,body,BetaBrowserMouseMoveConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`mouse_move`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.navigate,body,BetaBrowserNavigateConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`navigate`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.new_tab,body,BetaBrowserNewTabConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`new_tab`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.read_console,body,BetaBrowserReadConsoleConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`read_console`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.read_network,body,BetaBrowserReadNetworkConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`read_network`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.read_page,body,BetaBrowserReadPageConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`read_page`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.right_click,body,BetaBrowserRightClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`right_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.screenshot,body,BetaBrowserScreenshotConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`screenshot`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.scroll,body,BetaBrowserScrollConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`scroll`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.scroll_to,body,BetaBrowserScrollToConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`scroll_to`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.switch_tab,body,BetaBrowserSwitchTabConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`switch_tab`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.triple_click,body,BetaBrowserTripleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`triple_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.wait,body,BetaBrowserWaitConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`wait`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.zoom,body,BetaBrowserZoomConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`zoom`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""computer_20241022""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].display_height_px,body,number,true,,1,,,,message-batches-2024-09-24,DOCUMENTED,The height of the display in pixels.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].display_width_px,body,number,true,,1,,,,message-batches-2024-09-24,DOCUMENTED,The width of the display in pixels.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""computer""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].display_number,body,number or null,false,,0,,,,message-batches-2024-09-24,DOCUMENTED,"The X11 display number (e.g. 0, 1) for the display.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""memory_20250818""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""memory""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""computer_20250124""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""text_editor_20241022""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""str_replace_editor""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""computer_20251124""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].enable_zoom,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Whether to enable an action to take a zoomed-in screenshot of the screen.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""computer_toolset_20260801""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs,body,BetaComputerToolsetConfigs or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Per-member configuration for `computer_toolset_20260801`: one optional field per member tool, keyed by the member name — the same name the member's `tool_use` blocks carry. Every member is an accepted key, and a member's defaults apply wherever its key is absent. Unknown keys are rejected: the field set is this toolset version's complete member set.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.type,body,BetaComputerTypeConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`type`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.cursor_position,body,BetaComputerCursorPositionConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`cursor_position`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.double_click,body,BetaComputerDoubleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`double_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.hold_key,body,BetaComputerHoldKeyConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`hold_key`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.key,body,BetaComputerKeyConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`key`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_click,body,BetaComputerLeftClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_click_drag,body,BetaComputerLeftClickDragConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_click_drag`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_mouse_down,body,BetaComputerLeftMouseDownConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_mouse_down`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.left_mouse_up,body,BetaComputerLeftMouseUpConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`left_mouse_up`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.middle_click,body,BetaComputerMiddleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`middle_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.mouse_move,body,BetaComputerMouseMoveConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`mouse_move`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.right_click,body,BetaComputerRightClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`right_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.screenshot,body,BetaComputerScreenshotConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`screenshot`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.scroll,body,BetaComputerScrollConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`scroll`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.triple_click,body,BetaComputerTripleClickConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`triple_click`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.wait,body,BetaComputerWaitConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`wait`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.zoom,body,BetaComputerZoomConfig or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,`zoom`'s config overrides.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""text_editor_20250124""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""text_editor_20250429""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""str_replace_based_edit_tool""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""text_editor_20250728""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].max_characters,body,number or null,false,,1,,,,message-batches-2024-09-24,DOCUMENTED,"Maximum number of characters to display when viewing a file. If not specified, defaults to displaying the full file.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_search_20250305""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""web_search""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].allowed_domains,body,array of string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,List of domains to allow fetching from,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].blocked_domains,body,array of string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,List of domains to block fetching from,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].max_uses,body,number or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Maximum number of times the tool can be used in the API request. exclusiveMinimum: 0,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location,body,BetaUserLocation or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Parameters for the user's location. Used to provide more relevant search results.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location.type,body,"""approximate""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location.city,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,The city of the user.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location.country,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,The two letter [ISO country code](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2) of the user.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location.region,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,The region of the user.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].user_location.timezone,body,string or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,The [IANA timezone](https://nodatime.org/TimeZones) of the user.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_fetch_20250910""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""web_fetch""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].citations,body,BetaCitationsConfigParam or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Citations configuration for fetched documents. Citations are disabled by default.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].max_content_tokens,body,number or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Maximum number of tokens used by including web page text content in the context. The limit is approximate and does not apply to binary content such as PDFs. exclusiveMinimum: 0,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].url_sources,body,BetaWebFetchURLSources or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Which sources contribute to the set of URLs web fetch may fetch. Each key is a tagged variant: `user_input` is `all` or `none`; the two tool filters are `all`, `none`, `only` (only the named tools' results) or `except` (every result but the named tools'). A named tool must be declared in this request's `tools[]`.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].url_sources.client_tool_results,body,BetaWebFetchURLSourceAll or BetaWebFetchURLSourceNone or BetaWebFetchURLSourceOnly or BetaWebFetchURLSourceExcept,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Which client tools' results contribute fetchable URLs: ""all"", ""none"", or an only or except list of client tool names from tools[].",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].url_sources.server_tool_results,body,BetaWebFetchURLSourceAll or BetaWebFetchURLSourceNone or BetaWebFetchURLSourceOnly or BetaWebFetchURLSourceExcept,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Which server tools' results contribute fetchable URLs: ""all"", ""none"", or an only or except list of server tool names from tools[]; only web_search and web_fetch results ever contribute.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].url_sources.user_input,body,BetaWebFetchURLSourceAll or BetaWebFetchURLSourceNone,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Whether URLs in user messages are fetchable: ""all"" or ""none"".",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_search_20260209""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_fetch_20260209""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_fetch_20260309""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].use_cache,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Whether to use cached content. Set to false to bypass the cache and fetch fresh content. Only set to false when the user explicitly requests fresh content or when fetching rapidly-changing sources.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_search_20260318""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].response_inclusion,body,"""full"" or ""excluded""",false,,,,full|excluded,,message-batches-2024-09-24,DOCUMENTED,"How this tool's result blocks appear in the API response when the result was consumed by a completed code_execution call in the same turn. 'full' returns the complete content (default). 'excluded' drops the nested server_tool_use and result block pair entirely. Results from direct calls, or from code_execution calls that paused before completing, are always returned in full so they can be sent back on the next turn.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""web_fetch_20260318""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""advisor_20260301""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].model,body,Model,true,,,,,,message-batches-2024-09-24,DOCUMENTED,The model that will complete your prompt. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""advisor""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].caching,body,BetaCacheControlEphemeral or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Caching for the advisor's own prompt. When set, each advisor call writes a cache entry at the given TTL so subsequent calls in the same conversation read the stable prefix. When omitted, the advisor prompt is not cached.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].max_tokens,body,number or null,false,,1024,,,,message-batches-2024-09-24,DOCUMENTED,"Bounds the advisor's total output (thinking + text) per call. When the advisor hits this cap, the returned advisor_result or advisor_redacted_result block carries stop_reason='max_tokens', and a truncation note is appended to the advice text the worker model sees (inside the encrypted blob in redacted mode). When set, the server also emits a remaining-tokens budget block in the advisor's prompt so the advisor self-shapes toward the cap. When omitted, the advisor model's default output cap applies and no budget block is emitted.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""tool_search_tool_bm25_20251119"" or ""tool_search_tool_bm25""",true,,,,tool_search_tool_bm25_20251119|tool_search_tool_bm25,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""tool_search_tool_bm25""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""tool_search_tool_regex_20251119"" or ""tool_search_tool_regex""",true,,,,tool_search_tool_regex_20251119|tool_search_tool_regex,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].name,body,"""tool_search_tool_regex""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the tool. This is how the tool will be called by the model and in `tool_use` blocks.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].type,body,"""mcp_toolset""",true,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].mcp_server_name,body,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,Name of the MCP server to configure tools for,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs,body,map[BetaMCPToolConfig] or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,"Configuration overrides for specific tools, keyed by tool name",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.defer_loading,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].configs.enabled,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].default_config,body,BetaMCPToolDefaultConfig,false,,,,,,message-batches-2024-09-24,DOCUMENTED,Default configuration applied to all tools from this server,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].default_config.defer_loading,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.tools[].default_config.enabled,body,boolean,false,,,,,,message-batches-2024-09-24,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.output_format,body,BetaJSONOutputFormat or null,false,,,,,,message-batches-2024-09-24,DOCUMENTED,**Deprecated** Deprecated: Use `output_config.format` instead. See [structured outputs](https://platform.claude.com/docs/en/build-with-claude/structured-outputs) A schema to specify Claude's output format in responses. This parameter will be removed in a future release.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.temperature,body,number,false,,0,1,,,message-batches-2024-09-24,DOCUMENTED,"**Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not support setting temperature. A value of 1.0 will be accepted for backwards compatibility, all other values will be rejected with a 400 error. Amount of randomness injected into the response. Defaults to `1.0`. Ranges from `0.0` to `1.0`. Use `temperature` closer to `0.0` for analytical / multiple choice, and closer to `1.0` for creative and generative tasks. Note that even with `temperature` of `0.0`, the results will not be fully deterministic.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.top_k,body,number,false,,0,,,,message-batches-2024-09-24,DOCUMENTED,"**Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not accept top_k; any value will be rejected with a 400 error. Only sample from the top K options for each subsequent token. Used to remove ""long tail"" low probability responses. [Learn more technical details here](https://towardsdatascience.com/how-to-sample-from-language-models-682bceb97277). Recommended for advanced use cases only.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,requests[].params.top_p,body,number,false,,0,1,,,message-batches-2024-09-24,DOCUMENTED,"**Deprecated**: Deprecated. Models released after Claude Opus 4.6 do not support setting top_p. A value >= 0.99 will be accepted for backwards compatibility, all other values will be rejected with a 400 error. Use nucleus sampling. In nucleus sampling, we compute the cumulative distribution over all the options for each subsequent token in decreasing probability order and cut it off once it reaches a particular probability specified by `top_p`. Recommended for advanced use cases only.",https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,anthropic-user-profile-id,header,string,false,,,,,,,DOCUMENTED,The user profile ID to attribute the requests in this batch to. Use when acting on behalf of a party other than your organization. Requires the `user-profiles` beta header. Applies to every request in the batch; an individual request whose `user_profile_id` body field conflicts with this header is errored.,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,POST /v1/messages/batches,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/create
anthropic,DELETE /v1/messages/batches/{message_batch_id},message_batch_id,path,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,ID of the Message Batch.,https://platform.claude.com/docs/en/api/beta/messages/batches/delete
anthropic,DELETE /v1/messages/batches/{message_batch_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/delete
anthropic,GET /v1/messages/batches/{message_batch_id},message_batch_id,path,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,ID of the Message Batch.,https://platform.claude.com/docs/en/api/beta/messages/batches/retrieve
anthropic,GET /v1/messages/batches/{message_batch_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/retrieve
anthropic,POST /v1/messages/batches/{message_batch_id}/cancel,message_batch_id,path,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,ID of the Message Batch.,https://platform.claude.com/docs/en/api/beta/messages/batches/cancel
anthropic,POST /v1/messages/batches/{message_batch_id}/cancel,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/cancel
anthropic,GET /v1/messages/batches/{message_batch_id}/results,message_batch_id,path,string,true,,,,,,message-batches-2024-09-24,DOCUMENTED,ID of the Message Batch.,https://platform.claude.com/docs/en/api/beta/messages/batches/results
anthropic,GET /v1/messages/batches/{message_batch_id}/results,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/messages/batches/results
anthropic,POST /v1/complete,model,body,string,true,,,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED|FAILED_VERIFICATION,"Model id. Live (2026-09-18): every request, with claude-2.1 or claude-haiku-4-5-20251001, returns 400 'The /v1/complete endpoint has been deprecated. Please use the /v1/messages endpoint instead.'",https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,prompt,body,string,true,,1,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Prompt with alternating '\n\nHuman:' / '\n\nAssistant:' turns (minLength 1).,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,max_tokens_to_sample,body,integer,true,,1,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Max tokens to generate (≥1).,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,stop_sequences,body,array<string>,false,,,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Extra stop strings; the model always stops on '\n\nHuman:'.,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,stream,body,boolean,false,false,,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,SSE streaming of `completion` events.,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,temperature,body,number,false,1.0,0,1,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,"0..1, default 1.0 (deprecated semantics as in Messages).",https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,top_p,body,number,false,,0,1,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Nucleus sampling.,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,top_k,body,integer,false,,0,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Top-K sampling.,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,metadata.user_id,body,string|null,false,,,512,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Opaque end-user id (≤512).,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,anthropic-beta,header,string,false,,,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Optional beta header(s).,https://platform.claude.com/docs/en/api/completions
anthropic,POST /v1/complete,anthropic-workspace-id,header,string,false,,,,,all current models unless noted,,DOCUMENTED|LEGACY|DEPRECATED,Workspace id.,https://platform.claude.com/docs/en/api/completions
anthropic,GET /v1/compliance/activities,activity_types[],query,"array of ""abuse_decision_received"" or ""account_deleted"" or ""admin_api_key_created"" or 503 more",false,,,,abuse_decision_received|account_deleted|admin_api_key_created|admin_api_key_deleted|admin_api_key_updated|admin_connector_request_resolved|admin_request_created|admin_setup_checklist_step_delegated|admin_setup_checklist_step_delegation_cancelled|age_verified|anonymous_mobile_login_attempted|api_key_created|audit_log_export_accessed|audit_log_export_started|billing_emails_updated|ccr_agent_created|ccr_agent_deleted|ccr_agent_proxy_anthropic_oidc_token_exchanged|ccr_agent_proxy_anthropic_oidc_token_minted|ccr_agent_proxy_credential_created|ccr_agent_proxy_credential_deleted|ccr_agent_proxy_credential_rotated|ccr_agent_proxy_credential_updated|ccr_agent_proxy_destination_deleted|ccr_agent_proxy_network_events_listed|ccr_agent_proxy_profile_bound|ccr_agent_proxy_profile_created|ccr_agent_proxy_profile_deleted|ccr_agent_proxy_profile_unbound|ccr_agent_proxy_profile_updated|ccr_agent_proxy_provisioning_credential_rejected|ccr_agent_proxy_provisioning_link_enabled|ccr_agent_proxy_provisioning_link_generated|ccr_agent_proxy_provisioning_link_revoked|ccr_agent_proxy_provisioning_link_submitted|ccr_agent_proxy_rule_created|ccr_agent_proxy_rule_deleted|ccr_agent_proxy_rule_updated|ccr_agent_slack_access_scope_created|ccr_agent_slack_access_scope_deleted|ccr_agent_slack_binding_created|ccr_agent_slack_binding_deleted|ccr_agent_updated|ccr_channel_manager_added|ccr_channel_manager_removed|ccr_role_channel_assignment_deleted|ccr_role_channel_assignment_updated|ccr_session_created|ccr_session_deleted|ccr_session_updated|ccr_slack_channel_joined|claude_artifact_access_failed|claude_artifact_commented|claude_artifact_comments_viewed|claude_artifact_created|claude_artifact_duplicated|claude_artifact_external_sharing_permission_updated|claude_artifact_published|claude_artifact_sharing_updated|claude_artifact_viewed|claude_chat_access_failed|claude_chat_created|claude_chat_deleted|claude_chat_deletion_failed|claude_chat_settings_updated|claude_chat_snapshot_created|claude_chat_snapshot_deleted|claude_chat_snapshot_viewed|claude_chat_sync_source_created|claude_chat_sync_source_deleted|claude_chat_sync_source_updated|claude_chat_updated|claude_chat_viewed|claude_code_credential_revoked|claude_code_review_config_updated|claude_code_review_repository_added|claude_code_review_repository_removed|claude_code_review_repository_updated|claude_code_runner_deleted|claude_code_runner_pool_created|claude_code_runner_pool_deleted|claude_code_runner_pool_secret_minted|claude_code_runner_pool_session_queue_updated|claude_code_runner_pool_updated|claude_code_security_center_config_updated|claude_code_security_scan_cancelled|claude_code_security_scan_created|claude_code_security_scan_project_member_updated|claude_code_security_scan_project_updated|claude_code_security_scan_project_visibility_updated|claude_code_security_scan_run_updated|claude_code_security_scan_schedule_deleted|claude_code_security_scan_schedule_updated|claude_code_security_vulnerability_deleted|claude_code_security_vulnerability_fix_session_created|claude_code_security_vulnerability_updated|claude_code_security_webhook_created|claude_code_security_webhook_deleted|claude_code_security_webhook_secret_updated|claude_code_security_webhook_updated|claude_code_team_memory_acl_updated|claude_code_team_memory_updated|claude_code_team_onboarding_guide_updated|claude_code_user_marketplaces_updated|claude_code_user_memory_updated|claude_code_user_plugins_updated|claude_code_user_settings_updated|claude_command_created|claude_command_deleted|claude_command_replaced|claude_enterprise_upgrade_credit_updated|claude_file_access_failed|claude_file_deleted|claude_file_exported|claude_file_uploaded|claude_file_viewed|claude_gdrive_integration_created|claude_gdrive_integration_deleted|claude_gdrive_integration_updated|claude_github_integration_created|claude_github_integration_deleted|claude_github_integration_updated|claude_organization_settings_updated|claude_plugin_created|claude_plugin_deleted|claude_plugin_disabled|claude_plugin_enabled|claude_plugin_replaced|claude_plugin_security_scan_completed|claude_plugin_updated|claude_project_archived|claude_project_created|claude_project_deleted|claude_project_document_access_failed|claude_project_document_bulk_deletion_audit_truncated|claude_project_document_deleted|claude_project_document_deletion_failed|claude_project_document_updated|claude_project_document_uploaded|claude_project_document_viewed|claude_project_file_access_failed|claude_project_file_bulk_deletion_audit_truncated|claude_project_file_deleted|claude_project_file_deletion_failed|claude_project_file_uploaded|claude_project_reported|claude_project_sharing_updated|claude_project_sync_source_created|claude_project_sync_source_deleted|claude_project_sync_source_updated|claude_project_viewed|claude_published_artifact_deleted|claude_pubsec_identity_configured|claude_skill_created|claude_skill_deleted|claude_skill_disabled|claude_skill_enabled|claude_skill_replaced|claude_skill_security_scan_completed|claude_user_role_updated|claude_user_seat_tier_updated|claude_user_settings_updated|cli_plugin_exec_policy_updated|compliance_api_accessed|cowork_session_updated|design_project_artifact_published|design_project_created|design_project_deleted|design_project_member_added|design_project_member_removed|design_project_member_role_updated|design_project_published|design_project_sharing_updated|design_project_unpublished|design_project_updated|design_project_version_restored|design_project_viewed|desktop_extension_allowlisted|desktop_extension_blocklisted|desktop_extension_deleted|desktop_extension_removed_from_allowlist|desktop_extension_unblocked|desktop_extension_uploaded|desktop_extension_version_uploaded|domain_claim_initiated|end_user_invite_requested|extra_usage_billing_enabled|extra_usage_credit_granted|extra_usage_spend_limit_created|extra_usage_spend_limit_deleted|extra_usage_spend_limit_increase_request_approved|extra_usage_spend_limit_increase_request_denied|extra_usage_spend_limit_updated|ghe_configuration_created|ghe_configuration_deleted|ghe_configuration_updated|ghe_user_connected|ghe_user_disconnected|ghe_webhook_signature_invalid|github_app_installation_linked|github_app_installation_unlinked|github_token_import|group_created|group_deleted|group_list_viewed|group_member_added|group_member_addition_failed|group_member_list_viewed|group_member_removal_failed|group_member_removed|group_project_shares_revoked|group_skill_shares_revoked|group_updated|group_viewed|group_visibility_updated|inference_hooks_circuit_breaker_tripped|inference_hooks_config_deleted|inference_hooks_config_updated|inference_hooks_request_denied|inference_hooks_request_failed_open|inference_hooks_signing_secret_generated|integration_user_connected|integration_user_disconnected|invoice_collection_method_updated|lti_launch_initiated|lti_launch_success|lti_platform_created|lti_platform_updated|magic_link_login_failed|magic_link_login_initiated|magic_link_login_succeeded|managed_organization_setup_completed|marketplace_created|marketplace_deleted|marketplace_updated|marketplace_webhook_deleted|marketplace_webhook_provisioned|mcp_directory_server_published|mcp_server_created|mcp_server_deleted|mcp_server_managed_auth_token_exchanged|mcp_server_managed_auth_updated|mcp_server_updated|mcp_tool_policy_updated|org_analytics_api_capability_updated|org_bulk_delete_initiated|org_capability_grant_added|org_capability_grant_removed|org_claude_code_data_sharing_disabled|org_claude_code_data_sharing_enabled|org_claude_code_desktop_disabled|org_claude_code_desktop_enabled|org_claude_code_zero_data_retention_disabled|org_compliance_api_settings_updated|org_connector_domain_guard_updated|org_cowork_act_without_asking_mode_disabled|org_cowork_act_without_asking_mode_enabled|org_cowork_agent_disabled|org_cowork_agent_enabled|org_cowork_auto_mode_disabled|org_cowork_auto_mode_enabled|org_cowork_browser_pane_disabled|org_cowork_browser_pane_enabled|org_cowork_disabled|org_cowork_enabled|org_cowork_mcp_always_allow_disabled|org_cowork_mcp_always_allow_enabled|org_cowork_otlp_settings_updated|org_cowork_remote_disabled|org_cowork_remote_enabled|org_creation_blocked|org_data_export_accessed|org_data_export_completed|org_data_export_started|org_data_residency_updated|org_deleted_via_bulk|org_deletion_requested|org_directory_resync_completed|org_directory_resync_failed|org_directory_resync_started|org_directory_sync_activated|org_directory_sync_add_initiated|org_directory_sync_deleted|org_discoverability_disabled|org_discoverability_enabled|org_discoverability_settings_updated|org_domain_add_initiated|org_domain_removed|org_domain_verified|org_external_key_created|org_external_key_deleted|org_external_key_updated|org_external_key_validated|org_hipaa_self_serve_enabled|org_invite_link_disabled|org_invite_link_generated|org_invite_link_regenerated|org_invite_viewed|org_invites_listed|org_ip_restriction_created|org_ip_restriction_deleted|org_ip_restriction_updated|org_join_proposal_decided|org_join_request_approved|org_join_request_created|org_join_request_dismissed|org_join_request_instant_approved|org_join_requests_bulk_dismissed|org_magic_link_second_factor_toggled|org_member_invites_disabled|org_member_invites_enabled|org_members_exported|org_model_default_updated|org_parent_join_proposal_created|org_parent_search_performed|org_sso_add_initiated|org_sso_connection_activated|org_sso_connection_deactivated|org_sso_connection_deleted|org_sso_group_role_mappings_updated|org_sso_provisioning_mode_changed|org_sso_scim_welcome_email_toggled|org_sso_seat_tier_assignment_toggled|org_sso_seat_tier_mappings_updated|org_sso_toggled|org_sync_deleting_synchronized_files_started|org_sync_synchronized_files_deleted|org_taint_added|org_taint_removed|org_user_deleted|org_user_invite_accepted|org_user_invite_deleted|org_user_invite_re_sent|org_user_invite_rejected|org_user_invite_sent|org_user_left|org_user_shares_retained|org_user_trusted_devices_revoked|org_user_viewed|org_users_listed|org_work_across_apps_disabled|org_work_across_apps_enabled|organization_address_updated|organization_icon_deleted|organization_icon_updated|owned_projects_access_restored|payment_method_updated|pending_share_created|pending_share_revoked|phone_code_sent|phone_code_verified|platform_agent_archived|platform_agent_created|platform_agent_deleted|platform_agent_deployment_archived|platform_agent_deployment_created|platform_agent_deployment_deleted|platform_agent_deployment_paused|platform_agent_deployment_run_triggered|platform_agent_deployment_unpaused|platform_agent_deployment_updated|platform_agent_session_archived|platform_agent_session_created|platform_agent_session_deleted|platform_agent_session_resource_added|platform_agent_session_resource_deleted|platform_agent_session_resource_updated|platform_agent_session_thread_archived|platform_agent_session_updated|platform_agent_updated|platform_api_key_created|platform_api_key_updated|platform_app_attest_authentication|platform_billing_upgraded_to_prepaid|platform_clearance_workspace_program_request_cleared|platform_clearance_workspace_program_request_set|platform_cost_report_viewed|platform_dream_archived|platform_dream_cancelled|platform_dream_created|platform_federated_authentication|platform_federation_issuer_archived|platform_federation_issuer_created|platform_federation_issuer_updated|platform_federation_rule_archived|platform_federation_rule_created|platform_federation_rule_updated|platform_federation_rule_workspace_added|platform_federation_rule_workspace_removed|platform_file_content_downloaded|platform_file_deleted|platform_file_uploaded|platform_memory_created|platform_memory_deleted|platform_memory_store_archived|platform_memory_store_created|platform_memory_store_deleted|platform_memory_store_updated|platform_memory_updated|platform_memory_version_redacted|platform_oauth_app_created|platform_oauth_app_revoked|platform_oauth_app_updated|platform_plugin_directory_submission_created|platform_plugin_directory_submission_deleted|platform_plugin_directory_submission_updated|platform_service_account_archived|platform_service_account_created|platform_service_account_updated|platform_service_account_workspace_member_added|platform_service_account_workspace_member_removed|platform_service_account_workspace_member_updated|platform_signing_key_created|platform_signing_key_deleted|platform_signing_key_rotated|platform_skill_version_content_downloaded|platform_skill_version_created|platform_skill_version_deleted|platform_spend_limit_alert_emails_updated|platform_spend_limit_created|platform_spend_limit_deleted|platform_spend_limit_updated|platform_usage_report_claude_code_viewed|platform_usage_report_messages_viewed|platform_workspace_archived|platform_workspace_created|platform_workspace_inference_data_retention_disabled|platform_workspace_inference_data_retention_enabled|platform_workspace_member_added|platform_workspace_member_removed|platform_workspace_member_updated|platform_workspace_member_viewed|platform_workspace_members_listed|platform_workspace_rate_limit_deleted|platform_workspace_rate_limit_updated|platform_workspace_updated|plugin_installation_preference_updated|prepaid_auto_recharge_disabled|prepaid_auto_recharge_updated|prepaid_extra_usage_auto_reload_disabled|prepaid_extra_usage_auto_reload_enabled|prepaid_extra_usage_auto_reload_settings_updated|primary_owner_transferred|rbac_role_assigned|rbac_role_created|rbac_role_deleted|rbac_role_grant_updated|rbac_role_permission_added|rbac_role_permission_removed|rbac_role_unassigned|rbac_role_updated|role_assignment_granted|role_assignment_revoked|scim_user_created|scim_user_deleted|scim_user_updated|scoped_api_key_deleted|scoped_api_key_updated|seat_tier_changes_cancelled|seat_tiers_purchased|service_created|service_deleted|service_key_created|service_key_revoked|session_revoked|session_share_accessed|session_share_created|session_share_revoked|slack_workspace_claim_revoked|slack_workspace_claimed|social_login_succeeded|sso_login_failed|sso_login_initiated|sso_login_succeeded|sso_second_factor_magic_link|step_up_authentication_failed|step_up_authentication_succeeded|step_up_credential_enrolled|subscription_cancellation_scheduled|subscription_quantity_updated|subscription_renewed|subscription_resumed|subscription_started|subscription_upgraded|trusted_device_credential_rotated|trusted_device_enrolled|trusted_device_revoked|tunnel_archived|tunnel_certificate_added|tunnel_certificate_revoked|tunnel_created|tunnel_token_minted|tunnel_token_revealed|tunnel_token_revoked|tunnel_token_rotated|user_consent_recorded|user_consent_revoked|user_logged_out|verification_evidence_submitted|verification_program_application_created|workspace_member_spend_limit_created|workspace_member_spend_limit_deleted|workspace_member_spend_limit_updated|workspace_spend_limit_alert_emails_updated|workspace_spend_limit_created|workspace_spend_limit_deleted,,,DOCUMENTED,Filter activities by type. See the response `data` schema for the additional fields each type returns. Cannot be combined with `exclude_activity_types[]`.,https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,actor_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter activities by actor IDs (currently only `user_...` IDs are supported). Enumerate IDs via `GET /v1/compliance/organizations/{org_uuid}/users`.,https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,after_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the next page of results. To paginate, pass the `last_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,before_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the previous page of results. To paginate, pass the `first_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,created_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter activities created after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter activities created at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter activities created before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,created_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter activities created at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,exclude_activity_types[],query,"array of ""abuse_decision_received"" or ""account_deleted"" or ""admin_api_key_created"" or 503 more",false,,,,abuse_decision_received|account_deleted|admin_api_key_created|admin_api_key_deleted|admin_api_key_updated|admin_connector_request_resolved|admin_request_created|admin_setup_checklist_step_delegated|admin_setup_checklist_step_delegation_cancelled|age_verified|anonymous_mobile_login_attempted|api_key_created|audit_log_export_accessed|audit_log_export_started|billing_emails_updated|ccr_agent_created|ccr_agent_deleted|ccr_agent_proxy_anthropic_oidc_token_exchanged|ccr_agent_proxy_anthropic_oidc_token_minted|ccr_agent_proxy_credential_created|ccr_agent_proxy_credential_deleted|ccr_agent_proxy_credential_rotated|ccr_agent_proxy_credential_updated|ccr_agent_proxy_destination_deleted|ccr_agent_proxy_network_events_listed|ccr_agent_proxy_profile_bound|ccr_agent_proxy_profile_created|ccr_agent_proxy_profile_deleted|ccr_agent_proxy_profile_unbound|ccr_agent_proxy_profile_updated|ccr_agent_proxy_provisioning_credential_rejected|ccr_agent_proxy_provisioning_link_enabled|ccr_agent_proxy_provisioning_link_generated|ccr_agent_proxy_provisioning_link_revoked|ccr_agent_proxy_provisioning_link_submitted|ccr_agent_proxy_rule_created|ccr_agent_proxy_rule_deleted|ccr_agent_proxy_rule_updated|ccr_agent_slack_access_scope_created|ccr_agent_slack_access_scope_deleted|ccr_agent_slack_binding_created|ccr_agent_slack_binding_deleted|ccr_agent_updated|ccr_channel_manager_added|ccr_channel_manager_removed|ccr_role_channel_assignment_deleted|ccr_role_channel_assignment_updated|ccr_session_created|ccr_session_deleted|ccr_session_updated|ccr_slack_channel_joined|claude_artifact_access_failed|claude_artifact_commented|claude_artifact_comments_viewed|claude_artifact_created|claude_artifact_duplicated|claude_artifact_external_sharing_permission_updated|claude_artifact_published|claude_artifact_sharing_updated|claude_artifact_viewed|claude_chat_access_failed|claude_chat_created|claude_chat_deleted|claude_chat_deletion_failed|claude_chat_settings_updated|claude_chat_snapshot_created|claude_chat_snapshot_deleted|claude_chat_snapshot_viewed|claude_chat_sync_source_created|claude_chat_sync_source_deleted|claude_chat_sync_source_updated|claude_chat_updated|claude_chat_viewed|claude_code_credential_revoked|claude_code_review_config_updated|claude_code_review_repository_added|claude_code_review_repository_removed|claude_code_review_repository_updated|claude_code_runner_deleted|claude_code_runner_pool_created|claude_code_runner_pool_deleted|claude_code_runner_pool_secret_minted|claude_code_runner_pool_session_queue_updated|claude_code_runner_pool_updated|claude_code_security_center_config_updated|claude_code_security_scan_cancelled|claude_code_security_scan_created|claude_code_security_scan_project_member_updated|claude_code_security_scan_project_updated|claude_code_security_scan_project_visibility_updated|claude_code_security_scan_run_updated|claude_code_security_scan_schedule_deleted|claude_code_security_scan_schedule_updated|claude_code_security_vulnerability_deleted|claude_code_security_vulnerability_fix_session_created|claude_code_security_vulnerability_updated|claude_code_security_webhook_created|claude_code_security_webhook_deleted|claude_code_security_webhook_secret_updated|claude_code_security_webhook_updated|claude_code_team_memory_acl_updated|claude_code_team_memory_updated|claude_code_team_onboarding_guide_updated|claude_code_user_marketplaces_updated|claude_code_user_memory_updated|claude_code_user_plugins_updated|claude_code_user_settings_updated|claude_command_created|claude_command_deleted|claude_command_replaced|claude_enterprise_upgrade_credit_updated|claude_file_access_failed|claude_file_deleted|claude_file_exported|claude_file_uploaded|claude_file_viewed|claude_gdrive_integration_created|claude_gdrive_integration_deleted|claude_gdrive_integration_updated|claude_github_integration_created|claude_github_integration_deleted|claude_github_integration_updated|claude_organization_settings_updated|claude_plugin_created|claude_plugin_deleted|claude_plugin_disabled|claude_plugin_enabled|claude_plugin_replaced|claude_plugin_security_scan_completed|claude_plugin_updated|claude_project_archived|claude_project_created|claude_project_deleted|claude_project_document_access_failed|claude_project_document_bulk_deletion_audit_truncated|claude_project_document_deleted|claude_project_document_deletion_failed|claude_project_document_updated|claude_project_document_uploaded|claude_project_document_viewed|claude_project_file_access_failed|claude_project_file_bulk_deletion_audit_truncated|claude_project_file_deleted|claude_project_file_deletion_failed|claude_project_file_uploaded|claude_project_reported|claude_project_sharing_updated|claude_project_sync_source_created|claude_project_sync_source_deleted|claude_project_sync_source_updated|claude_project_viewed|claude_published_artifact_deleted|claude_pubsec_identity_configured|claude_skill_created|claude_skill_deleted|claude_skill_disabled|claude_skill_enabled|claude_skill_replaced|claude_skill_security_scan_completed|claude_user_role_updated|claude_user_seat_tier_updated|claude_user_settings_updated|cli_plugin_exec_policy_updated|compliance_api_accessed|cowork_session_updated|design_project_artifact_published|design_project_created|design_project_deleted|design_project_member_added|design_project_member_removed|design_project_member_role_updated|design_project_published|design_project_sharing_updated|design_project_unpublished|design_project_updated|design_project_version_restored|design_project_viewed|desktop_extension_allowlisted|desktop_extension_blocklisted|desktop_extension_deleted|desktop_extension_removed_from_allowlist|desktop_extension_unblocked|desktop_extension_uploaded|desktop_extension_version_uploaded|domain_claim_initiated|end_user_invite_requested|extra_usage_billing_enabled|extra_usage_credit_granted|extra_usage_spend_limit_created|extra_usage_spend_limit_deleted|extra_usage_spend_limit_increase_request_approved|extra_usage_spend_limit_increase_request_denied|extra_usage_spend_limit_updated|ghe_configuration_created|ghe_configuration_deleted|ghe_configuration_updated|ghe_user_connected|ghe_user_disconnected|ghe_webhook_signature_invalid|github_app_installation_linked|github_app_installation_unlinked|github_token_import|group_created|group_deleted|group_list_viewed|group_member_added|group_member_addition_failed|group_member_list_viewed|group_member_removal_failed|group_member_removed|group_project_shares_revoked|group_skill_shares_revoked|group_updated|group_viewed|group_visibility_updated|inference_hooks_circuit_breaker_tripped|inference_hooks_config_deleted|inference_hooks_config_updated|inference_hooks_request_denied|inference_hooks_request_failed_open|inference_hooks_signing_secret_generated|integration_user_connected|integration_user_disconnected|invoice_collection_method_updated|lti_launch_initiated|lti_launch_success|lti_platform_created|lti_platform_updated|magic_link_login_failed|magic_link_login_initiated|magic_link_login_succeeded|managed_organization_setup_completed|marketplace_created|marketplace_deleted|marketplace_updated|marketplace_webhook_deleted|marketplace_webhook_provisioned|mcp_directory_server_published|mcp_server_created|mcp_server_deleted|mcp_server_managed_auth_token_exchanged|mcp_server_managed_auth_updated|mcp_server_updated|mcp_tool_policy_updated|org_analytics_api_capability_updated|org_bulk_delete_initiated|org_capability_grant_added|org_capability_grant_removed|org_claude_code_data_sharing_disabled|org_claude_code_data_sharing_enabled|org_claude_code_desktop_disabled|org_claude_code_desktop_enabled|org_claude_code_zero_data_retention_disabled|org_compliance_api_settings_updated|org_connector_domain_guard_updated|org_cowork_act_without_asking_mode_disabled|org_cowork_act_without_asking_mode_enabled|org_cowork_agent_disabled|org_cowork_agent_enabled|org_cowork_auto_mode_disabled|org_cowork_auto_mode_enabled|org_cowork_browser_pane_disabled|org_cowork_browser_pane_enabled|org_cowork_disabled|org_cowork_enabled|org_cowork_mcp_always_allow_disabled|org_cowork_mcp_always_allow_enabled|org_cowork_otlp_settings_updated|org_cowork_remote_disabled|org_cowork_remote_enabled|org_creation_blocked|org_data_export_accessed|org_data_export_completed|org_data_export_started|org_data_residency_updated|org_deleted_via_bulk|org_deletion_requested|org_directory_resync_completed|org_directory_resync_failed|org_directory_resync_started|org_directory_sync_activated|org_directory_sync_add_initiated|org_directory_sync_deleted|org_discoverability_disabled|org_discoverability_enabled|org_discoverability_settings_updated|org_domain_add_initiated|org_domain_removed|org_domain_verified|org_external_key_created|org_external_key_deleted|org_external_key_updated|org_external_key_validated|org_hipaa_self_serve_enabled|org_invite_link_disabled|org_invite_link_generated|org_invite_link_regenerated|org_invite_viewed|org_invites_listed|org_ip_restriction_created|org_ip_restriction_deleted|org_ip_restriction_updated|org_join_proposal_decided|org_join_request_approved|org_join_request_created|org_join_request_dismissed|org_join_request_instant_approved|org_join_requests_bulk_dismissed|org_magic_link_second_factor_toggled|org_member_invites_disabled|org_member_invites_enabled|org_members_exported|org_model_default_updated|org_parent_join_proposal_created|org_parent_search_performed|org_sso_add_initiated|org_sso_connection_activated|org_sso_connection_deactivated|org_sso_connection_deleted|org_sso_group_role_mappings_updated|org_sso_provisioning_mode_changed|org_sso_scim_welcome_email_toggled|org_sso_seat_tier_assignment_toggled|org_sso_seat_tier_mappings_updated|org_sso_toggled|org_sync_deleting_synchronized_files_started|org_sync_synchronized_files_deleted|org_taint_added|org_taint_removed|org_user_deleted|org_user_invite_accepted|org_user_invite_deleted|org_user_invite_re_sent|org_user_invite_rejected|org_user_invite_sent|org_user_left|org_user_shares_retained|org_user_trusted_devices_revoked|org_user_viewed|org_users_listed|org_work_across_apps_disabled|org_work_across_apps_enabled|organization_address_updated|organization_icon_deleted|organization_icon_updated|owned_projects_access_restored|payment_method_updated|pending_share_created|pending_share_revoked|phone_code_sent|phone_code_verified|platform_agent_archived|platform_agent_created|platform_agent_deleted|platform_agent_deployment_archived|platform_agent_deployment_created|platform_agent_deployment_deleted|platform_agent_deployment_paused|platform_agent_deployment_run_triggered|platform_agent_deployment_unpaused|platform_agent_deployment_updated|platform_agent_session_archived|platform_agent_session_created|platform_agent_session_deleted|platform_agent_session_resource_added|platform_agent_session_resource_deleted|platform_agent_session_resource_updated|platform_agent_session_thread_archived|platform_agent_session_updated|platform_agent_updated|platform_api_key_created|platform_api_key_updated|platform_app_attest_authentication|platform_billing_upgraded_to_prepaid|platform_clearance_workspace_program_request_cleared|platform_clearance_workspace_program_request_set|platform_cost_report_viewed|platform_dream_archived|platform_dream_cancelled|platform_dream_created|platform_federated_authentication|platform_federation_issuer_archived|platform_federation_issuer_created|platform_federation_issuer_updated|platform_federation_rule_archived|platform_federation_rule_created|platform_federation_rule_updated|platform_federation_rule_workspace_added|platform_federation_rule_workspace_removed|platform_file_content_downloaded|platform_file_deleted|platform_file_uploaded|platform_memory_created|platform_memory_deleted|platform_memory_store_archived|platform_memory_store_created|platform_memory_store_deleted|platform_memory_store_updated|platform_memory_updated|platform_memory_version_redacted|platform_oauth_app_created|platform_oauth_app_revoked|platform_oauth_app_updated|platform_plugin_directory_submission_created|platform_plugin_directory_submission_deleted|platform_plugin_directory_submission_updated|platform_service_account_archived|platform_service_account_created|platform_service_account_updated|platform_service_account_workspace_member_added|platform_service_account_workspace_member_removed|platform_service_account_workspace_member_updated|platform_signing_key_created|platform_signing_key_deleted|platform_signing_key_rotated|platform_skill_version_content_downloaded|platform_skill_version_created|platform_skill_version_deleted|platform_spend_limit_alert_emails_updated|platform_spend_limit_created|platform_spend_limit_deleted|platform_spend_limit_updated|platform_usage_report_claude_code_viewed|platform_usage_report_messages_viewed|platform_workspace_archived|platform_workspace_created|platform_workspace_inference_data_retention_disabled|platform_workspace_inference_data_retention_enabled|platform_workspace_member_added|platform_workspace_member_removed|platform_workspace_member_updated|platform_workspace_member_viewed|platform_workspace_members_listed|platform_workspace_rate_limit_deleted|platform_workspace_rate_limit_updated|platform_workspace_updated|plugin_installation_preference_updated|prepaid_auto_recharge_disabled|prepaid_auto_recharge_updated|prepaid_extra_usage_auto_reload_disabled|prepaid_extra_usage_auto_reload_enabled|prepaid_extra_usage_auto_reload_settings_updated|primary_owner_transferred|rbac_role_assigned|rbac_role_created|rbac_role_deleted|rbac_role_grant_updated|rbac_role_permission_added|rbac_role_permission_removed|rbac_role_unassigned|rbac_role_updated|role_assignment_granted|role_assignment_revoked|scim_user_created|scim_user_deleted|scim_user_updated|scoped_api_key_deleted|scoped_api_key_updated|seat_tier_changes_cancelled|seat_tiers_purchased|service_created|service_deleted|service_key_created|service_key_revoked|session_revoked|session_share_accessed|session_share_created|session_share_revoked|slack_workspace_claim_revoked|slack_workspace_claimed|social_login_succeeded|sso_login_failed|sso_login_initiated|sso_login_succeeded|sso_second_factor_magic_link|step_up_authentication_failed|step_up_authentication_succeeded|step_up_credential_enrolled|subscription_cancellation_scheduled|subscription_quantity_updated|subscription_renewed|subscription_resumed|subscription_started|subscription_upgraded|trusted_device_credential_rotated|trusted_device_enrolled|trusted_device_revoked|tunnel_archived|tunnel_certificate_added|tunnel_certificate_revoked|tunnel_created|tunnel_token_minted|tunnel_token_revealed|tunnel_token_revoked|tunnel_token_rotated|user_consent_recorded|user_consent_revoked|user_logged_out|verification_evidence_submitted|verification_program_application_created|workspace_member_spend_limit_created|workspace_member_spend_limit_deleted|workspace_member_spend_limit_updated|workspace_spend_limit_alert_emails_updated|workspace_spend_limit_created|workspace_spend_limit_deleted,,,DOCUMENTED,Exclude activities of these types. Cannot be combined with `activity_types[]`.,https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,limit,query,number,false,100,1,5000,,,,DOCUMENTED,"Maximum results (default: 100, max: 5000)",https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,order,query,"""asc"" or ""desc""",false,desc,,,asc|desc,,,DOCUMENTED,"Sort direction by `created_at`. `desc` (default) returns newest-first; `asc` returns oldest-first for incremental sync. Activities become queryable after a short asynchronous ingestion delay. When using `asc` with `after_id` for incremental sync, late-arriving rows with timestamps behind the cursor will be skipped; consumers that need at-least-once delivery should periodically re-poll an overlap window via `created_at.gte` and deduplicate by `id`. `after_id` and `before_id` are relative to this order.",https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,organization_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter activities by organization IDs (accepts `org_...` or organization UUID). Enumerate IDs via `GET /v1/compliance/organizations`.,https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/activities,user_ids[],query,array of string,false,,,,,,,DOCUMENTED,"Alias for `actor_ids[]`, for consistency with other compliance routes. If both are provided, the lists are merged.",https://platform.claude.com/docs/en/api/compliance/activities/list
anthropic,GET /v1/compliance/apps/artifacts/{artifact_version_id}/content,artifact_version_id,path,string,true,,,,,,,DOCUMENTED,"The artifact version ID (tagged ID, e.g., claude_artifact_version_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/artifacts/download
anthropic,GET /v1/compliance/apps/artifacts/{artifact_version_id},artifact_version_id,path,string,true,,,,,,,DOCUMENTED,"The artifact version ID (tagged ID, e.g., claude_artifact_version_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/artifacts/retrieve
anthropic,DELETE /v1/compliance/apps/chats/{claude_chat_id},claude_chat_id,path,string,true,,,,,,,DOCUMENTED,"The chat ID (tagged ID, e.g., claude_chat_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/chats/delete
anthropic,DELETE /v1/compliance/apps/chats/files/{claude_file_id},claude_file_id,path,string,true,,,,,,,DOCUMENTED,"The file ID (tagged ID, e.g., claude_file_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/chats/files/delete
anthropic,GET /v1/compliance/apps/chats/files/{claude_file_id}/content,claude_file_id,path,string,true,,,,,,,DOCUMENTED,"The file ID (tagged ID, e.g., claude_file_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/chats/files/download
anthropic,GET /v1/compliance/apps/chats/files/{claude_file_id},claude_file_id,path,string,true,,,,,,,DOCUMENTED,"The file ID (tagged ID, e.g., claude_file_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/chats/files/retrieve
anthropic,GET /v1/compliance/apps/chats/generated-files/{claude_gen_file_id}/content,claude_gen_file_id,path,string,true,,,,,,,DOCUMENTED,"The generated-file id (e.g., 'claude_gen_file_abc123') as returned in `chat_messages[].generated_files[].id` from GET /apps/chats/{claude_chat_id}/messages.",https://platform.claude.com/docs/en/api/compliance/apps/chats/generated_files/download
anthropic,GET /v1/compliance/apps/chats/generated-files/{claude_gen_file_id},claude_gen_file_id,path,string,true,,,,,,,DOCUMENTED,"The generated-file id (e.g., 'claude_gen_file_abc123') as returned in `chat_messages[].generated_files[].id` from GET /apps/chats/{claude_chat_id}/messages.",https://platform.claude.com/docs/en/api/compliance/apps/chats/generated_files/retrieve
anthropic,GET /v1/compliance/apps/chats,after_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the next page of results. To paginate, pass the `last_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,before_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the previous page of results. To paginate, pass the `first_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,created_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter chats created after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter chats created at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter chats created before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,created_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter chats created at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,limit,query,number,false,100,1,1000,,,,DOCUMENTED,"Maximum results (default: 100, max: 1000)",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,order_by,query,"""created_at"" or ""updated_at""",false,created_at,,,created_at|updated_at,,,DOCUMENTED,"Sort key for results. `created_at` (default) sorts by chat creation time. `updated_at` sorts by last update time and is only supported for org-wide queries (omit user_ids[]). For org-wide queries, any time filter must match the sort key: `created_at.*` filters require `order_by=created_at`, and `updated_at.*` filters require `order_by=updated_at`.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,organization_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter by organization IDs (accepts `org_...` or organization UUID). Enumerate IDs via `GET /v1/compliance/organizations`.,https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,project_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter by project IDs (accepts `claude_proj_...`). Enumerate IDs via `GET /v1/compliance/apps/projects`. Requires user_ids[]; not supported for org-wide queries.,https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,updated_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,updated_at.gt,query,string,false,,,,,,,DOCUMENTED,"Filter chats updated after this time (RFC 3339 format). Combining updated_at filters with `user_ids[]` is deprecated and will be rejected after 2026-09-22; for updated_at-windowed polling, omit `user_ids[]` and use `order_by=updated_at` with `after_id` pagination.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,updated_at.gte,query,string,false,,,,,,,DOCUMENTED,"Filter chats updated at or after this time (RFC 3339 format). Combining updated_at filters with `user_ids[]` is deprecated and will be rejected after 2026-09-22; for updated_at-windowed polling, omit `user_ids[]` and use `order_by=updated_at` with `after_id` pagination.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,updated_at.lt,query,string,false,,,,,,,DOCUMENTED,"Filter chats updated before this time (RFC 3339 format). Combining updated_at filters with `user_ids[]` is deprecated and will be rejected after 2026-09-22; for updated_at-windowed polling, omit `user_ids[]` and use `order_by=updated_at` with `after_id` pagination.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,updated_at.lte,query,string,false,,,,,,,DOCUMENTED,"Filter chats updated at or before this time (RFC 3339 format). Combining updated_at filters with `user_ids[]` is deprecated and will be rejected after 2026-09-22; for updated_at-windowed polling, omit `user_ids[]` and use `order_by=updated_at` with `after_id` pagination.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats,user_ids[],query,array of string,false,,,,,,,DOCUMENTED,"Filter to chats created by specific users (max 10 per request). Omit for an org-wide query. Enumerate IDs via `GET /v1/compliance/organizations/{org_uuid}/users`. Deprecated combination: passing `user_ids[]` together with any `updated_at.*` filter is deprecated and will be rejected after 2026-09-22. For `updated_at`-windowed polling, omit `user_ids[]` and use `order_by=updated_at` with `after_id` pagination.",https://platform.claude.com/docs/en/api/compliance/apps/chats/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,claude_chat_id,path,string,true,,,,,,,DOCUMENTED,"The chat ID (tagged ID, e.g., claude_chat_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,after_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the next page of results. To paginate, pass the `last_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,before_id,query,string,false,,,,,,,DOCUMENTED,"Pagination cursor for retrieving the previous page of results. To paginate, pass the `first_id` value from the most recent response. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,created_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter messages created after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter messages created at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter messages created before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,created_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter messages created at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,limit,query,number,false,,1,1000,,,,DOCUMENTED,"Maximum results (max: 1000). When omitted, the full result set is returned in one response.",https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,order,query,"""asc"" or ""desc""",false,asc,,,asc|desc,,,DOCUMENTED,Sort direction for messages within the response. `asc` (the default) returns oldest-first; `desc` returns newest-first.,https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,tool_result_max_chars,query,number,false,10000,-1,,,,,DOCUMENTED,Maximum characters returned per tool-result text item. Items longer than this are shortened and the block's `truncated` field is set. Pass -1 to disable the limit.,https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,tool_use_input_max_chars,query,number,false,10000,-1,,,,,DOCUMENTED,Maximum characters of JSON-encoded tool input returned per tool_use block. Inputs longer than this are shortened and the block's `truncated` field is set. Pass -1 to disable the limit.,https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,updated_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,updated_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter messages updated after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,updated_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter messages updated at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,updated_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter messages updated before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/chats/{claude_chat_id}/messages,updated_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter messages updated at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/attachments,project_id,path,string,true,,,,,,,DOCUMENTED,"The project ID (tagged ID, e.g., claude_proj_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/attachments/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/attachments,limit,query,number,false,20,1,100,,,,DOCUMENTED,"Maximum results (default: 20, max: 100)",https://platform.claude.com/docs/en/api/compliance/apps/projects/attachments/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/attachments,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/projects/attachments/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/collaborators,project_id,path,string,true,,,,,,,DOCUMENTED,"The project ID (tagged ID, e.g., claude_proj_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/collaborators/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/collaborators,limit,query,number,false,20,1,100,,,,DOCUMENTED,"Maximum results (default: 20, max: 100)",https://platform.claude.com/docs/en/api/compliance/apps/projects/collaborators/list
anthropic,GET /v1/compliance/apps/projects/{project_id}/collaborators,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/projects/collaborators/list
anthropic,DELETE /v1/compliance/apps/projects/{project_id},project_id,path,string,true,,,,,,,DOCUMENTED,"The project ID (tagged ID, e.g., claude_proj_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/delete
anthropic,DELETE /v1/compliance/apps/projects/documents/{document_id},document_id,path,string,true,,,,,,,DOCUMENTED,"The document ID (tagged ID, e.g., claude_proj_doc_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/documents/delete
anthropic,GET /v1/compliance/apps/projects/documents/{document_id}/metadata,document_id,path,string,true,,,,,,,DOCUMENTED,"The document ID (tagged ID, e.g., claude_proj_doc_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/documents/metadata
anthropic,GET /v1/compliance/apps/projects/documents/{document_id},document_id,path,string,true,,,,,,,DOCUMENTED,"The document ID (tagged ID, e.g., claude_proj_doc_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/documents/retrieve
anthropic,GET /v1/compliance/apps/projects,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,created_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter projects created after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter projects created at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter projects created before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,created_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter projects created at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,limit,query,number,false,20,1,100,,,,DOCUMENTED,"Maximum results (default: 20, max: 100)",https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,organization_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter by organization IDs (accepts `org_...` or organization UUID). Enumerate IDs via `GET /v1/compliance/organizations`.,https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,updated_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,updated_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter projects updated after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,updated_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter projects updated at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,updated_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter projects updated before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,updated_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter projects updated at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects,user_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter by user IDs. Enumerate IDs via `GET /v1/compliance/organizations/{org_uuid}/users`.,https://platform.claude.com/docs/en/api/compliance/apps/projects/list
anthropic,GET /v1/compliance/apps/projects/{project_id},project_id,path,string,true,,,,,,,DOCUMENTED,"The project ID (tagged ID, e.g., claude_proj_abc123)",https://platform.claude.com/docs/en/api/compliance/apps/projects/retrieve
anthropic,GET /v1/compliance/apps/sessions/local,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Only return sessions whose first inference call is at or after this time (RFC 3339; a UTC offset is required).,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Only return sessions whose first inference call is strictly before this time (RFC 3339; a UTC offset is required).,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,limit,query,number,false,100,1,500,,,,DOCUMENTED,"Maximum results (default: 100, max: 500)",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,updated_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local,updated_at.gte,query,string,false,,,,,,,DOCUMENTED,"Only return sessions whose last inference call is at or after this time (RFC 3339; a UTC offset is required). Combines with `created_at.gte` / `created_at.lt`; the ordering and pagination are unchanged. Use it to poll for sessions that have been active since a previous pass — a session that becomes active later can only enter the result, never leave it.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,local_session_id,path,string,true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,limit,query,number,false,100,1,1000,,,,DOCUMENTED,"Maximum results (default: 100, max: 1000)",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,order,query,"""asc"" or ""desc""",false,asc,,,asc|desc,,,DOCUMENTED,"Sort direction. `asc` (oldest-first, default) or `desc`.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,tool_result_max_bytes,query,number,false,10000,-1,2147483647,,,,DOCUMENTED,Truncate each text item inside a tool result to at most this many bytes (cut on a code-point boundary). Pass `-1` to request the server maximum (approximately 1 MiB); larger values are clamped to it. `0` is not a valid value.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id}/messages,tool_use_input_max_bytes,query,number,false,10000,-1,2147483647,,,,DOCUMENTED,Truncate each tool-use input to at most this many bytes (cut on a code-point boundary so the result is valid UTF-8). Pass `-1` to request the server maximum (approximately 1 MiB); larger values are clamped to it. `0` is not a valid value.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/messages/list
anthropic,GET /v1/compliance/apps/sessions/local/{local_session_id},local_session_id,path,string,true,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/sessions/local/retrieve
anthropic,GET /v1/compliance/apps/sessions/remote,created_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,created_at.gt,query,string,false,,,,,,,DOCUMENTED,Filter remote sessions created after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,created_at.gte,query,string,false,,,,,,,DOCUMENTED,Filter remote sessions created at or after this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,created_at.lt,query,string,false,,,,,,,DOCUMENTED,Filter remote sessions created before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,created_at.lte,query,string,false,,,,,,,DOCUMENTED,Filter remote sessions created at or before this time (RFC 3339 format),https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,limit,query,number,false,100,1,500,,,,DOCUMENTED,"Maximum results (default: 100, max: 500)",https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,organization_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter to specific child organization identifiers. Omit to enumerate every child organization the key may read.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote,user_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter to sessions owned by specific users (max 10 per request). Agent-owned sessions are excluded when this filter is set.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,claude_remote_session_id,path,string,true,,,,,,,DOCUMENTED,The remote session identifier (`cse_...`) to retrieve,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,limit,query,number,false,100,1,1000,,,,DOCUMENTED,"Maximum results (default: 100, max: 1000)",https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,order,query,"""asc"" or ""desc""",false,asc,,,asc|desc,,,DOCUMENTED,Sort direction. `asc` (oldest-first) or `desc`.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,tool_result_max_bytes,query,number,false,10000,-1,2147483647,,,,DOCUMENTED,Truncate each text item inside a tool result to at most this many bytes (cut on a code-point boundary). Pass `-1` to request the server maximum. `0` is not a valid value.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,GET /v1/compliance/apps/sessions/remote/{claude_remote_session_id}/messages,tool_use_input_max_bytes,query,number,false,10000,-1,2147483647,,,,DOCUMENTED,Truncate each tool-use input to at most this many bytes (cut on a code-point boundary so the result is valid UTF-8). Pass `-1` to request the server maximum. `0` is not a valid value.,https://platform.claude.com/docs/en/api/compliance/apps/sessions/remote/messages/list
anthropic,DELETE /v1/compliance/apps/code/artifacts/{artifact_id},artifact_id,path,string,true,,,,,,,DOCUMENTED,"The Artifact ID (tagged ID, e.g., cart_abc123)",https://platform.claude.com/docs/en/api/compliance/code/artifacts/delete
anthropic,GET /v1/compliance/apps/code/artifacts,limit,query,number,false,20,1,100,,,,DOCUMENTED,"Maximum results (default: 20, max: 100)",https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,organization_ids[],query,array of string,false,,,,,,,DOCUMENTED,"Filter by organization IDs (accepts `org_...` or organization UUID, up to 500). Enumerate IDs via `GET /v1/compliance/organizations`.",https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,updated_at,query,object,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,updated_at.gt,query,string,false,,,,,,,DOCUMENTED,Return only Artifacts updated after this time (RFC 3339 format). See `updated_at.gte` for the completeness caveat.,https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,updated_at.gte,query,string,false,,,,,,,DOCUMENTED,"Return only Artifacts updated at or after this time (RFC 3339 format). Time filters match an eventually-consistent index and Artifacts published before this field was recorded never match — omit the time filter for compliance-complete enumeration. For incremental export, apply a generous overlap margin between windows and dedupe by `id`: adjacent tiling silently misses items whose index update lagged their publish.",https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,updated_at.lt,query,string,false,,,,,,,DOCUMENTED,Return only Artifacts updated before this time (RFC 3339 format). Multiple time operators are AND-ed to the tightest bound. See `updated_at.gte` for the completeness caveat.,https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,updated_at.lte,query,string,false,,,,,,,DOCUMENTED,Return only Artifacts updated at or before this time (RFC 3339 format). See `updated_at.gte` for the completeness caveat.,https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts,user_ids[],query,array of string,false,,,,,,,DOCUMENTED,Filter by owner user IDs (up to 200). Enumerate IDs via `GET /v1/compliance/organizations/{org_uuid}/users`.,https://platform.claude.com/docs/en/api/compliance/code/artifacts/list
anthropic,GET /v1/compliance/apps/code/artifacts/{artifact_id}/versions/{version_id},artifact_id,path,string,true,,,,,,,DOCUMENTED,"The Artifact ID (tagged ID, e.g., cart_abc123)",https://platform.claude.com/docs/en/api/compliance/code/artifacts/retrieve_version
anthropic,GET /v1/compliance/apps/code/artifacts/{artifact_id}/versions/{version_id},version_id,path,string,true,,,,,,,DOCUMENTED,Opaque version identifier from the Artifact's `versions` list,https://platform.claude.com/docs/en/api/compliance/code/artifacts/retrieve_version
anthropic,GET /v1/compliance/groups,limit,query,number,false,500,1,1000,,,,DOCUMENTED,"Maximum results (default: 500, max: 1000)",https://platform.claude.com/docs/en/api/compliance/groups/list
anthropic,GET /v1/compliance/groups,name_prefix,query,string,false,"""""",,,,,,DOCUMENTED,Filter groups by name prefix,https://platform.claude.com/docs/en/api/compliance/groups/list
anthropic,GET /v1/compliance/groups,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/groups/list
anthropic,GET /v1/compliance/groups/{group_id}/members,group_id,path,string,true,,,,,,,DOCUMENTED,"The group ID (tagged ID, e.g., rbac_group_abc123)",https://platform.claude.com/docs/en/api/compliance/groups/members/list
anthropic,GET /v1/compliance/groups/{group_id}/members,limit,query,number,false,500,1,1000,,,,DOCUMENTED,"Maximum results (default: 500, max: 1000)",https://platform.claude.com/docs/en/api/compliance/groups/members/list
anthropic,GET /v1/compliance/groups/{group_id}/members,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/groups/members/list
anthropic,GET /v1/compliance/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,"The group ID (tagged ID, e.g., rbac_group_abc123)",https://platform.claude.com/docs/en/api/compliance/groups/retrieve
anthropic,GET /v1/compliance/organizations,limit,query,number,false,1000,1,1000,,,,DOCUMENTED,"Maximum results (default: 1000, max: 1000)",https://platform.claude.com/docs/en/api/compliance/organizations/list
anthropic,GET /v1/compliance/organizations,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/organizations/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles,org_uuid,path,string,true,,,,,,,DOCUMENTED,The organization UUID,https://platform.claude.com/docs/en/api/compliance/organizations/roles/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles,limit,query,number,false,500,1,1000,,,,DOCUMENTED,"Maximum results (default: 500, max: 1000)",https://platform.claude.com/docs/en/api/compliance/organizations/roles/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/organizations/roles/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id}/permissions,org_uuid,path,string,true,,,,,,,DOCUMENTED,The organization UUID,https://platform.claude.com/docs/en/api/compliance/organizations/roles/permissions/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id}/permissions,role_id,path,string,true,,,,,,,DOCUMENTED,"The role ID (tagged ID, e.g., rbac_role_abc123)",https://platform.claude.com/docs/en/api/compliance/organizations/roles/permissions/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id}/permissions,limit,query,number,false,500,1,1000,,,,DOCUMENTED,"Maximum results (default: 500, max: 1000)",https://platform.claude.com/docs/en/api/compliance/organizations/roles/permissions/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id}/permissions,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/organizations/roles/permissions/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id},org_uuid,path,string,true,,,,,,,DOCUMENTED,The organization UUID,https://platform.claude.com/docs/en/api/compliance/organizations/roles/retrieve
anthropic,GET /v1/compliance/organizations/{org_uuid}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,"The role ID (tagged ID, e.g., rbac_role_abc123)",https://platform.claude.com/docs/en/api/compliance/organizations/roles/retrieve
anthropic,GET /v1/compliance/organizations/{organization_id}/settings,organization_id,path,string,true,,,,,,,DOCUMENTED,The organization's UUID,https://platform.claude.com/docs/en/api/compliance/organizations/settings/retrieve
anthropic,GET /v1/compliance/organizations/{org_uuid}/users,org_uuid,path,string,true,,,,,,,DOCUMENTED,The organization UUID,https://platform.claude.com/docs/en/api/compliance/organizations/users/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/users,limit,query,number,false,500,1,1000,,,,DOCUMENTED,"Maximum results (default: 500, max: 1000)",https://platform.claude.com/docs/en/api/compliance/organizations/users/list
anthropic,GET /v1/compliance/organizations/{org_uuid}/users,page,query,string,false,,,,,,,DOCUMENTED,"Opaque pagination token from a previous response's `next_page` field. Pass this to retrieve the next page of results. Clients should treat this value as an opaque string and not attempt to parse or interpret its contents, as the format may change without notice.",https://platform.claude.com/docs/en/api/compliance/organizations/users/list
anthropic,POST /v1/organizations/compliance_settings,state.type,body,"""enabled"" | ""disabled""",true,,,,enabled|disabled,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/organization/compliance_settings/update
anthropic,POST /v1/messages/count_tokens,model,body,string,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Model whose tokenizer is used (Claude 4.7+/Fable/Mythos tokenizer counts ~30% higher).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,messages,body,array<MessageParam>,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Same shape as /v1/messages. image/document blocks must be base64 (url/file sources rejected).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,messages[].role,body,string,true,,,,user|assistant|system,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Turn role.,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,messages[].content,body,string|array<ContentBlockParam>,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Same content block union as /v1/messages (see anthropic-messages.json).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,system,body,string|array<TextBlockParam>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,System prompt (counted). Live: +6 tokens for a short system string.,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,tools,body,array<MessageCountTokensTool>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Client tools (and the advisor tool) are counted; other server tools → 400 'Server tools are not supported in the count_tokens endpoint: web_search_20250305' (live).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,tool_choice,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Counted (live: any adds +92 tokens over auto on haiku 4.5).,https://platform.claude.com/docs/en/build-with-claude/token-counting
anthropic,POST /v1/messages/count_tokens,thinking,body,object,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Thinking config (affects count; live enabled/1024 → 40 vs 11 tokens).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,output_config,body,object,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,effort / format (json_schema). Live: a small schema added ~140 tokens.,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,cache_control,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Accepted but ignored: token counting never uses the cache.,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,anthropic-beta,header,string,false,,,,,all current models unless noted,,DOCUMENTED,Optional beta header(s).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,anthropic-workspace-id,header,string,false,,,,,all current models unless noted,,DOCUMENTED,Workspace id.,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,POST /v1/messages/count_tokens,anthropic-user-profile-id,header,string,false,,,,,all current models unless noted,user-profiles-2026-09-04,DOCUMENTED|BETA,User profile attribution (user-profiles beta).,https://platform.claude.com/docs/en/api/messages/count_tokens
anthropic,GET /v1/agents,created_at[gte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return agents created at or after this time (inclusive).,https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,GET /v1/agents,created_at[lte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return agents created at or before this time (inclusive).,https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,GET /v1/agents,include_archived,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Include archived agents in results. Defaults to false.,https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,GET /v1/agents,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum results per page. Default 20, maximum 100.",https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,GET /v1/agents,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor from a previous response.,https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,GET /v1/agents,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/list
anthropic,POST /v1/agents,model,body,BetaManagedAgentsModel or BetaManagedAgentsModelConfigParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Model identifier. Accepts the [model string](https://platform.claude.com/docs/en/about-claude/models/overview#latest-models-comparison), e.g. `claude-opus-5`, or a `model_config` object for additional configuration control",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.id,body,BetaManagedAgentsModel,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort,body,"""low"" or ""medium"" or ""high"" or 2 more or BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or 3 more or null",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"How hard Claude works on each inference call. Accepts a bare level string (`""high""`) or `{""type"": ""high""}`. On create, omitting it resolves the per-model default; on update, omitting it leaves the stored value unchanged.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort.type,body,"""low""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort.type,body,"""medium""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort.type,body,"""high""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort.type,body,"""xhigh""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.effort.type,body,"""max""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.inference_geo,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Geographic region for model inference. When unset, requests fall through to the workspace's default_inference_geo. On update, `model` is whole-object replacement — omitting inference_geo clears it.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,model.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,managed-agents-2026-04-01,DOCUMENTED|BETA,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name for the agent.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Description of what the agent does.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,mcp_servers,body,array of BetaManagedAgentsURLMCPServerParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,MCP servers this agent connects to. Maximum 20. Names must be unique within the array. Every server must be referenced by an `mcp_toolset` in `tools`; unreferenced servers are rejected. See the [MCP connector guide](https://platform.claude.com/docs/en/managed-agents/mcp-connector).,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,mcp_servers[].type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,mcp_servers[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,mcp_servers[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Endpoint URL for the MCP server.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Arbitrary key-value metadata. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent,body,BetaManagedAgentsMultiagentParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the `agents` roster.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.type,body,"""coordinator""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents,body,array of BetaManagedAgentsMultiagentRosterEntryParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Agents the coordinator may spawn as session threads. 1–20 entries. Each entry is an agent ID string, a versioned `{""type"":""agent"",""id"",""version""}` reference, or `{""type"":""self""}` to allow recursive self-invocation. Entries must reference distinct agents (after resolving `self` and string forms); at most one `self`. Referenced agents must exist, must not be archived, and must not themselves have `multiagent` set (depth limit 1).",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].type,body,"""agent""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The `agent` ID.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The specific `agent` version to use. Omit to use the latest version. Must be at least 1 if specified.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].type,body,"""self""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].type,body,"""advisor""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,multiagent.agents[].model,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A Claude model id. The model must be permitted as an advisor for this agent's model — see the sessions/threads/advisor spec.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,skills,body,array of BetaManagedAgentsSkillParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Skills available to the agent.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,skills[].type,body,"""anthropic""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,skills[].skill_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Tagged ID of the custom skill (e.g., ""skill_01XJ5..."").",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,skills[].version,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Version to pin. Defaults to latest if omitted.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,skills[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,system,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,System prompt for the agent.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools,body,array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Tool configurations available to the agent. Maximum of 128 tools across all toolsets allowed.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].type,body,"""agent_toolset_20260401""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs,body,array of BetaManagedAgentsAgentToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""bash""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""bash""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""bash"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether this tool is enabled. Overrides the `default_config` setting.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].permission_policy.type,body,"""always_allow""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].permission_policy.type,body,"""always_ask""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].permission_policy.type,body,"""auto""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""edit""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""edit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""edit"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""read""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""read""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""read"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""write""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""write""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""write"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""glob""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""glob""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""glob"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""grep""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""grep""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""grep"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""web_fetch""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""web_fetch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_fetch"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].allowed_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Only return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""docs.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].blocked_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Never return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""ads.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].max_content_tokens,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of tokens of fetched text content to include in context per call. Does not apply to binary content such as PDFs.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].type,body,"""web_search""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,"""web_search""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_search"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location,body,BetaManagedAgentsUserLocation or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Approximate user location for search result localization.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location.type,body,"""approximate""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Location precision. Only ""approximate"" is supported.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location.city,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,City name.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location.country,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Two-letter ISO 3166-1 country code, uppercase.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location.region,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Region or state name.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].user_location.timezone,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"IANA timezone identifier, e.g. ""America/Los_Angeles"".",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].default_config,body,BetaManagedAgentsAgentToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools in a toolset.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].default_config.enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether tools are enabled by default. Defaults to true if not specified.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].default_config.permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].type,body,"""mcp_toolset""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].mcp_server_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs,body,array of BetaManagedAgentsMCPToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].configs[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP tool to configure. 1-128 characters.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].default_config,body,BetaManagedAgentsMCPToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools from an MCP server.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Description of what the tool does, shown to the agent to help it decide when to use the tool.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].input_schema,body,BetaManagedAgentsCustomToolInputSchema,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,JSON Schema for custom tool input parameters.,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].input_schema.type,body,"""object""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].input_schema.properties,body,map[unknown] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].input_schema.required,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,tools[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.",https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,POST /v1/agents,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/create
anthropic,GET /v1/agents/{agent_id},agent_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/retrieve
anthropic,GET /v1/agents/{agent_id},version,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Agent version. Omit for the most recent version. Must be at least 1 if specified.,https://platform.claude.com/docs/en/api/beta/agents/retrieve
anthropic,GET /v1/agents/{agent_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/retrieve
anthropic,POST /v1/agents/{agent_id},agent_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Description. Omit to preserve; send empty string or null to clear.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},mcp_servers,body,array of BetaManagedAgentsURLMCPServerParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,MCP servers. Full replacement. Omit to preserve; send empty array or `null` to clear. Names must be unique. Maximum 20. Every server must be referenced by an `mcp_toolset` in the agent's resulting `tools`; unreferenced servers are rejected. See the [MCP connector guide](https://platform.claude.com/docs/en/managed-agents/mcp-connector).,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},mcp_servers[].type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},mcp_servers[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},mcp_servers[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Endpoint URL for the MCP server.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},metadata,body,map[string] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve. The stored bag is limited to 16 keys (up to 64 chars each) with values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model,body,BetaManagedAgentsModel or BetaManagedAgentsModelConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Model identifier. Accepts the [model string](https://platform.claude.com/docs/en/about-claude/models/overview#latest-models-comparison), e.g. `claude-opus-5`, or a `model_config` object for additional configuration control. Omit to preserve. Cannot be cleared.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.id,body,BetaManagedAgentsModel,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort,body,"""low"" or ""medium"" or ""high"" or 2 more or BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or 3 more or null",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"How hard Claude works on each inference call. Accepts a bare level string (`""high""`) or `{""type"": ""high""}`. On create, omitting it resolves the per-model default; on update, omitting it leaves the stored value unchanged.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort.type,body,"""low""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort.type,body,"""medium""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort.type,body,"""high""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort.type,body,"""xhigh""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.effort.type,body,"""max""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.inference_geo,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Geographic region for model inference. When unset, requests fall through to the workspace's default_inference_geo. On update, `model` is whole-object replacement — omitting inference_geo clears it.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},model.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,managed-agents-2026-04-01,DOCUMENTED|BETA,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent,body,BetaManagedAgentsMultiagentParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the `agents` roster.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.type,body,"""coordinator""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents,body,array of BetaManagedAgentsMultiagentRosterEntryParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Agents the coordinator may spawn as session threads. 1–20 entries. Each entry is an agent ID string, a versioned `{""type"":""agent"",""id"",""version""}` reference, or `{""type"":""self""}` to allow recursive self-invocation. Entries must reference distinct agents (after resolving `self` and string forms); at most one `self`. Referenced agents must exist, must not be archived, and must not themselves have `multiagent` set (depth limit 1).",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].type,body,"""agent""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The `agent` ID.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The specific `agent` version to use. Omit to use the latest version. Must be at least 1 if specified.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].type,body,"""self""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].type,body,"""advisor""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},multiagent.agents[].model,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A Claude model id. The model must be permitted as an advisor for this agent's model — see the sessions/threads/advisor spec.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},name,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name. Must be non-empty. Omit to preserve. Cannot be cleared.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},skills,body,array of BetaManagedAgentsSkillParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Skills. Full replacement. Omit to preserve; send empty array or null to clear.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},skills[].type,body,"""anthropic""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},skills[].skill_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Tagged ID of the custom skill (e.g., ""skill_01XJ5..."").",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},skills[].version,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Version to pin. Defaults to latest if omitted.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},skills[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},system,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,System prompt. Omit to preserve; send empty string or null to clear.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools,body,array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Tool configurations available to the agent. Full replacement. Omit to preserve; send empty array or null to clear. Maximum of 128 tools across all toolsets allowed.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].type,body,"""agent_toolset_20260401""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs,body,array of BetaManagedAgentsAgentToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""bash""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""bash""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""bash"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether this tool is enabled. Overrides the `default_config` setting.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].permission_policy.type,body,"""always_allow""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].permission_policy.type,body,"""always_ask""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].permission_policy.type,body,"""auto""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""edit""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""edit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""edit"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""read""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""read""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""read"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""write""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""write""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""write"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""glob""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""glob""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""glob"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""grep""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""grep""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""grep"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""web_fetch""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""web_fetch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_fetch"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].allowed_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Only return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""docs.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].blocked_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Never return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""ads.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].max_content_tokens,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of tokens of fetched text content to include in context per call. Does not apply to binary content such as PDFs.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].type,body,"""web_search""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,"""web_search""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_search"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location,body,BetaManagedAgentsUserLocation or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Approximate user location for search result localization.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location.type,body,"""approximate""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Location precision. Only ""approximate"" is supported.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location.city,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,City name.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location.country,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Two-letter ISO 3166-1 country code, uppercase.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location.region,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Region or state name.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].user_location.timezone,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"IANA timezone identifier, e.g. ""America/Los_Angeles"".",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].default_config,body,BetaManagedAgentsAgentToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools in a toolset.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].default_config.enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether tools are enabled by default. Defaults to true if not specified.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].default_config.permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].type,body,"""mcp_toolset""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].mcp_server_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs,body,array of BetaManagedAgentsMCPToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].configs[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP tool to configure. 1-128 characters.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].default_config,body,BetaManagedAgentsMCPToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools from an MCP server.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Description of what the tool does, shown to the agent to help it decide when to use the tool.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].input_schema,body,BetaManagedAgentsCustomToolInputSchema,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,JSON Schema for custom tool input parameters.,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].input_schema.type,body,"""object""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].input_schema.properties,body,map[unknown] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].input_schema.required,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},tools[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"The agent's current version, used to prevent concurrent overwrites. Obtain this value from a create or retrieve response. Must be at least 1 if specified. When supplied, the request fails if it does not match the server's current version; omit to apply the update unconditionally.",https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/update
anthropic,POST /v1/agents/{agent_id}/archive,agent_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/archive
anthropic,POST /v1/agents/{agent_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/archive
anthropic,GET /v1/agents/{agent_id}/versions,agent_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/agents/versions/list
anthropic,GET /v1/agents/{agent_id}/versions,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum results per page. Default 20, maximum 100.",https://platform.claude.com/docs/en/api/beta/agents/versions/list
anthropic,GET /v1/agents/{agent_id}/versions,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor.,https://platform.claude.com/docs/en/api/beta/agents/versions/list
anthropic,GET /v1/agents/{agent_id}/versions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/agents/versions/list
anthropic,GET /v1/deployment_runs,created_at[gt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return runs created strictly after this time (exclusive).,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,created_at[gte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return runs created at or after this time (inclusive).,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,created_at[lt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return runs created strictly before this time (exclusive).,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,created_at[lte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return runs created at or before this time (inclusive).,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,deployment_id,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter to a specific deployment. Omit to list across all deployments in the workspace. Filtering by a non-existent `deployment_id` returns 200 with empty data.,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,has_error,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Filter: true for runs with non-null `error`, false for runs with non-null `session_id`. Omit for all.",https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum results per page. Default 20, maximum 1000.",https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor. Pass `next_page` from the previous response. Invalid or expired cursors return 400.,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,trigger_type,query,BetaManagedAgentsTriggerType,false,,,,schedule|manual,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter runs by what triggered them. Omit to return all runs.,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployment_runs/list
anthropic,GET /v1/deployment_runs/{deployment_run_id},deployment_run_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployment_runs/retrieve
anthropic,GET /v1/deployment_runs/{deployment_run_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployment_runs/retrieve
anthropic,GET /v1/deployments,agent_id,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter by agent ID.,https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,created_at[gte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return deployments created at or after this time (inclusive).,https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,created_at[lte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return deployments created at or before this time (inclusive).,https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,include_archived,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"When true, includes archived deployments. Default: false (exclude archived).",https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum results per page. Default 20, maximum 100.",https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor.,https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,status,query,BetaManagedAgentsDeploymentStatus,false,,,,active|paused,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Filter by status: `active` or `paused`. Omit for both. To include archived deployments, use `include_archived` instead; the two cannot be combined.",https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,GET /v1/deployments,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/list
anthropic,POST /v1/deployments,agent,body,string or BetaManagedAgentsAgentParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Agent to deploy. Accepts the `agent` ID string, which pins the latest version, or an `agent` object with both id and version specified. The agent must exist and not be archived.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,agent.type,body,"""agent""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,agent.id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The `agent` ID.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,agent.version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The specific `agent` version to use. Omit to use the latest version. Must be at least 1 if specified.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,environment_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the `environment` defining the container configuration for sessions created from this deployment.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events,body,array of BetaManagedAgentsDeploymentInitialEventParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Events to send to each session immediately after creation. At least 1, maximum 50.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].type,body,"""user.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content,body,array of BetaManagedAgentsTextBlock or BetaManagedAgentsImageBlock or BetaManagedAgentsDocumentBlock or BetaManagedAgentsRedactedBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Array of content blocks for the user message.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].text,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The text content.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].type,body,"""image""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source,body,BetaManagedAgentsBase64ImageSource or BetaManagedAgentsURLImageSource or BetaManagedAgentsFileImageSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for image source variants.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.type,body,"""base64""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.data,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The plain text content.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.media_type,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the document (e.g., ""application/pdf"").",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,URL of the document to fetch.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].type,body,"""document""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source,body,BetaManagedAgentsBase64DocumentSource or BetaManagedAgentsPlainTextDocumentSource or BetaManagedAgentsURLDocumentSource or BetaManagedAgentsFileDocumentSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for document source variants.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].source.media_type,body,"""text/plain""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the text content. Must be ""text/plain"".",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].context,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Additional context about the document for the model.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The title of the document.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content[].type,body,"""redacted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].type,body,"""user.define_outcome""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,What the agent should produce. This is the task specification.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].rubric,body,BetaManagedAgentsFileRubricParams or BetaManagedAgentsTextRubricParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric for grading the quality of an outcome.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].rubric.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].rubric.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the rubric file.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].rubric.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].rubric.content,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric content. Plain text or markdown — the grader treats it as freeform text. Maximum 262144 characters.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].max_iterations,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Eval→revision cycles before giving up. Default 3, max 20.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].type,body,"""system.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,initial_events[].content,body,array of BetaManagedAgentsSystemContentBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,System content blocks to append. Text-only.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name for the deployment.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,budget,body,BetaManagedAgentsBudgetLimit or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,budget.type,body,"""limit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,budget.max_list_cost,body,BetaMonetaryAmount,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A monetary amount in a specific currency.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,budget.max_list_cost.amount,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Amount in minor units of the currency, as an integer decimal string with no leading zeros: ""2500"" is $25.00 and ""50"" is fifty cents. A string rather than a number so no float rounding is ever applied.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,budget.max_list_cost.currency,body,BetaCurrency,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Description of what the deployment does.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Arbitrary key-value metadata. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources,body,array of BetaManagedAgentsGitHubRepositoryResourceParams or BetaManagedAgentsFileResourceParams or BetaManagedAgentsMemoryStoreResourceParam,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Resources (e.g. repositories, files) to mount into each session's container. Maximum 500.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].type,body,"""github_repository""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Github URL of the repository,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].authorization_token,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,GitHub authorization token used to clone the repository. Required for private repositories; optional for public ones.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].checkout,body,BetaManagedAgentsBranchCheckout or BetaManagedAgentsCommitCheckout or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch or commit to check out. Defaults to the repository's default branch.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].checkout.type,body,"""branch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].checkout.name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch name to check out.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].checkout.type,body,"""commit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].checkout.sha,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Full commit SHA to check out.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].mount_path,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Mount path in the container. Defaults to `/mnt/session/uploads/<file_id>`.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].type,body,"""memory_store""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].memory_store_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The memory store ID (memstore_...). Must belong to the caller's organization and workspace.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].access,body,"""read_write"" or ""read_only"" or null",false,,,,read_write|read_only,,managed-agents-2026-04-01,DOCUMENTED|BETA,Access mode for an attached memory store.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,resources[].instructions,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,schedule,body,BetaManagedAgentsScheduleParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,5-field POSIX cron schedule. Literal wall-clock matching in the configured timezone.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,schedule.type,body,"""cron""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,schedule.expression,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"5-field POSIX cron expression: minute hour day-of-month month day-of-week (e.g., ""0 9 * * 1-5"" for weekdays at 9am). Day-of-week is 0-7 where 0 and 7 both mean Sunday. Extended cron syntax - seconds or year fields, and the special characters L, W, #, and ? - is not supported, nor are predefined shortcuts (@daily).",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,schedule.timezone,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Required. IANA timezone identifier (e.g., ""America/Los_Angeles"", ""UTC""). Validated against the IANA timezone database.",https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,vault_ids,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Vault IDs for stored credentials the agent can use during sessions created from this deployment. Maximum 50.,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,POST /v1/deployments,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/create
anthropic,GET /v1/deployments/{deployment_id},deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/retrieve
anthropic,GET /v1/deployments/{deployment_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/retrieve
anthropic,POST /v1/deployments/{deployment_id},deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},agent,body,string or BetaManagedAgentsAgentParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Agent to deploy. Accepts the `agent` ID string, which re-pins to the latest version, or an `agent` object with both id and version specified. Omit to preserve. Cannot be cleared.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},agent.type,body,"""agent""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},agent.id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The `agent` ID.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},agent.version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The specific `agent` version to use. Omit to use the latest version. Must be at least 1 if specified.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},budget,body,BetaManagedAgentsBudgetLimit or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},budget.type,body,"""limit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},budget.max_list_cost,body,BetaMonetaryAmount,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A monetary amount in a specific currency.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},budget.max_list_cost.amount,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Amount in minor units of the currency, as an integer decimal string with no leading zeros: ""2500"" is $25.00 and ""50"" is fifty cents. A string rather than a number so no float rounding is ever applied.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},budget.max_list_cost.currency,body,BetaCurrency,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Description. Omit to preserve; send empty string or null to clear.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},environment_id,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the `environment` where sessions run. Omit to preserve. Cannot be cleared.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events,body,array of BetaManagedAgentsDeploymentInitialEventParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Initial events. Full replacement. Omit to preserve. Cannot be cleared. At least 1, maximum 50.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].type,body,"""user.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content,body,array of BetaManagedAgentsTextBlock or BetaManagedAgentsImageBlock or BetaManagedAgentsDocumentBlock or BetaManagedAgentsRedactedBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Array of content blocks for the user message.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].text,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The text content.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].type,body,"""image""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source,body,BetaManagedAgentsBase64ImageSource or BetaManagedAgentsURLImageSource or BetaManagedAgentsFileImageSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for image source variants.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.type,body,"""base64""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.data,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The plain text content.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.media_type,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the document (e.g., ""application/pdf"").",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,URL of the document to fetch.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].type,body,"""document""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source,body,BetaManagedAgentsBase64DocumentSource or BetaManagedAgentsPlainTextDocumentSource or BetaManagedAgentsURLDocumentSource or BetaManagedAgentsFileDocumentSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for document source variants.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].source.media_type,body,"""text/plain""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the text content. Must be ""text/plain"".",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].context,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Additional context about the document for the model.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The title of the document.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content[].type,body,"""redacted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].type,body,"""user.define_outcome""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,What the agent should produce. This is the task specification.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].rubric,body,BetaManagedAgentsFileRubricParams or BetaManagedAgentsTextRubricParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric for grading the quality of an outcome.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].rubric.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].rubric.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the rubric file.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].rubric.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].rubric.content,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric content. Plain text or markdown — the grader treats it as freeform text. Maximum 262144 characters.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].max_iterations,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Eval→revision cycles before giving up. Default 3, max 20.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].type,body,"""system.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},initial_events[].content,body,array of BetaManagedAgentsSystemContentBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,System content blocks to append. Text-only.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},metadata,body,map[string] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve. The stored bag is limited to 16 keys (up to 64 chars each) with values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},name,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name. Must be non-empty. Omit to preserve. Cannot be cleared.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources,body,array of BetaManagedAgentsGitHubRepositoryResourceParams or BetaManagedAgentsFileResourceParams or BetaManagedAgentsMemoryStoreResourceParam or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Session resources. Full replacement. Omit to preserve; send empty array or null to clear. Maximum 500.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].type,body,"""github_repository""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Github URL of the repository,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].authorization_token,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,GitHub authorization token used to clone the repository. Required for private repositories; optional for public ones.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].checkout,body,BetaManagedAgentsBranchCheckout or BetaManagedAgentsCommitCheckout or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch or commit to check out. Defaults to the repository's default branch.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].checkout.type,body,"""branch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].checkout.name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch name to check out.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].checkout.type,body,"""commit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].checkout.sha,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Full commit SHA to check out.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].mount_path,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Mount path in the container. Defaults to `/mnt/session/uploads/<file_id>`.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].type,body,"""memory_store""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].memory_store_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The memory store ID (memstore_...). Must belong to the caller's organization and workspace.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].access,body,"""read_write"" or ""read_only"" or null",false,,,,read_write|read_only,,managed-agents-2026-04-01,DOCUMENTED|BETA,Access mode for an attached memory store.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},resources[].instructions,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},schedule,body,BetaManagedAgentsScheduleParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,5-field POSIX cron schedule. Literal wall-clock matching in the configured timezone.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},schedule.type,body,"""cron""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},schedule.expression,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"5-field POSIX cron expression: minute hour day-of-month month day-of-week (e.g., ""0 9 * * 1-5"" for weekdays at 9am). Day-of-week is 0-7 where 0 and 7 both mean Sunday. Extended cron syntax - seconds or year fields, and the special characters L, W, #, and ? - is not supported, nor are predefined shortcuts (@daily).",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},schedule.timezone,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Required. IANA timezone identifier (e.g., ""America/Los_Angeles"", ""UTC""). Validated against the IANA timezone database.",https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},vault_ids,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Vault IDs. Full replacement. Omit to preserve; send empty array or null to clear. Maximum 50.,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/update
anthropic,POST /v1/deployments/{deployment_id}/archive,deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/archive
anthropic,POST /v1/deployments/{deployment_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/archive
anthropic,POST /v1/deployments/{deployment_id}/pause,deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/pause
anthropic,POST /v1/deployments/{deployment_id}/pause,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/pause
anthropic,POST /v1/deployments/{deployment_id}/run,deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/run
anthropic,POST /v1/deployments/{deployment_id}/run,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/run
anthropic,POST /v1/deployments/{deployment_id}/unpause,deployment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/deployments/unpause
anthropic,POST /v1/deployments/{deployment_id}/unpause,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/deployments/unpause
anthropic,GET /v1/dreams,created_at[gt],query,string,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,"Return dreams with `created_at` strictly after this timestamp (exclusive lower bound, RFC 3339). Unset applies no lower bound.",https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,created_at[lt],query,string,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,"Return dreams with `created_at` strictly before this timestamp (exclusive upper bound, RFC 3339). Unset applies no upper bound.",https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,include_archived,query,boolean,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,Query parameter for include_archived,https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,limit,query,number,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,Query parameter for limit,https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,page,query,string,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,Query parameter for page,https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,statuses,query,array of BetaDreamStatus,false,,,,pending|running|completed|failed|canceled,,dreaming-2026-04-21,DOCUMENTED|BETA,Filter by lifecycle status. Repeat the parameter to match any of multiple statuses. Empty applies no status filter.,https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,GET /v1/dreams,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/dreams/list
anthropic,POST /v1/dreams,inputs,body,array of BetaDreamInput,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,inputs[].type,body,"""memory_store""",true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,inputs[].memory_store_id,body,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,inputs[].type,body,"""sessions""",true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,inputs[].session_ids,body,array of string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,model,body,string or BetaDreamModelConfigParam,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,model.id,body,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,"Model identifier, e.g. ""claude-opus-5"". 1-256 characters.",https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,model.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,dreaming-2026-04-21,DOCUMENTED|BETA,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,instructions,body,string or null,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,output_behavior,body,BetaOutputBehavior,false,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,output_behavior.type,body,"""create_new""",true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,output_behavior.type,body,"""update_existing""",true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,output_behavior.memory_store_id,body,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,POST /v1/dreams,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/dreams/create
anthropic,GET /v1/dreams/{dream_id},dream_id,path,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/retrieve
anthropic,GET /v1/dreams/{dream_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/dreams/retrieve
anthropic,POST /v1/dreams/{dream_id}/archive,dream_id,path,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/archive
anthropic,POST /v1/dreams/{dream_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/dreams/archive
anthropic,POST /v1/dreams/{dream_id}/cancel,dream_id,path,string,true,,,,,,dreaming-2026-04-21,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/dreams/cancel
anthropic,POST /v1/dreams/{dream_id}/cancel,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/dreams/cancel
anthropic,GET /v1/environments,include_archived,query,boolean,false,false,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Include archived environments in the response,https://platform.claude.com/docs/en/api/beta/environments/list
anthropic,GET /v1/environments,limit,query,number,false,20,1,1000,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of environments to return,https://platform.claude.com/docs/en/api/beta/environments/list
anthropic,GET /v1/environments,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque cursor from previous response for pagination. Pass the `next_page` value from the previous response.,https://platform.claude.com/docs/en/api/beta/environments/list
anthropic,GET /v1/environments,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/list
anthropic,POST /v1/environments,name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name for the environment,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config,body,BetaCloudConfigParams or BetaSelfHostedConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Environment configuration,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.type,body,"""cloud""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Environment type,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking,body,BetaUnrestrictedNetwork or BetaLimitedNetworkParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network configuration policy. Omit on update to preserve the existing value.,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking.type,body,"""unrestricted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network policy type,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking.type,body,"""limited""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network policy type,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking.allow_mcp_servers,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Permits outbound access to MCP server endpoints configured on the agent, beyond those listed in the `allowed_hosts` array. Defaults to `false`.",https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking.allow_package_managers,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Permits outbound access to public package registries (PyPI, npm, etc.) beyond those listed in the `allowed_hosts` array. Defaults to `false` on creation. Must be `true` when `packages` are specified.",https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.networking.allowed_hosts,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Specifies domains the container can reach.,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages,body,BetaPackagesParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Specify packages (and optionally their versions) available in this environment. When versioning, use the version semantics relevant for the package manager, e.g. for `pip` use `package==1.0.0`. You are responsible for validating the package and version exist. Unversioned installs the latest. Under `limited` networking, requires `networking.allow_package_managers` to be `true`.",https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.type,body,"""packages""",false,packages,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Package configuration type,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.apt,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Ubuntu/Debian packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.cargo,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rust packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.gem,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Ruby packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.go,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Go packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.npm,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Node.js packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.packages.pip,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Python packages to install,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,config.type,body,"""self_hosted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Environment type,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Optional description of the environment,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,User-provided metadata key-value pairs,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,scope,body,"""organization"" or ""account"" or null",false,,,,organization|account,,managed-agents-2026-04-01,DOCUMENTED|BETA,"The visibility scope for this environment. 'organization' makes the environment visible to all accounts. 'account' restricts visibility to the owning account only. API organizations support only 'organization'; 'account' is rejected. If not specified, defaults based on organization type.",https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,POST /v1/environments,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/create
anthropic,DELETE /v1/environments/{environment_id},environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/delete
anthropic,DELETE /v1/environments/{environment_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/delete
anthropic,GET /v1/environments/{environment_id},environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/retrieve
anthropic,GET /v1/environments/{environment_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/retrieve
anthropic,POST /v1/environments/{environment_id},environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config,body,BetaCloudConfigParams or BetaSelfHostedConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated environment configuration,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.type,body,"""cloud""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Environment type,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking,body,BetaUnrestrictedNetwork or BetaLimitedNetworkParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network configuration policy. Omit on update to preserve the existing value.,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking.type,body,"""unrestricted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network policy type,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking.type,body,"""limited""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Network policy type,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking.allow_mcp_servers,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Permits outbound access to MCP server endpoints configured on the agent, beyond those listed in the `allowed_hosts` array. Defaults to `false`.",https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking.allow_package_managers,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Permits outbound access to public package registries (PyPI, npm, etc.) beyond those listed in the `allowed_hosts` array. Defaults to `false` on creation. Must be `true` when `packages` are specified.",https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.networking.allowed_hosts,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Specifies domains the container can reach.,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages,body,BetaPackagesParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Specify packages (and optionally their versions) available in this environment. When versioning, use the version semantics relevant for the package manager, e.g. for `pip` use `package==1.0.0`. You are responsible for validating the package and version exist. Unversioned installs the latest. Under `limited` networking, requires `networking.allow_package_managers` to be `true`.",https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.type,body,"""packages""",false,packages,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Package configuration type,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.apt,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Ubuntu/Debian packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.cargo,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rust packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.gem,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Ruby packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.go,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Go packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.npm,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Node.js packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.packages.pip,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Python packages to install,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},config.type,body,"""self_hosted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Environment type,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},description,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated description of the environment. Omit to preserve; null clears to null; an empty string is stored as an empty string.,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,User-provided metadata key-value pairs. Set a value to null or empty string to delete the key.,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},name,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated name for the environment,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},scope,body,"""organization"" or ""account"" or null",false,,,,organization|account,,managed-agents-2026-04-01,DOCUMENTED|BETA,The visibility scope for this environment. 'organization' makes the environment visible to all accounts. 'account' restricts visibility to the owning account only.,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/update
anthropic,POST /v1/environments/{environment_id}/archive,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/archive
anthropic,POST /v1/environments/{environment_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/archive
anthropic,GET /v1/environments/{environment_id}/work,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/list
anthropic,GET /v1/environments/{environment_id}/work,limit,query,number,false,20,1,1000,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of work items to return,https://platform.claude.com/docs/en/api/beta/environments/work/list
anthropic,GET /v1/environments/{environment_id}/work,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque cursor from previous response for pagination,https://platform.claude.com/docs/en/api/beta/environments/work/list
anthropic,GET /v1/environments/{environment_id}/work/poll,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/poll
anthropic,GET /v1/environments/{environment_id}/work/poll,block_ms,query,number,false,,1,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,How long to wait for work to arrive before returning. Must be 1-999 in milliseconds. Defaults to non-blocking (returns immediately if no work is available).,https://platform.claude.com/docs/en/api/beta/environments/work/poll
anthropic,GET /v1/environments/{environment_id}/work/poll,reclaim_older_than_ms,query,number,false,,1,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Reclaim unacknowledged work items older than this many milliseconds. If omitted, uses the default (5000ms).",https://platform.claude.com/docs/en/api/beta/environments/work/poll
anthropic,GET /v1/environments/{environment_id}/work/poll,Anthropic-Worker-ID,header,string,false,,,,,,,DOCUMENTED,"Unique identifier for the specific worker polling, used to track aggregated environment-level work metrics in Console",https://platform.claude.com/docs/en/api/beta/environments/work/poll
anthropic,GET /v1/environments/{environment_id}/work/stats,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/stats
anthropic,GET /v1/environments/{environment_id}/work/stats,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/work/stats
anthropic,GET /v1/environments/{environment_id}/work/{work_id},environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/retrieve
anthropic,GET /v1/environments/{environment_id}/work/{work_id},work_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/retrieve
anthropic,GET /v1/environments/{environment_id}/work/{work_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/work/retrieve
anthropic,POST /v1/environments/{environment_id}/work/{work_id},environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/update
anthropic,POST /v1/environments/{environment_id}/work/{work_id},work_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/update
anthropic,POST /v1/environments/{environment_id}/work/{work_id},metadata,body,map[string],true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve existing metadata.",https://platform.claude.com/docs/en/api/beta/environments/work/update
anthropic,POST /v1/environments/{environment_id}/work/{work_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/work/update
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/ack,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/ack
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/ack,work_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/ack
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/heartbeat,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/heartbeat
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/heartbeat,work_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/heartbeat
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/heartbeat,desired_ttl_seconds,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Desired TTL in seconds,https://platform.claude.com/docs/en/api/beta/environments/work/heartbeat
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/heartbeat,expected_last_heartbeat,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Expected last_heartbeat for conditional update (optimistic concurrency). Use literal 'NO_HEARTBEAT' to claim an unclaimed lease (first heartbeat). For subsequent heartbeats, echo the server's previous last_heartbeat value exactly. Returns 412 Precondition Failed if the actual value doesn't match.",https://platform.claude.com/docs/en/api/beta/environments/work/heartbeat
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/stop,environment_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/stop
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/stop,work_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/environments/work/stop
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/stop,force,body,boolean,false,false,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"If true, immediately stop work without graceful shutdown",https://platform.claude.com/docs/en/api/beta/environments/work/stop
anthropic,POST /v1/environments/{environment_id}/work/{work_id}/stop,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/environments/work/stop
anthropic,GET /v1/memory_stores,created_at[gte],query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Return only stores whose `created_at` is at or after this time (inclusive). Sent on the wire as `created_at[gte]`.,https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,GET /v1/memory_stores,created_at[lte],query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Return only stores whose `created_at` is at or before this time (inclusive). Sent on the wire as `created_at[lte]`.,https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,GET /v1/memory_stores,include_archived,query,boolean,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"When `true`, archived stores are included in the results. Defaults to `false` (archived stores are excluded).",https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,GET /v1/memory_stores,limit,query,number,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Maximum number of stores to return per page. Must be between 1 and 100. Defaults to 20 when omitted.,https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,GET /v1/memory_stores,page,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Opaque pagination cursor (a `page_...` value). Pass the `next_page` value from a previous response to fetch the next page; omit for the first page.,https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,GET /v1/memory_stores,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/list
anthropic,POST /v1/memory_stores,name,body,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Human-readable name for the store. Required; 1–255 characters; no control characters. The mount-path slug under `/mnt/memory/` is derived from this name (lowercased, non-alphanumeric runs collapsed to a hyphen). Names need not be unique within a workspace.",https://platform.claude.com/docs/en/api/beta/memory_stores/create
anthropic,POST /v1/memory_stores,description,body,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Free-text description of what the store contains, up to 1024 characters. Included in the agent's system prompt when the store is attached, so word it to be useful to the agent.",https://platform.claude.com/docs/en/api/beta/memory_stores/create
anthropic,POST /v1/memory_stores,metadata,body,map[string],false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Arbitrary key-value tags for your own bookkeeping (such as the end user a store belongs to). Up to 16 pairs; keys 1–64 characters; values up to 512 characters. Not visible to the agent.,https://platform.claude.com/docs/en/api/beta/memory_stores/create
anthropic,POST /v1/memory_stores,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/create
anthropic,DELETE /v1/memory_stores/{memory_store_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/delete
anthropic,DELETE /v1/memory_stores/{memory_store_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/delete
anthropic,GET /v1/memory_stores/{memory_store_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/retrieve
anthropic,POST /v1/memory_stores/{memory_store_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/update
anthropic,POST /v1/memory_stores/{memory_store_id},description,body,string or null,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"New description for the store, up to 1024 characters. Pass an empty string to clear it.",https://platform.claude.com/docs/en/api/beta/memory_stores/update
anthropic,POST /v1/memory_stores/{memory_store_id},metadata,body,map[string] or null,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve. The stored bag is limited to 16 keys (up to 64 chars each) with values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/memory_stores/update
anthropic,POST /v1/memory_stores/{memory_store_id},name,body,string or null,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,New human-readable name for the store. 1–255 characters; no control characters. Renaming changes the slug used for the store's `mount_path` in sessions created after the update.,https://platform.claude.com/docs/en/api/beta/memory_stores/update
anthropic,POST /v1/memory_stores/{memory_store_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/update
anthropic,POST /v1/memory_stores/{memory_store_id}/archive,memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/archive
anthropic,POST /v1/memory_stores/{memory_store_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/archive
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,depth,query,number,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,`0` (or omitted) returns all descendants below `path_prefix` (recursive). `1` returns immediate children only; deeper entries roll up as `memory_prefix` items. `depth=1` behaves like `ls`; omitting `depth` behaves like `find`.,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,limit,query,number,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Maximum number of items to return per page. Must be between 1 and 100. Defaults to 20 when omitted. Capped at 20 when `view=full`. Both `memory` and `memory_prefix` items count toward the limit.,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,page,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Opaque pagination cursor (a `page_...` value). Pass the `next_page` value from a previous response to fetch the next page; omit for the first page.,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,path_prefix,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Optional path prefix filter. Must end with `/` (segment-aligned), e.g., `/notes/`. This value appears in request URLs. Do not include secrets or personally identifiable information.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Which projection of each `memory` to return. Defaults to `basic` (content omitted). `full` populates `content` on each item and caps `limit` at 20; use this as the bulk-read path for export and sync.,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memories,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/list
anthropic,POST /v1/memory_stores/{memory_store_id}/memories,memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/create
anthropic,POST /v1/memory_stores/{memory_store_id}/memories,view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for view,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/create
anthropic,POST /v1/memory_stores/{memory_store_id}/memories,content,body,string or null,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"UTF-8 text content for the new memory. Maximum 100 kB (102,400 bytes). Required; pass `""""` explicitly to create an empty memory.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/create
anthropic,POST /v1/memory_stores/{memory_store_id}/memories,path,body,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Hierarchical path for the new memory, e.g. `/projects/foo/notes.md`. Must start with `/`, contain at least one non-empty segment, and be at most 1,024 bytes. Must not contain empty segments, `.` or `..` segments, control or format characters, or the Unicode line and paragraph separators (U+2028, U+2029), and must be NFC-normalized. Paths are case-sensitive.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/create
anthropic,POST /v1/memory_stores/{memory_store_id}/memories,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/create
anthropic,DELETE /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/delete
anthropic,DELETE /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/delete
anthropic,DELETE /v1/memory_stores/{memory_store_id}/memories/{memory_id},expected_content_sha256,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for expected_content_sha256,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/delete
anthropic,DELETE /v1/memory_stores/{memory_store_id}/memories/{memory_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/delete
anthropic,GET /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memories/{memory_id},view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for view,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memories/{memory_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/retrieve
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},memory_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for view,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},content,body,string or null,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"New UTF-8 text content for the memory. Maximum 100 kB (102,400 bytes). Omit to leave the content unchanged (e.g., for a rename-only update).",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},path,body,string or null,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"New path for the memory (a rename). Must start with `/`, contain at least one non-empty segment, and be at most 1,024 bytes. Must not contain empty segments, `.` or `..` segments, control or format characters, or the Unicode line and paragraph separators (U+2028, U+2029), and must be NFC-normalized. Paths are case-sensitive. The memory's `id` is preserved across renames. Omit to leave the path unchanged.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},precondition,body,BetaManagedAgentsPrecondition,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Optimistic-concurrency precondition: the update applies only if the memory's stored `content_sha256` equals the supplied value. On mismatch, the request returns `memory_precondition_failed_error` (HTTP 409); re-read the memory and retry against the fresh state. If the precondition fails but the stored state already exactly matches the requested `content` and `path`, the server returns 200 instead of 409.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},precondition.type,body,"""content_sha256""",true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},precondition.content_sha256,body,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,"Expected `content_sha256` of the stored memory (64 lowercase hexadecimal characters). Typically the `content_sha256` returned by a prior read or list call. Because the server applies no content normalization, clients can also compute this locally as the SHA-256 of the UTF-8 content bytes.",https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,POST /v1/memory_stores/{memory_store_id}/memories/{memory_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memories/update
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,api_key_id,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for api_key_id,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,created_at[gte],query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Return versions created at or after this time (inclusive).,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,created_at[lte],query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Return versions created at or before this time (inclusive).,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,limit,query,number,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for limit,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,memory_id,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for memory_id,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,operation,query,BetaManagedAgentsMemoryVersionOperation,false,,,,created|modified|deleted,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for operation,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,page,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for page,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,service_account_id,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for service_account_id,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,session_id,query,string,false,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for session_id,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for view,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/list
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id},memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id},memory_version_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id},view,query,BetaManagedAgentsMemoryView,false,,,,basic|full,,agent-memory-2026-07-22,DOCUMENTED|BETA,Query parameter for view,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/retrieve
anthropic,GET /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/retrieve
anthropic,POST /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id}/redact,memory_store_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/redact
anthropic,POST /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id}/redact,memory_version_id,path,string,true,,,,,,agent-memory-2026-07-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/redact
anthropic,POST /v1/memory_stores/{memory_store_id}/memory_versions/{memory_version_id}/redact,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/memory_stores/memory_versions/redact
anthropic,GET /v1/sessions,agent_id,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter sessions created with this agent ID.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,agent_version,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter by agent version. Only applies when `agent_id` is also set.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,created_at[gt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return sessions created after this time (exclusive).,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,created_at[gte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return sessions created at or after this time (inclusive).,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,created_at[lt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return sessions created before this time (exclusive).,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,created_at[lte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return sessions created at or before this time (inclusive).,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,deployment_id,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter sessions created by this deployment ID.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,include_archived,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"When true, includes archived sessions. Default: false (exclude archived).",https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of results to return.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,memory_store_id,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter sessions whose resources contain a `memory_store` with this memory store ID.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Sort direction for results, ordered by `created_at`. Defaults to `desc` (newest first).",https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor from a previous response.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,statuses,query,"array of ""rescheduling"" or ""running"" or ""idle"" or ""terminated""",false,,,,rescheduling|running|idle|terminated,,managed-agents-2026-04-01,DOCUMENTED|BETA,Filter by session status. Repeat the parameter to match any of multiple statuses.,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,GET /v1/sessions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/list
anthropic,POST /v1/sessions,agent,body,string or BetaManagedAgentsAgentParams or BetaManagedAgentsAgentWithOverridesParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Agent identifier. Accepts the `agent` ID string, which pins the latest version for the session, or an `agent` object with both id and version specified.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.type,body,"""agent""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The `agent` ID.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.version,body,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The specific `agent` version to use. Omit to use the latest version.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.type,body,"""agent_with_overrides""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.mcp_servers,body,array of BetaManagedAgentsURLMCPServerParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Replacement MCP server list. Full replacement: the provided array becomes the MCP servers. Send an empty array to clear; omit to preserve the agent's servers.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.mcp_servers[].type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.mcp_servers[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.mcp_servers[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Endpoint URL for the MCP server.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model,body,BetaManagedAgentsModel or BetaManagedAgentsModelConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Replacement model. Accepts the model string, e.g. `claude-opus-5`, or a `model_config` object. Omit to use the agent's model.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.id,body,BetaManagedAgentsModel,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort,body,"""low"" or ""medium"" or ""high"" or 2 more or BetaManagedAgentsEffortLow or BetaManagedAgentsEffortMedium or 3 more or null",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"How hard Claude works on each inference call. Accepts a bare level string (`""high""`) or `{""type"": ""high""}`. On create, omitting it resolves the per-model default; on update, omitting it leaves the stored value unchanged.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort.type,body,"""low""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort.type,body,"""medium""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort.type,body,"""high""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort.type,body,"""xhigh""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.effort.type,body,"""max""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.inference_geo,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Geographic region for model inference. When unset, requests fall through to the workspace's default_inference_geo. On update, `model` is whole-object replacement — omitting inference_geo clears it.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.model.speed,body,"""standard"" or ""fast"" or null",false,,,,standard|fast,,managed-agents-2026-04-01,DOCUMENTED|BETA,Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.skills,body,array of BetaManagedAgentsSkillParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Replacement skill list. Full replacement: the provided array becomes the skills. Send an empty array to clear; omit to preserve the agent's skills.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.skills[].type,body,"""anthropic""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.skills[].skill_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Tagged ID of the custom skill (e.g., ""skill_01XJ5..."").",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.skills[].version,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Version to pin. Defaults to latest if omitted.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.skills[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.system,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Replacement system prompt. Up to 100,000 characters. Set to null to clear the agent's system prompt; omit to preserve it.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools,body,array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Replacement tool list. Full replacement: the provided array becomes the tool configuration. Send an empty array to clear; omit to preserve the agent's tools.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].type,body,"""agent_toolset_20260401""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs,body,array of BetaManagedAgentsAgentToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""bash""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""bash""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""bash"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether this tool is enabled. Overrides the `default_config` setting.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""edit""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""edit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""edit"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""read""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""read""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""read"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""write""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""write""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""write"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""glob""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""glob""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""glob"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""grep""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""grep""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""grep"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""web_fetch""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""web_fetch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_fetch"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].allowed_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Only return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""docs.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].blocked_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Never return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""ads.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].max_content_tokens,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of tokens of fetched text content to include in context per call. Does not apply to binary content such as PDFs.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].type,body,"""web_search""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,"""web_search""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_search"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location,body,BetaManagedAgentsUserLocation or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Approximate user location for search result localization.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location.type,body,"""approximate""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Location precision. Only ""approximate"" is supported.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location.city,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,City name.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location.country,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Two-letter ISO 3166-1 country code, uppercase.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location.region,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Region or state name.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].user_location.timezone,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"IANA timezone identifier, e.g. ""America/Los_Angeles"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].default_config,body,BetaManagedAgentsAgentToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools in a toolset.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].default_config.enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether tools are enabled by default. Defaults to true if not specified.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].default_config.permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].type,body,"""mcp_toolset""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].mcp_server_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs,body,array of BetaManagedAgentsMCPToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].configs[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP tool to configure. 1-128 characters.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].default_config,body,BetaManagedAgentsMCPToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools from an MCP server.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Description of what the tool does, shown to the agent to help it decide when to use the tool.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].input_schema,body,BetaManagedAgentsCustomToolInputSchema,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,JSON Schema for custom tool input parameters.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].input_schema.type,body,"""object""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].input_schema.properties,body,map[unknown] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].input_schema.required,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,agent.tools[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,environment_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the `environment` defining the container configuration for this session.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,budget,body,BetaManagedAgentsBudgetLimit,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,budget.type,body,"""limit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,budget.max_list_cost,body,BetaMonetaryAmount,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A monetary amount in a specific currency.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,budget.max_list_cost.amount,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Amount in minor units of the currency, as an integer decimal string with no leading zeros: ""2500"" is $25.00 and ""50"" is fifty cents. A string rather than a number so no float rounding is ever applied.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,budget.max_list_cost.currency,body,BetaCurrency,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events,body,array of BetaManagedAgentsUserMessageEventParams or BetaManagedAgentsUserDefineOutcomeEventParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Initial events to send to the `session` at creation, processed in order. Supports `user.message` and `user.define_outcome` events. Maximum 50 events.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].type,body,"""user.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content,body,array of BetaManagedAgentsTextBlock or BetaManagedAgentsImageBlock or BetaManagedAgentsDocumentBlock or BetaManagedAgentsRedactedBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Array of content blocks for the user message.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].text,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The text content.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].type,body,"""image""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source,body,BetaManagedAgentsBase64ImageSource or BetaManagedAgentsURLImageSource or BetaManagedAgentsFileImageSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for image source variants.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.type,body,"""base64""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.data,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The plain text content.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.media_type,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the document (e.g., ""application/pdf"").",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,URL of the document to fetch.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].type,body,"""document""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source,body,BetaManagedAgentsBase64DocumentSource or BetaManagedAgentsPlainTextDocumentSource or BetaManagedAgentsURLDocumentSource or BetaManagedAgentsFileDocumentSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for document source variants.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].source.media_type,body,"""text/plain""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the text content. Must be ""text/plain"".",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].context,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Additional context about the document for the model.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The title of the document.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].content[].type,body,"""redacted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].type,body,"""user.define_outcome""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,What the agent should produce. This is the task specification.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].rubric,body,BetaManagedAgentsFileRubricParams or BetaManagedAgentsTextRubricParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric for grading the quality of an outcome.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].rubric.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].rubric.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the rubric file.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].rubric.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].rubric.content,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric content. Plain text or markdown — the grader treats it as freeform text. Maximum 262144 characters.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,initial_events[].max_iterations,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Eval→revision cycles before giving up. Default 3, max 20.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Arbitrary key-value metadata attached to the session. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources,body,array of BetaManagedAgentsGitHubRepositoryResourceParams or BetaManagedAgentsFileResourceParams or BetaManagedAgentsMemoryStoreResourceParam,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Resources (e.g. repositories, files) to mount into the session's container.",https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].type,body,"""github_repository""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Github URL of the repository,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].authorization_token,body,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,GitHub authorization token used to clone the repository. Required for private repositories; optional for public ones.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].checkout,body,BetaManagedAgentsBranchCheckout or BetaManagedAgentsCommitCheckout or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch or commit to check out. Defaults to the repository's default branch.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].checkout.type,body,"""branch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].checkout.name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Branch name to check out.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].checkout.type,body,"""commit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].checkout.sha,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Full commit SHA to check out.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].mount_path,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Mount path in the container. Defaults to `/mnt/session/uploads/<file_id>`.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].type,body,"""memory_store""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].memory_store_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The memory store ID (memstore_...). Must belong to the caller's organization and workspace.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].access,body,"""read_write"" or ""read_only"" or null",false,,,,read_write|read_only,,managed-agents-2026-04-01,DOCUMENTED|BETA,Access mode for an attached memory store.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,resources[].instructions,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-attachment guidance for the agent on how to use this store. Rendered into the memory section of the system prompt. Max 4096 chars.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable session title.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,vault_ids,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Vault IDs for stored credentials the agent can use during the session.,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,POST /v1/sessions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/create
anthropic,DELETE /v1/sessions/{session_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/delete
anthropic,DELETE /v1/sessions/{session_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/delete
anthropic,GET /v1/sessions/{session_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/retrieve
anthropic,GET /v1/sessions/{session_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/retrieve
anthropic,POST /v1/sessions/{session_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent,body,BetaManagedAgentsSessionAgentUpdate,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Mid-session agent configuration update. Only `tools` and `mcp_servers` are updatable. Full replacement: the provided array becomes the new value. To preserve existing entries, GET the session, modify the array, and POST it back.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.mcp_servers,body,array of BetaManagedAgentsURLMCPServerParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Replacement MCP server list. Full replacement: the provided array becomes the new value. Send an empty array to clear; omit to preserve.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.mcp_servers[].type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.mcp_servers[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.mcp_servers[].url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Endpoint URL for the MCP server.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools,body,array of BetaManagedAgentsAgentToolset20260401Params or BetaManagedAgentsMCPToolsetParams or BetaManagedAgentsCustomToolParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Replacement tool list. Full replacement: the provided array becomes the new value. Send an empty array to clear; omit to preserve.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].type,body,"""agent_toolset_20260401""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs,body,array of BetaManagedAgentsAgentToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""bash""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""bash""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""bash"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether this tool is enabled. Overrides the `default_config` setting.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""edit""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""edit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""edit"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""read""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""read""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""read"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""write""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""write""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""write"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""glob""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""glob""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""glob"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""grep""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""grep""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""grep"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""web_fetch""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""web_fetch""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_fetch"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].allowed_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Only return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""docs.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with blocked_domains.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].blocked_domains,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Never return search results whose host is one of these domains or a subdomain of one. Each entry is a plain hostname like ""ads.example.com"" (no scheme or port; an optional path suffix is accepted). At most 64 entries; an empty list is rejected (omit the field instead). Cannot be combined with allowed_domains.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].max_content_tokens,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum number of tokens of fetched text content to include in context per call. Does not apply to binary content such as PDFs.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].type,body,"""web_search""",false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,"""web_search""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Must be ""web_search"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location,body,BetaManagedAgentsUserLocation or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Approximate user location for search result localization.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location.type,body,"""approximate""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Location precision. Only ""approximate"" is supported.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location.city,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,City name.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location.country,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Two-letter ISO 3166-1 country code, uppercase.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location.region,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Region or state name.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].user_location.timezone,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"IANA timezone identifier, e.g. ""America/Los_Angeles"".",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].default_config,body,BetaManagedAgentsAgentToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools in a toolset.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].default_config.enabled,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether tools are enabled by default. Defaults to true if not specified.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].default_config.permission_policy,body,BetaManagedAgentsAlwaysAllowPolicy or BetaManagedAgentsAlwaysAskPolicy or BetaManagedAgentsAutoPolicy or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Permission policy for tool execution.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].type,body,"""mcp_toolset""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].mcp_server_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs,body,array of BetaManagedAgentsMCPToolConfigParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Per-tool configuration overrides.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].configs[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the MCP tool to configure. 1-128 characters.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].default_config,body,BetaManagedAgentsMCPToolsetDefaultConfigParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Default configuration for all tools from an MCP server.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].type,body,"""custom""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Description of what the tool does, shown to the agent to help it decide when to use the tool.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].input_schema,body,BetaManagedAgentsCustomToolInputSchema,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,JSON Schema for custom tool input parameters.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].input_schema.type,body,"""object""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].input_schema.properties,body,map[unknown] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].input_schema.required,body,array of string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},agent.tools[].name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},budget,body,BetaManagedAgentsBudgetLimit or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A hard spend ceiling. The session stops issuing new model requests once the tracked list cost reaches `max_list_cost`.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},budget.type,body,"""limit""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},budget.max_list_cost,body,BetaMonetaryAmount,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A monetary amount in a specific currency.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},budget.max_list_cost.amount,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Amount in minor units of the currency, as an integer decimal string with no leading zeros: ""2500"" is $25.00 and ""50"" is fifty cents. A string rather than a number so no float rounding is ever applied.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},budget.max_list_cost.currency,body,BetaCurrency,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Uppercase ISO-4217 currency code. `USD` is the only currency currently supported; the accepted set is closed and grows only when a new currency is priced.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},metadata,body,map[string] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omit the field to preserve.",https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable session title.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},vault_ids,body,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Vault IDs (`vlt_*`) to attach to the session. Not yet supported; requests setting this field are rejected. Reserved for future use.,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/update
anthropic,POST /v1/sessions/{session_id}/archive,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/archive
anthropic,POST /v1/sessions/{session_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/archive
anthropic,GET /v1/sessions/{session_id}/events,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,created_at[gt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return events created after this time (exclusive). Compared against the event's `processed_at` value.,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,created_at[gte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return events created at or after this time (inclusive). Compared against the event's `processed_at` value.,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,created_at[lt],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return events created before this time (exclusive). Compared against the event's `processed_at` value.,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,created_at[lte],query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Return events created at or before this time (inclusive). Compared against the event's `processed_at` value.,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Query parameter for limit,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Sort direction for results, ordered by the event's `processed_at`. Defaults to `asc` (chronological).",https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor from a previous response's `next_page`.,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,types,query,array of string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Filter by event type. Values match the `type` field on returned events (for example, `user.message` or `agent.tool_use`). Omit to return all event types.",https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,GET /v1/sessions/{session_id}/events,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/events/list
anthropic,POST /v1/sessions/{session_id}/events,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events,body,array of BetaManagedAgentsEventParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Events to send to the `session`.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content,body,array of BetaManagedAgentsTextBlock or BetaManagedAgentsImageBlock or BetaManagedAgentsDocumentBlock or BetaManagedAgentsRedactedBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Array of content blocks for the user message.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].text,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The text content.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].type,body,"""image""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source,body,BetaManagedAgentsBase64ImageSource or BetaManagedAgentsURLImageSource or BetaManagedAgentsFileImageSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for image source variants.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.type,body,"""base64""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.data,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The plain text content.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.media_type,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the document (e.g., ""application/pdf"").",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.type,body,"""url""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,URL of the document to fetch.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].type,body,"""document""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source,body,BetaManagedAgentsBase64DocumentSource or BetaManagedAgentsPlainTextDocumentSource or BetaManagedAgentsURLDocumentSource or BetaManagedAgentsFileDocumentSource,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Union type for document source variants.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source.media_type,body,"""text/plain""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"MIME type of the text content. Must be ""text/plain"".",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].context,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Additional context about the document for the model.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].title,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The title of the document.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].type,body,"""redacted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.interrupt""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].session_thread_id,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"If absent, interrupts every non-archived thread in a multiagent session (or the primary alone in a single-agent session). If present, interrupts only the named thread.",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.tool_confirmation""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].result,body,"""allow"" or ""deny""",true,,,,allow|deny,,managed-agents-2026-04-01,DOCUMENTED|BETA,UserToolConfirmationResult enum,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].tool_use_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"The id of the `agent.tool_use` event this result corresponds to, which can be found in the last `session.status_idle` [event's](https://platform.claude.com/docs/en/api/beta/sessions/events/list#beta_managed_agents_session_requires_action.event_ids) `stop_reason.event_ids` field.",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].deny_message,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Optional message providing context for a 'deny' decision. Only allowed when result is 'deny'.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.custom_tool_result""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].custom_tool_use_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"The id of the `agent.custom_tool_use` event this result corresponds to, which can be found in the last `session.status_idle` [event's](https://platform.claude.com/docs/en/api/beta/sessions/events/list#beta_managed_agents_session_requires_action.event_ids) `stop_reason.event_ids` field.",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content,body,array of BetaManagedAgentsTextBlock or BetaManagedAgentsImageBlock or BetaManagedAgentsDocumentBlock or BetaManagedAgentsSearchResultBlock,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The result content returned by the tool.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].type,body,"""search_result""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].citations,body,BetaManagedAgentsSearchResultCitations,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Citation settings for a search result.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].citations.enabled,body,boolean,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether citations are enabled for this search result.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].content,body,array of BetaManagedAgentsSearchResultContent,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Array of text content blocks from the search result.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].content[].type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].content[].text,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The text content.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].source,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The URL source of the search result.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content[].title,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,The title of the search result.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].is_error,body,boolean or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether the tool execution resulted in an error.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.define_outcome""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].description,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,What the agent should produce. This is the task specification.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].rubric,body,BetaManagedAgentsFileRubricParams or BetaManagedAgentsTextRubricParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric for grading the quality of an outcome.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].rubric.type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].rubric.file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of the rubric file.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].rubric.type,body,"""text""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].rubric.content,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Rubric content. Plain text or markdown — the grader treats it as freeform text. Maximum 262144 characters.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].max_iterations,body,number or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Eval→revision cycles before giving up. Default 3, max 20.",https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""user.tool_result""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].type,body,"""system.message""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,events[].content,body,array of BetaManagedAgentsSystemContentBlock,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,System content blocks to append. Text-only.,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,POST /v1/sessions/{session_id}/events,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/events/send
anthropic,GET /v1/sessions/{session_id}/events/stream,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/events/stream
anthropic,GET /v1/sessions/{session_id}/events/stream,event_deltas,query,array of BetaManagedAgentsDeltaType,false,,,,agent.message|agent.thinking,,managed-agents-2026-04-01,DOCUMENTED|BETA,"When set, this connection also receives streaming deltas (`event_start`, `event_delta`) while an event is being produced, before the event itself arrives. Deltas are best-effort; when the final event is produced it carries the complete content. A model request that ends early (an error or interrupt) produces no final event — its terminal `span.model_request_end` closes the preview. Accepts one or more event types to preview and may be repeated: `agent.message` streams `content_delta` fragments; `agent.thinking` is start-only — a signal that the agent has begun extended thinking, concluded by the `agent.thinking` event itself. Only previews of the requested event types are sent.",https://platform.claude.com/docs/en/api/beta/sessions/events/stream
anthropic,GET /v1/sessions/{session_id}/events/stream,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/events/stream
anthropic,GET /v1/sessions/{session_id}/resources,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/list
anthropic,GET /v1/sessions/{session_id}/resources,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum number of resources to return per page (max 1000). If omitted, returns all resources.",https://platform.claude.com/docs/en/api/beta/sessions/resources/list
anthropic,GET /v1/sessions/{session_id}/resources,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque cursor from a previous response's `next_page` field.,https://platform.claude.com/docs/en/api/beta/sessions/resources/list
anthropic,GET /v1/sessions/{session_id}/resources,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/resources/list
anthropic,POST /v1/sessions/{session_id}/resources,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/add
anthropic,POST /v1/sessions/{session_id}/resources,file_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,ID of a previously uploaded file.,https://platform.claude.com/docs/en/api/beta/sessions/resources/add
anthropic,POST /v1/sessions/{session_id}/resources,type,body,"""file""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/add
anthropic,POST /v1/sessions/{session_id}/resources,mount_path,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Mount path in the container. Defaults to `/mnt/session/uploads/<file_id>`.,https://platform.claude.com/docs/en/api/beta/sessions/resources/add
anthropic,POST /v1/sessions/{session_id}/resources,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/resources/add
anthropic,DELETE /v1/sessions/{session_id}/resources/{resource_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/delete
anthropic,DELETE /v1/sessions/{session_id}/resources/{resource_id},resource_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/delete
anthropic,DELETE /v1/sessions/{session_id}/resources/{resource_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/resources/delete
anthropic,GET /v1/sessions/{session_id}/resources/{resource_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/retrieve
anthropic,GET /v1/sessions/{session_id}/resources/{resource_id},resource_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/retrieve
anthropic,GET /v1/sessions/{session_id}/resources/{resource_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/resources/retrieve
anthropic,POST /v1/sessions/{session_id}/resources/{resource_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/update
anthropic,POST /v1/sessions/{session_id}/resources/{resource_id},resource_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/resources/update
anthropic,POST /v1/sessions/{session_id}/resources/{resource_id},authorization_token,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,New authorization token for the resource. Currently only `github_repository` resources support token rotation.,https://platform.claude.com/docs/en/api/beta/sessions/resources/update
anthropic,POST /v1/sessions/{session_id}/resources/{resource_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/resources/update
anthropic,GET /v1/sessions/{session_id}/threads,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/list
anthropic,GET /v1/sessions/{session_id}/threads,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Maximum results per page. Defaults to 1000.,https://platform.claude.com/docs/en/api/beta/sessions/threads/list
anthropic,GET /v1/sessions/{session_id}/threads,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination cursor from a previous response's `next_page`. Forward-only.,https://platform.claude.com/docs/en/api/beta/sessions/threads/list
anthropic,GET /v1/sessions/{session_id}/threads,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/threads/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id},session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/retrieve
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id},thread_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/retrieve
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/threads/retrieve
anthropic,POST /v1/sessions/{session_id}/threads/{thread_id}/archive,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/archive
anthropic,POST /v1/sessions/{session_id}/threads/{thread_id}/archive,thread_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/archive
anthropic,POST /v1/sessions/{session_id}/threads/{thread_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/threads/archive
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/events,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/events,thread_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/events,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Query parameter for limit,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/events,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Query parameter for page,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/events,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/list
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/stream,session_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/stream
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/stream,thread_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/stream
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/stream,event_deltas,query,array of BetaManagedAgentsDeltaType,false,,,,agent.message|agent.thinking,,managed-agents-2026-04-01,DOCUMENTED|BETA,"When set, this connection also receives streaming deltas (`event_start`, `event_delta`) while an event is being produced, before the event itself arrives. Deltas are best-effort; when the final event is produced it carries the complete content. A model request that ends early (an error or interrupt) produces no final event — its terminal `span.model_request_end` closes the preview. Accepts one or more event types to preview and may be repeated: `agent.message` streams `content_delta` fragments; `agent.thinking` is start-only — a signal that the agent has begun extended thinking, concluded by the `agent.thinking` event itself. Only previews of the requested event types are sent.",https://platform.claude.com/docs/en/api/beta/sessions/threads/events/stream
anthropic,GET /v1/sessions/{session_id}/threads/{thread_id}/stream,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/sessions/threads/events/stream
anthropic,GET /v1/tunnels,include_archived,query,boolean,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,Whether to include archived tunnels in the results. Defaults to false.,https://platform.claude.com/docs/en/api/beta/tunnels/list
anthropic,GET /v1/tunnels,limit,query,number,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,"Maximum number of tunnels to return per page. Defaults to 20, maximum 1000.",https://platform.claude.com/docs/en/api/beta/tunnels/list
anthropic,GET /v1/tunnels,page,query,string,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,Opaque pagination cursor from a previous `list_tunnels` response.,https://platform.claude.com/docs/en/api/beta/tunnels/list
anthropic,GET /v1/tunnels,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/list
anthropic,POST /v1/tunnels,display_name,body,string or null,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,Optional human-readable name for the tunnel (1-255 characters).,https://platform.claude.com/docs/en/api/beta/tunnels/create
anthropic,POST /v1/tunnels,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/create
anthropic,GET /v1/tunnels/{tunnel_id},tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/retrieve
anthropic,GET /v1/tunnels/{tunnel_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/retrieve
anthropic,POST /v1/tunnels/{tunnel_id}/archive,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/archive
anthropic,POST /v1/tunnels/{tunnel_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/archive
anthropic,GET /v1/tunnels/{tunnel_id}/certificates,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/list
anthropic,GET /v1/tunnels/{tunnel_id}/certificates,include_archived,query,boolean,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,Whether to include archived certificates in the results. Defaults to false.,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/list
anthropic,GET /v1/tunnels/{tunnel_id}/certificates,limit,query,number,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,"Maximum number of certificates to return per page. Defaults to 20, maximum 1000.",https://platform.claude.com/docs/en/api/beta/tunnels/certificates/list
anthropic,GET /v1/tunnels/{tunnel_id}/certificates,page,query,string,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,Opaque pagination cursor from a previous `list_tunnel_certificates` response.,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/list
anthropic,GET /v1/tunnels/{tunnel_id}/certificates,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/list
anthropic,POST /v1/tunnels/{tunnel_id}/certificates,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/create
anthropic,POST /v1/tunnels/{tunnel_id}/certificates,ca_certificate_pem,body,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,PEM-encoded X.509 CA certificate. Must contain exactly one certificate and no private-key material. Maximum 8KB.,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/create
anthropic,POST /v1/tunnels/{tunnel_id}/certificates,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/create
anthropic,GET /v1/tunnels/{tunnel_id}/certificates/{certificate_id},tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/retrieve
anthropic,GET /v1/tunnels/{tunnel_id}/certificates/{certificate_id},certificate_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/retrieve
anthropic,GET /v1/tunnels/{tunnel_id}/certificates/{certificate_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/retrieve
anthropic,POST /v1/tunnels/{tunnel_id}/certificates/{certificate_id}/archive,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/archive
anthropic,POST /v1/tunnels/{tunnel_id}/certificates/{certificate_id}/archive,certificate_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/archive
anthropic,POST /v1/tunnels/{tunnel_id}/certificates/{certificate_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/certificates/archive
anthropic,POST /v1/tunnels/{tunnel_id}/reveal_token,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/reveal_token
anthropic,POST /v1/tunnels/{tunnel_id}/reveal_token,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/reveal_token
anthropic,POST /v1/tunnels/{tunnel_id}/rotate_token,tunnel_id,path,string,true,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/tunnels/rotate_token
anthropic,POST /v1/tunnels/{tunnel_id}/rotate_token,reason,body,string or null,false,,,,,,mcp-tunnels-2026-06-22,DOCUMENTED|BETA,"Optional free-text reason for the rotation, recorded for audit.",https://platform.claude.com/docs/en/api/beta/tunnels/rotate_token
anthropic,POST /v1/tunnels/{tunnel_id}/rotate_token,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/tunnels/rotate_token
anthropic,GET /v1/user_profiles,limit,query,number,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,Query parameter for limit,https://platform.claude.com/docs/en/api/beta/user_profiles/list
anthropic,GET /v1/user_profiles,order,query,"""asc"" or ""desc""",false,,,,asc|desc,,user-profiles-2026-08-18,DOCUMENTED|BETA,Query parameter for order,https://platform.claude.com/docs/en/api/beta/user_profiles/list
anthropic,GET /v1/user_profiles,order_by,query,"""created_at"" or ""name""",false,,,,created_at|name,,user-profiles-2026-08-18,DOCUMENTED|BETA,Query parameter for order_by,https://platform.claude.com/docs/en/api/beta/user_profiles/list
anthropic,GET /v1/user_profiles,page,query,string,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,Query parameter for page,https://platform.claude.com/docs/en/api/beta/user_profiles/list
anthropic,GET /v1/user_profiles,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/user_profiles/list
anthropic,POST /v1/user_profiles,access_type,body,"""application"" or ""passthrough""",false,,,,application|passthrough,,user-profiles-2026-08-18,DOCUMENTED|BETA,"How the platform uses the API on behalf of the entity this profile represents. `application`: the platform sells a product that uses the API behind the scenes, and the profile represents an individual end-user of that product. `passthrough`: the platform resells raw inference, and the profile identifies the resold-to company.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_id,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,Platform's own identifier for this user. Not enforced unique. Maximum 255 characters. Accepted under the `user-profiles-2026-03-24` and `user-profiles-2026-08-18` beta headers; under `user-profiles-2026-09-04` send `external_user_details.reference_id` instead.,https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details,body,BetaUserProfileExternalUserDetailsParams,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"Details about the entity this profile represents, as the platform states them. Every field is optional. Accepted under the `user-profiles-2026-09-04` beta header only.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.account_status,body,"""active"" or ""suspended"" or ""blocked"" or null",false,,,,active|suspended|blocked,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The status of the entity's account on the platform, as the platform states it: `active`; `suspended`, when the platform has restricted the account and may restore it; or `blocked`, when the platform has barred it. It records the platform's decision only; the statuses in `trust_grants` are Anthropic's and do not follow it.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.country,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The country of the entity (not of the platform), as the platform determines it: an ISO 3166-1 alpha-2 code in upper case, for example `US`. Only the form, two uppercase ASCII letters, is checked.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.email_hash,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"A hash of the entity's email address, computed by the platform. Anthropic treats it as an opaque string and does not prescribe the hash function. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.entity_type,body,"""individual"" or ""business"" or ""non_profit"" or ""government"" or null",false,,,,individual|business|non_profit|government,,user-profiles-2026-08-18,DOCUMENTED|BETA,"What kind of entity the profile represents, as the platform states it: `individual`, `business`, `non_profit` or `government`.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.name_hash,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"A hash of the entity's name, computed by the platform. Anthropic treats it as an opaque string and does not prescribe the hash function. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.onboarded_at,body,string,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_details.reference_id,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The platform's own reference for the entity, for example the key of the end-user's row in the platform's database. Not interpreted by Anthropic and not enforced unique. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,external_user_onboarded_at,body,string,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,metadata,body,map[string],false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"Free-form key-value data to attach to this user profile. Maximum 16 keys, with keys up to 64 characters and values up to 512 characters. Values must be non-empty strings.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,name,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"Optional for all profiles. Real-world name of the entity this profile represents (company or individual); for a company the platform resells Claude access to (`access_type` `passthrough`), that company's name where known. Maximum 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,POST /v1/user_profiles,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/user_profiles/create
anthropic,GET /v1/user_profiles/{user_profile_id},user_profile_id,path,string,true,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/user_profiles/retrieve
anthropic,GET /v1/user_profiles/{user_profile_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/user_profiles/retrieve
anthropic,POST /v1/user_profiles/{user_profile_id},user_profile_id,path,string,true,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},access_type,body,"""application"" or ""passthrough"" or null",false,,,,application|passthrough,,user-profiles-2026-08-18,DOCUMENTED|BETA,"How the platform uses the API on behalf of the entity this profile represents. `application`: the platform sells a product that uses the API behind the scenes, and the profile represents an individual end-user of that product. `passthrough`: the platform resells raw inference, and the profile identifies the resold-to company.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_id,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"If present, replaces the stored external_id. Omit to leave unchanged. Maximum 255 characters. Accepted under the `user-profiles-2026-03-24` and `user-profiles-2026-08-18` beta headers; under `user-profiles-2026-09-04` send `external_user_details.reference_id` instead.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details,body,BetaUserProfileExternalUserDetailsParams,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"Details about the entity this profile represents, as the platform states them. Each field sent replaces the stored value; omit a field to leave it unchanged. Once set, a value cannot be cleared and `null` is rejected. Accepted under the `user-profiles-2026-09-04` beta header only.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.account_status,body,"""active"" or ""suspended"" or ""blocked"" or null",false,,,,active|suspended|blocked,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The status of the entity's account on the platform, as the platform states it: `active`; `suspended`, when the platform has restricted the account and may restore it; or `blocked`, when the platform has barred it. It records the platform's decision only; the statuses in `trust_grants` are Anthropic's and do not follow it.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.country,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The country of the entity (not of the platform), as the platform determines it: an ISO 3166-1 alpha-2 code in upper case, for example `US`. Only the form, two uppercase ASCII letters, is checked.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.email_hash,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"A hash of the entity's email address, computed by the platform. Anthropic treats it as an opaque string and does not prescribe the hash function. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.entity_type,body,"""individual"" or ""business"" or ""non_profit"" or ""government"" or null",false,,,,individual|business|non_profit|government,,user-profiles-2026-08-18,DOCUMENTED|BETA,"What kind of entity the profile represents, as the platform states it: `individual`, `business`, `non_profit` or `government`.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.name_hash,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"A hash of the entity's name, computed by the platform. Anthropic treats it as an opaque string and does not prescribe the hash function. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.onboarded_at,body,string,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_details.reference_id,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"The platform's own reference for the entity, for example the key of the end-user's row in the platform's database. Not interpreted by Anthropic and not enforced unique. 1 to 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},external_user_onboarded_at,body,string,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},metadata,body,map[string],false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"Key-value pairs to merge into the stored metadata. Keys provided overwrite existing values. To remove a key, set its value to an empty string. Keys not provided are left unchanged. Maximum 16 keys, with keys up to 64 characters and values up to 512 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},name,body,string or null,false,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,"If present, replaces the stored name. Omit to leave unchanged. Maximum 255 characters.",https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/user_profiles/update
anthropic,POST /v1/user_profiles/{user_profile_id}/enrollment_url,user_profile_id,path,string,true,,,,,,user-profiles-2026-08-18,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/user_profiles/create_enrollment_url
anthropic,POST /v1/user_profiles/{user_profile_id}/enrollment_url,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/user_profiles/create_enrollment_url
anthropic,GET /v1/vaults,include_archived,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether to include archived vaults in the results.,https://platform.claude.com/docs/en/api/beta/vaults/list
anthropic,GET /v1/vaults,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum number of vaults to return per page. Defaults to 20, maximum 100.",https://platform.claude.com/docs/en/api/beta/vaults/list
anthropic,GET /v1/vaults,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination token from a previous `list_vaults` response.,https://platform.claude.com/docs/en/api/beta/vaults/list
anthropic,GET /v1/vaults,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/list
anthropic,POST /v1/vaults,display_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name for the vault. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/vaults/create
anthropic,POST /v1/vaults,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Arbitrary key-value metadata to attach to the vault. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/vaults/create
anthropic,POST /v1/vaults,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/create
anthropic,DELETE /v1/vaults/{vault_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/delete
anthropic,DELETE /v1/vaults/{vault_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/delete
anthropic,GET /v1/vaults/{vault_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/retrieve
anthropic,GET /v1/vaults/{vault_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/retrieve
anthropic,POST /v1/vaults/{vault_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/update
anthropic,POST /v1/vaults/{vault_id},display_name,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated human-readable name for the vault. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/vaults/update
anthropic,POST /v1/vaults/{vault_id},metadata,body,map[string] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omitted keys are preserved.",https://platform.claude.com/docs/en/api/beta/vaults/update
anthropic,POST /v1/vaults/{vault_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/update
anthropic,POST /v1/vaults/{vault_id}/archive,vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/archive
anthropic,POST /v1/vaults/{vault_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/archive
anthropic,GET /v1/vaults/{vault_id}/credentials,vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/list
anthropic,GET /v1/vaults/{vault_id}/credentials,include_archived,query,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Whether to include archived credentials in the results.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/list
anthropic,GET /v1/vaults/{vault_id}/credentials,limit,query,number,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Maximum number of credentials to return per page. Defaults to 20, maximum 100.",https://platform.claude.com/docs/en/api/beta/vaults/credentials/list
anthropic,GET /v1/vaults/{vault_id}/credentials,page,query,string,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Opaque pagination token from a previous `list_credentials` response.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/list
anthropic,GET /v1/vaults/{vault_id}/credentials,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/list
anthropic,POST /v1/vaults/{vault_id}/credentials,vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth,body,BetaManagedAgentsMCPOAuthCreateParams or BetaManagedAgentsStaticBearerCreateParams or BetaManagedAgentsEnvironmentVariableCreateParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Authentication details for creating a credential.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.type,body,"""mcp_oauth""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.access_token,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth access token.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.mcp_server_url,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,URL of the MCP server this credential authenticates against.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.expires_at,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh,body,BetaManagedAgentsMCPOAuthRefreshParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth refresh token parameters for creating a credential with refresh support.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.client_id,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth client ID.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.refresh_token,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth refresh token.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Token endpoint URL used to refresh the access token.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth,body,BetaManagedAgentsTokenEndpointAuthNoneParam or BetaManagedAgentsTokenEndpointAuthBasicParam or BetaManagedAgentsTokenEndpointAuthPostParam,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.type,body,"""none""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.type,body,"""client_secret_basic""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.client_secret,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth client secret.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.type,body,"""client_secret_post""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.resource,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth resource indicator.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.refresh.scope,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,OAuth scope for the refresh request.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.type,body,"""static_bearer""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.token,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Static bearer token value.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.type,body,"""environment_variable""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.networking,body,BetaManagedAgentsCredentialNetworkingParams,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Outbound hosts the secret value is substituted on.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.networking.type,body,"""unrestricted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.networking.type,body,"""limited""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.networking.allowed_hosts,body,array of string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Hostnames on which the secret will be substituted. Each entry is a bare hostname (`api.example.com`), an IPv4 address (`192.0.2.1`), or a `*.`-prefixed wildcard (`*.example.com`). URLs, ports, paths, and IPv6 addresses are not accepted. At most 16 entries.",https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.secret_name,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Name of the environment variable. Immutable after create.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.secret_value,body,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Secret value. Write-only; never returned in responses.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.injection_location,body,BetaManagedAgentsInjectionLocationParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Where in the outbound request the secret value may be substituted.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.injection_location.body,body,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Substitute when the placeholder appears in the request body.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,auth.injection_location.header,body,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Substitute when the placeholder appears in a request header value.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,display_name,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Human-readable name for the credential. Up to 255 characters.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,metadata,body,map[string],false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Arbitrary key-value metadata to attach to the credential. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars.",https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,POST /v1/vaults/{vault_id}/credentials,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/create
anthropic,DELETE /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/delete
anthropic,DELETE /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/delete
anthropic,DELETE /v1/vaults/{vault_id}/credentials/{credential_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/delete
anthropic,GET /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/retrieve
anthropic,GET /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/retrieve
anthropic,GET /v1/vaults/{vault_id}/credentials/{credential_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/retrieve
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth,body,BetaManagedAgentsMCPOAuthUpdateParams or BetaManagedAgentsStaticBearerUpdateParams or BetaManagedAgentsEnvironmentVariableUpdateParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated authentication details for a credential.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.type,body,"""mcp_oauth""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.access_token,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated OAuth access token.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.expires_at,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,A timestamp in RFC 3339 format,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh,body,BetaManagedAgentsMCPOAuthRefreshUpdateParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Parameters for updating OAuth refresh token configuration.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.refresh_token,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated OAuth refresh token.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.scope,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated OAuth scope for the refresh request.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.token_endpoint_auth,body,BetaManagedAgentsTokenEndpointAuthBasicUpdateParam or BetaManagedAgentsTokenEndpointAuthPostUpdateParam,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.token_endpoint_auth.type,body,"""client_secret_basic""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.token_endpoint_auth.client_secret,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated OAuth client secret.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.token_endpoint_auth.type,body,"""client_secret_post""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.type,body,"""static_bearer""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.token,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated static bearer token value.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.type,body,"""environment_variable""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.injection_location,body,BetaManagedAgentsInjectionLocationUpdateParams,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated injection location.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.injection_location.body,body,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Substitute when the placeholder appears in the request body.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.injection_location.header,body,boolean,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Substitute when the placeholder appears in a request header value.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.networking,body,BetaManagedAgentsCredentialNetworkingParams or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated networking scope. Full replacement.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.networking.type,body,"""unrestricted""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.networking.type,body,"""limited""",true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.networking.allowed_hosts,body,array of string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Hostnames on which the secret will be substituted. Each entry is a bare hostname (`api.example.com`), an IPv4 address (`192.0.2.1`), or a `*.`-prefixed wildcard (`*.example.com`). URLs, ports, paths, and IPv6 addresses are not accepted. At most 16 entries.",https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.secret_value,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated secret value.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},display_name,body,string or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,Updated human-readable name for the credential. 1-255 characters.,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},metadata,body,map[string] or null,false,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,"Metadata patch. Set a key to a string to upsert it, or to null to delete it. Omitted keys are preserved.",https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/update
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/archive,vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/archive
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/archive,credential_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/archive
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/archive,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/archive
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/mcp_oauth_validate,vault_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/mcp_oauth_validate
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/mcp_oauth_validate,credential_id,path,string,true,,,,,,managed-agents-2026-04-01,DOCUMENTED|BETA,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/mcp_oauth_validate
anthropic,POST /v1/vaults/{vault_id}/credentials/{credential_id}/mcp_oauth_validate,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/vaults/credentials/mcp_oauth_validate
anthropic,POST /v1/messages,model,body,string,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Model id or alias (e.g. claude-haiku-4-5-20251001, claude-sonnet-5, claude-fable-5-1). Unknown id → 404 not_found_error 'model: <id>'.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages,body,array<MessageParam>,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Input messages; alternating user/assistant turns (consecutive same-role turns are merged). Max 100,000 messages. Empty array → 400 'messages: at least one message is required'.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].role,body,string,true,,,,user|assistant|system,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Turn role. `system` appears in the schema but the docs state there is no system role — use top-level `system`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content,body,string|array<ContentBlockParam>,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"String shorthand for one text block, or an array of content blocks.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].type,body,string,true,,,,text|image|document|search_result|thinking|redacted_thinking|tool_use|tool_result|server_tool_use|web_search_tool_result|web_fetch_tool_result|code_execution_tool_result|bash_code_execution_tool_result|text_editor_code_execution_tool_result|tool_search_tool_result|container_upload,all current models unless noted,,DOCUMENTED,Discriminator of the request content block.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].text,body,string,true,,1,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Text of a `text` block (minLength 1).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].citations,body,array<TextCitationParam>|null,false,,,,,all current models unless noted,,DOCUMENTED,"Citations attached to a `text` block (char_location, page_location, content_block_location, web_search_result_location, search_result_location). Usually replayed from a prior response.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].cache_control,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Cache control breakpoint at this block. `type` is always `ephemeral`; `ttl` is `5m` (default) or `1h` (1h TTL is GA; the old `extended-cache-ttl-2025-04-11` beta header is no longer required).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].cache_control.type,body,string,true,,,,ephemeral,all current models unless noted,,DOCUMENTED,Always `ephemeral`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].cache_control.ttl,body,string,false,5m,,,5m|1h,all current models unless noted,,DOCUMENTED,Time-to-live of the cache entry.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source,body,object,true,,,,,all current models unless noted,,DOCUMENTED,Source of an `image` or `document` block. image: base64 | url | file. document: base64 (PDF) | text | content | url | file.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source.type,body,string,true,,,,base64|url|file|text|content,all current models unless noted,,DOCUMENTED,Source kind.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source.media_type,body,string,false,,,,image/jpeg|image/png|image/gif|image/webp|application/pdf|text/plain,all current models unless noted,,DOCUMENTED,MIME type. image: image/jpeg|image/png|image/gif|image/webp. document base64: application/pdf. document text: text/plain.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source.data,body,string (base64 or plain text),false,,,,,all current models unless noted,,DOCUMENTED,Base64 payload (image/PDF) or plain text (document text source).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source.url,body,string,false,,,,,all current models unless noted,,DOCUMENTED,Public URL (image/PDF `url` source). Not accepted by /v1/messages/count_tokens.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source.content,body,string|array<TextBlockParam|ImageBlockParam>,false,,,,,all current models unless noted,,DOCUMENTED,Custom content document (`content` source): string or array of text/image blocks; citations then use `content_block_location`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].transformations,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Image-only. Server-side transformations applied before the model sees the image.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].title,body,string|null,false,,,,,all current models unless noted,,DOCUMENTED,Document title (shown to the model; used in citations as document_title).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].context,body,string|null,false,,,,,all current models unless noted,,DOCUMENTED,Document context string passed to the model (not cited).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].source (search_result),body,string,true,,,,,all current models unless noted,,DOCUMENTED,`search_result` block: source identifier/URL of the result.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].title (search_result),body,string,true,,,,,all current models unless noted,,DOCUMENTED,`search_result` block: title.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].content (search_result),body,array<TextBlockParam>,true,,,,,all current models unless noted,,DOCUMENTED,`search_result` block: text blocks of the result.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].thinking,body,string,true,,,,,all current models unless noted,,DOCUMENTED,"`thinking` block text, replayed exactly as returned.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].signature,body,string,true,,,,,all current models unless noted,,DOCUMENTED,"`thinking` block signature, replayed unmodified (modified blocks → 400).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].data,body,string,true,,,,,all current models unless noted,,DOCUMENTED,"`redacted_thinking` opaque encrypted payload, replayed unchanged.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].id,body,string,true,,,,,all current models unless noted,,DOCUMENTED,`tool_use`/`server_tool_use` id (toolu_… / srvtoolu_…).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].name,body,string,true,,,,,all current models unless noted,,DOCUMENTED,`tool_use` tool name; `server_tool_use` name ∈ web_search|web_fetch|code_execution|bash_code_execution|text_editor_code_execution|tool_search_tool_regex|tool_search_tool_bm25.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].input,body,object,true,,,,,all current models unless noted,,DOCUMENTED,`tool_use` input object.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].caller,body,object,false,"{""type"": ""direct""}",,,,all current models unless noted,,DOCUMENTED,"Who invoked the tool: {type: direct} (default) or {type: code_execution_20250825|code_execution_20260120, tool_id}.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].toolset_name,body,string|null,false,,,,,all current models unless noted,,DOCUMENTED,"For toolset member tools (computer/browser toolsets), the toolset family name.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].tool_use_id,body,string,true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,`tool_result`: id of the tool_use being answered. Every tool_use in the previous assistant turn needs a tool_result in the next user message (else 400).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].content (tool_result),body,string|array<text|image|search_result|document|tool_reference|browser_state>,false,,,,,all current models unless noted,,DOCUMENTED,`tool_result` content: string or array of blocks.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].is_error,body,boolean,false,false,,,,all current models unless noted,,DOCUMENTED,`tool_result`: mark the result as an error so the model treats it accordingly.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].file_id,body,string,true,,,,,all current models unless noted,,DOCUMENTED,`container_upload`: Files API id to place in the code-execution container input directory.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,messages[].content[].content (server tool results),body,object|array,false,,,,,all current models unless noted,,DOCUMENTED,Replayed server-tool result payloads (web_search_tool_result content = array<web_search_result> or error object; web_fetch_tool_result content = web_fetch block or error; code_execution/bash/text_editor/tool_search results = result or error object with `error_code`).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system,body,string|array<TextBlockParam>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,System prompt: string or text blocks (each with optional cache_control and citations).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system[].type,body,string,false,,,,text,all current models unless noted,,DOCUMENTED,Always `text`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system[].text,body,string,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,System text.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system[].cache_control,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Cache control breakpoint at this block. `type` is always `ephemeral`; `ttl` is `5m` (default) or `1h` (1h TTL is GA; the old `extended-cache-ttl-2025-04-11` beta header is no longer required).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system[].cache_control.type,body,string,true,,,,ephemeral,all current models unless noted,,DOCUMENTED,Always `ephemeral`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,system[].cache_control.ttl,body,string,false,5m,,,5m|1h,all current models unless noted,,DOCUMENTED,Time-to-live of the cache entry.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,stop_sequences,body,array<string>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Custom stop strings. When hit: stop_reason `stop_sequence`, `stop_sequence` = matched string; the matched text is excluded from `content` (live: '1 2 ' with stop '3').",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,stream,body,boolean,false,false,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Stream the response as SSE (text/event-stream).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,metadata,body,object,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Request metadata.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,metadata.user_id,body,string|null,false,,,512,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Opaque end-user id (uuid/hash, no PII), ≤512 chars; used for abuse detection.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,service_tier,body,string,false,auto,,,auto|standard_only,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"`auto` (default: Priority Tier if the org has capacity, else standard) or `standard_only`. Echoed in usage.service_tier (live: 'standard' for both).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,cache_control (top-level).cache_control,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Cache control breakpoint at this block. `type` is always `ephemeral`; `ttl` is `5m` (default) or `1h` (1h TTL is GA; the old `extended-cache-ttl-2025-04-11` beta header is no longer required).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,cache_control (top-level).cache_control.type,body,string,true,,,,ephemeral,all current models unless noted,,DOCUMENTED,Always `ephemeral`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,cache_control (top-level).cache_control.ttl,body,string,false,5m,,,5m|1h,all current models unless noted,,DOCUMENTED,Time-to-live of the cache entry.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,cache_control (top-level),body,object|null,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Top-level cache_control automatically marks the last cacheable block in the request (live: accepted).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,inference_geo,body,string|null,false,,,,,Claude 4.6+ (US-only inference priced 1.1x),,DOCUMENTED|FAILED_VERIFICATION,"Geographic inference region (e.g. `us`); defaults to the workspace `default_inference_geo`. Live on Haiku 4.5: 400 ""'claude-haiku-4-5-20251001' does not support inference_geo."" Echoed in usage.inference_geo (live 'not_available').",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container,body,string|object|null,false,,,,,all current models unless noted,,DOCUMENTED,"Code-execution container to reuse: container id string or {id, skills[]}.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container.id,body,string|null,false,,,,,all current models unless noted,,DOCUMENTED,Existing container id.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container.skills,body,array<SkillParams>|null,false,,,,,all current models unless noted,,DOCUMENTED,Skills to load (≤20).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container.skills[].type,body,string,true,,,,anthropic|custom,all current models unless noted,,DOCUMENTED,anthropic (built-in) or custom.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container.skills[].skill_id,body,string,true,,,,,all current models unless noted,,DOCUMENTED,Skill id (1..64 chars).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,container.skills[].version,body,string,false,,,,,all current models unless noted,,DOCUMENTED,Skill version or `latest`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config,body,object,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Output configuration (effort, structured output format, task budget).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config.effort,body,string|null,false,,,,low|medium|high|xhigh|max,all current models unless noted,,DOCUMENTED,Effort level (models with `capabilities.effort.supported`; Haiku 4.5 reports unsupported).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config.format,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Structured outputs (GA; the `structured-outputs-2025-11-13` beta header and the old top-level `output_format` are still accepted on the beta surface). Live: `output_format` without beta → 400 'Extra inputs are not permitted'.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config.format.schema,body,object (JSON Schema),true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Schema the response must follow.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,output_config.task_budget,body,object|null,false,,,,,all current models unless noted,task-budgets-2026-03-13,DOCUMENTED|BETA,"Beta. Token budget across contexts: {type: tokens, total ≥1024, remaining?}.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,thinking,body,object,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Thinking configuration. Deep-dive owned by the thinking fragment.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,thinking.display,body,string|null,false,summarized,,,summarized|omitted|updates,all current models unless noted,,DOCUMENTED,`summarized` (default) returns thinking text; `omitted` redacts text but keeps the signature; `updates` (beta thinking-display-updates-2026-08-18) streams only progress updates.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,thinking.block_binding,body,object|null,false,,,,,all current models unless noted,thinking-binding-controls-2026-08-01,DOCUMENTED|BETA,Beta. Controls replay of thinking blocks bound to a conversation prefix.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,thinking.block_binding.prefix_mismatch_behavior,body,string|null,false,,,,error|drop_block,all current models unless noted,thinking-binding-controls-2026-08-01,DOCUMENTED|BETA,`error` (400 on mismatch) or `drop_block` (drop the block and continue).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,tools,body,array<ToolUnion>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Tool definitions (client tools with input_schema, and server/built-in tools by versioned `type`). Full catalogue owned by the tools fragment.",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].type,body,string|null,false,,,,,all current models unless noted,,DOCUMENTED,"Omitted or `custom` for client tools; otherwise a versioned server/built-in tool type (bash_20250124, text_editor_20250124/20250429/20250728, web_search_20250305/20260209/20260318, web_fetch_20250910/20260209/20260309/20260318, code_execution_20250522/20250825/20260120/20260521, memory_20250818, tool_search_tool_regex_20251119, tool_search_tool_bm25_20251119, computer_toolset_20260801, browser_toolset_20260801, …).",https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].name,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Custom: ^[a-zA-Z0-9_-]{1,128}$. Fixed for Anthropic tools: bash, str_replace_based_edit_tool (str_replace_editor for 20250124), memory, computer, web_search, web_fetch, code_execution, tool_search_tool_regex, tool_search_tool_bm25, advisor. Absent on computer_toolset/browser_toolset/mcp_toolset entries.",https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tools[].description,body,string,false,,,,,,,DOCUMENTED,Plaintext description; the most important lever for tool selection.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools
anthropic,POST /v1/messages,tools[].input_schema,body,object (JSON Schema 2020-12),true,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Client tools: schema of `input`. `type` must be `object`.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].cache_control,body,object|null,false,,,,,all current models unless noted,,DOCUMENTED,Cache control breakpoint at this block. `type` is always `ephemeral`; `ttl` is `5m` (default) or `1h` (1h TTL is GA; the old `extended-cache-ttl-2025-04-11` beta header is no longer required).,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].defer_loading,body,boolean,false,false,,,,,,DOCUMENTED|LIVE_VERIFIED,"Exclude from initial system prompt; loaded on tool_reference from tool search. Max 10,000 deferred tools. At least one tool must be non-deferred (400).",https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool
anthropic,POST /v1/messages,tools[].allowed_callers,body,array<string>,false,,,,direct|code_execution_20250825|code_execution_20260120|code_execution_20260521,all current models unless noted,,DOCUMENTED,Who may call the tool.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].eager_input_streaming,body,boolean|null,false,,,,,all current models unless noted,,DOCUMENTED,Fine-grained tool streaming for this tool (no server-side buffering/validation of `input_json_delta`). Replaces the legacy `fine-grained-tool-streaming-2025-05-14` beta header.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tools[].input_examples,body,array<object>,false,,,,,all current models unless noted,,DOCUMENTED,Example inputs shown to the model.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,tool_choice.type,body,string,false,,,,auto|any|tool|none,,,DOCUMENTED|LIVE_VERIFIED,auto | any | tool | none,https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tool_choice.name,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Required when type=tool. Cannot name a tool whose allowed_callers omits direct (400) nor a toolset/member.,https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tool_choice.disable_parallel_tool_use,body,boolean,false,false,,,,,,DOCUMENTED|LIVE_VERIFIED,auto: at most one tool_use; any/tool: exactly one. Unsupported with programmatic tool calling. Changing it invalidates the messages cache.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/parallel-tool-use
anthropic,POST /v1/messages,context_management,body,object|null,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,Beta. Server-side context edits applied before sampling. Response echoes `context_management.applied_edits[]` (live: []).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits,body,array<edit>,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA|LIVE_VERIFIED,Ordered edits (minItems 0).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].keep,body,object|string,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA,"clear_tool_uses: {type: tool_uses, value}; clear_thinking: {type: thinking_turns, value} | {type: all} | 'all'.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].clear_at_least,body,object|null,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA,"clear_tool_uses: {type: input_tokens, value} minimum to clear.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].clear_tool_inputs,body,boolean|array<string>|null,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA,clear_tool_uses: also clear tool inputs (all or named tools).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].exclude_tools,body,array<string>|null,false,,,,,all current models unless noted,context-management-2025-06-27,DOCUMENTED|BETA,clear_tool_uses: tools whose uses are never cleared.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,context_management.edits[].instructions,body,string|null,false,,,,,all current models unless noted,compact-2026-01-12,DOCUMENTED|BETA,compact_20260112: custom summarization instructions.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,compaction,body,object|null,false,,,,,all current models unless noted,compact-2026-09-04,DOCUMENTED|BETA,Beta. Compact the whole conversation and return a single signed `compaction` block; nothing is sampled after it.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,compaction.instructions,body,string|null,false,,,16384,,all current models unless noted,compact-2026-09-04,DOCUMENTED|BETA,Replacement summarization prompt (≤16384 chars; blank = absent).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers,body,array<object>,false,,,,,all current models unless noted,mcp-client-2025-11-20 (or mcp-client-2025-04-04),DOCUMENTED|BETA,Beta MCP connector: remote MCP servers Claude may call (≤20). Shape only; details owned by the tools fragment.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers[].type,body,string,true,,,,url,all current models unless noted,mcp-client-2025-11-20,DOCUMENTED|BETA,Always `url`.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers[].name,body,string,true,,,,,all current models unless noted,mcp-client-2025-11-20,DOCUMENTED|BETA,Server name (appears in mcp_tool_use.server_name).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers[].url,body,string,true,,,,,all current models unless noted,mcp-client-2025-11-20,DOCUMENTED|BETA,Server URL.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers[].authorization_token,body,string|null,false,,,,,all current models unless noted,mcp-client-2025-11-20,DOCUMENTED|BETA,OAuth bearer token forwarded to the server.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,mcp_servers[].tool_configuration,body,object|null,false,,,,,all current models unless noted,mcp-client-2025-11-20,DOCUMENTED|BETA,"{enabled?: boolean, allowed_tools?: string[]}.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,speed,body,string|null,false,,,,standard|fast,all current models unless noted,fast-mode-2026-02-01,DOCUMENTED|BETA|FAILED_VERIFICATION,"Beta fast mode: `fast` = much faster output at premium pricing on supported models. Live on Haiku 4.5: 400 ""does not support the `speed` parameter"". Echoed as usage.speed.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,diagnostics,body,object|null,false,,,,,all current models unless noted,cache-diagnosis-2026-04-07,DOCUMENTED|BETA,Beta. Prompt-cache divergence reporting; response gets `diagnostics` with cache-miss reasons.,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,diagnostics.previous_message_id,body,string|null,false,,,256,,all current models unless noted,cache-diagnosis-2026-04-07,DOCUMENTED|BETA,Id of the previous message to compare the cache prefix against (≤256).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,fallbacks,body,array<object>|'default',false,,,,,all current models unless noted,server-side-fallback-2026-07-01 (or -2026-06-01),DOCUMENTED|BETA,"Beta server-side fallback: ordered fallback models tried when the requested model refuses. Each entry: model, max_tokens?, output_config?, speed?, thinking?.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,fallback_credit_token,body,string|object|null,false,,,,,all current models unless noted,fallback-credit-2026-07-01 (or -2026-06-01),DOCUMENTED|BETA,"Beta. Token from a refused response's usage.fallback_credit; redeem on the client-side retry within 5 minutes so warm-prefix cache-creation tokens are billed at cache-read rate. Object form: {token, mode: strict|best_effort}.",https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,output_format,body,object|null,false,,,,,all current models unless noted,structured-outputs-2025-11-13,DOCUMENTED|BETA|DEPRECATED|LIVE_VERIFIED,Legacy beta field for structured outputs (moved to output_config.format). Only accepted on the beta surface; without the header → 400 'output_format: Extra inputs are not permitted' (live).,https://platform.claude.com/docs/en/api/beta/messages
anthropic,POST /v1/messages,betas,body,array<string>,false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,SDK-only convenience parameter (client.beta.messages.*): the SDK turns it into the `anthropic-beta` header. It is NOT an API body field — live: 400 'betas: Extra inputs are not permitted'.,https://platform.claude.com/docs/en/api/beta-headers
anthropic,POST /v1/messages,anthropic-beta,header,"string (comma-separated, repeatable)",false,,,,,all current models unless noted,,DOCUMENTED|LIVE_VERIFIED,Opt into beta features; multiple values comma-separated or repeated header. Unknown value → 400 'Unexpected value(s) … for the `anthropic-beta` header'.,https://platform.claude.com/docs/en/api/beta-headers
anthropic,POST /v1/messages,anthropic-workspace-id,header,string,false,,,,,all current models unless noted,,DOCUMENTED,Workspace to run in (wrkspc_…); required for multi-workspace keys.,https://platform.claude.com/docs/en/api/messages
anthropic,POST /v1/messages,anthropic-user-profile-id,header,string,false,,,,,all current models unless noted,user-profiles-2026-09-04,DOCUMENTED|BETA,Attribute the request to a user profile; requires a `user-profiles-*` beta header.,https://platform.claude.com/docs/en/api/messages
anthropic,GET /v1/skills,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of results to return per page. Ranges from `1` to `1000`. Defaults to `20`.,https://platform.claude.com/docs/en/api/beta/skills/list
anthropic,GET /v1/skills,page,query,string,false,,,,,,,DOCUMENTED,Pagination token for fetching a specific page of results. Pass the value from a previous response's `next_page` field to get the next page of results.,https://platform.claude.com/docs/en/api/beta/skills/list
anthropic,GET /v1/skills,source,query,string,false,,,,,,,DOCUMENTED,"Filter skills by source. If provided, only skills from the specified source will be returned: * `""custom""`: only return user-created skills * `""anthropic""`: only return Anthropic-created skills",https://platform.claude.com/docs/en/api/beta/skills/list
anthropic,GET /v1/skills,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/list
anthropic,POST /v1/skills,files,body,array of string,true,,,,,,,DOCUMENTED,Files to upload for the skill. All files must be in the same top-level directory and must include a SKILL.md file at the root of that directory.,https://platform.claude.com/docs/en/api/beta/skills/create
anthropic,POST /v1/skills,display_name,body,string or null,false,,,,,,,DOCUMENTED,"Human-readable, single-line label for the Skill. Maximum 255 characters. Always set: derived from the SKILL.md frontmatter `name` when omitted at creation. Not unique.",https://platform.claude.com/docs/en/api/beta/skills/create
anthropic,POST /v1/skills,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/create
anthropic,DELETE /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/delete
anthropic,DELETE /v1/skills/{skill_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/delete
anthropic,GET /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/retrieve
anthropic,GET /v1/skills/{skill_id},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/retrieve
anthropic,GET /v1/skills/{skill_id}/versions,skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/versions/list
anthropic,GET /v1/skills/{skill_id}/versions,limit,query,number,false,20,1,1000,,,,DOCUMENTED,Number of results to return per page. Ranges from `1` to `1000`. Defaults to `20`.,https://platform.claude.com/docs/en/api/beta/skills/versions/list
anthropic,GET /v1/skills/{skill_id}/versions,page,query,string,false,,,,,,,DOCUMENTED,Optionally set to the `next_page` token from the previous response.,https://platform.claude.com/docs/en/api/beta/skills/versions/list
anthropic,GET /v1/skills/{skill_id}/versions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/versions/list
anthropic,POST /v1/skills/{skill_id}/versions,skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/versions/create
anthropic,POST /v1/skills/{skill_id}/versions,files,body,array of string,true,,,,,,,DOCUMENTED,Files to upload for the skill. All files must be in the same top-level directory and must include a SKILL.md file at the root of that directory.,https://platform.claude.com/docs/en/api/beta/skills/versions/create
anthropic,POST /v1/skills/{skill_id}/versions,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/versions/create
anthropic,DELETE /v1/skills/{skill_id}/versions/{version},skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/versions/delete
anthropic,DELETE /v1/skills/{skill_id}/versions/{version},version,path,string,true,,,,,,,DOCUMENTED,"Identifies the skill version by its version ID. Requests carrying the `skills-2025-10-02` beta header address versions by their Unix epoch timestamp instead (e.g., ""1759178010641129"").",https://platform.claude.com/docs/en/api/beta/skills/versions/delete
anthropic,DELETE /v1/skills/{skill_id}/versions/{version},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/versions/delete
anthropic,GET /v1/skills/{skill_id}/versions/{version},skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/versions/retrieve
anthropic,GET /v1/skills/{skill_id}/versions/{version},version,path,string,true,,,,,,,DOCUMENTED,"Identifies the skill version: a version ID, or the literal `latest` for the skill's most recent version. Requests carrying the `skills-2025-10-02` beta header address versions by their Unix epoch timestamp instead (e.g., ""1759178010641129"").",https://platform.claude.com/docs/en/api/beta/skills/versions/retrieve
anthropic,GET /v1/skills/{skill_id}/versions/{version},anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/versions/retrieve
anthropic,GET /v1/skills/{skill_id}/versions/{version}/content,skill_id,path,string,true,,,,,,,DOCUMENTED,Unique identifier for the skill. The format and length of IDs may change over time.,https://platform.claude.com/docs/en/api/beta/skills/versions/download
anthropic,GET /v1/skills/{skill_id}/versions/{version}/content,version,path,string,true,,,,,,,DOCUMENTED,"Identifies the skill version by its version ID. Requests carrying the `skills-2025-10-02` beta header address versions by their Unix epoch timestamp instead (e.g., ""1759178010641129"").",https://platform.claude.com/docs/en/api/beta/skills/versions/download
anthropic,GET /v1/skills/{skill_id}/versions/{version}/content,anthropic-workspace-id,header,string,false,,,,,,,DOCUMENTED,,https://platform.claude.com/docs/en/api/beta/skills/versions/download
anthropic,POST /v1/messages,tools,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Tool definitions (custom + Anthropic-provided). Union of custom tool objects and dated `type` objects. Tools are rendered into a system prompt whose token cost depends on model and tool_choice.,https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tools[].type,body,string,false,,,,custom|bash_20250124|text_editor_20250124|text_editor_20250429|text_editor_20250728|memory_20250818|computer_toolset_20260801|browser_toolset_20260801|computer_20251124|computer_20250124|web_search_20250305|web_search_20260209|web_search_20260318|web_fetch_20250910|web_fetch_20260209|web_fetch_20260309|web_fetch_20260318|code_execution_20250522|code_execution_20250825|code_execution_20260120|code_execution_20260521|tool_search_tool_regex_20251119|tool_search_tool_regex|tool_search_tool_bm25_20251119|tool_search_tool_bm25|advisor_20260301|mcp_toolset,,,DOCUMENTED|LIVE_VERIFIED,Omitted or 'custom' for user-defined tools; otherwise an exact versioned type string (see tools fragment notes).,https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tools[].input_schema,body,object,true,,,,,,,DOCUMENTED,JSON Schema (type object) for custom tools. Strict mode: structured-outputs subset.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools
anthropic,POST /v1/messages,tools[].input_examples,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Schema-validated example inputs (~20-200 tokens each). Not on server tools or toolsets. Invalid example -> 400.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/define-tools
anthropic,POST /v1/messages,tools[].cache_control,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"{type: ephemeral, ttl?: 5m|1h}; put on the last tool to cache the whole tools prefix. Not with defer_loading:true (400).",https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-use-with-prompt-caching
anthropic,POST /v1/messages,tools[].allowed_callers,body,array,false,,,,direct|code_execution_20250825|code_execution_20260120|code_execution_20260521,,,DOCUMENTED|LIVE_VERIFIED,Any of direct | code_execution_20250825 | code_execution_20260120 | code_execution_20260521. Default [direct]; web_search/web_fetch _20260209+ default [code_execution_20260120]. Toolsets accept only [direct].,https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling
anthropic,POST /v1/messages,tools[].eager_input_streaming,body,boolean,false,false,,,,,,DOCUMENTED|LIVE_VERIFIED,Custom tools only. true = fine-grained tool streaming (no buffering/validation of input JSON). Replaces beta header fine-grained-tool-streaming-2025-05-14.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/fine-grained-tool-streaming
anthropic,POST /v1/messages,tools[].max_characters,body,integer,false,,,,,,,DOCUMENTED,text_editor_20250728+: truncation length for view.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/text-editor-tool
anthropic,POST /v1/messages,tools[].display_width_px,body,integer,true,,1,,,,computer-use-2025-11-24 | computer-use-2025-01-24,DOCUMENTED|LIVE_VERIFIED,computer_20251124 / computer_20250124 (beta): display width; required. Rejected on computer_toolset_20260801.,https://platform.claude.com/docs/en/api/beta/messages/create
anthropic,POST /v1/messages,tools[].display_height_px,body,integer,true,,1,,,,computer-use-2025-11-24 | computer-use-2025-01-24,DOCUMENTED|LIVE_VERIFIED,computer_2025*: display height; required.,https://platform.claude.com/docs/en/api/beta/messages/create
anthropic,POST /v1/messages,tools[].display_number,body,integer,false,,0,,,,computer-use-2025-11-24 | computer-use-2025-01-24,DOCUMENTED,computer_2025*: X11 display number.,https://platform.claude.com/docs/en/api/beta/messages/create
anthropic,POST /v1/messages,tools[].enable_zoom,body,boolean,false,false,,,,,computer-use-2025-11-24,DOCUMENTED,computer_20251124 only: enable the zoom action.,https://platform.claude.com/docs/en/api/beta/messages/create
anthropic,POST /v1/messages,tools[].configs,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"computer_toolset_20260801 / browser_toolset_20260801: member name -> {enabled?, defer_loading?}. mcp_toolset: tool name -> {enabled?, defer_loading?}.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-reference
anthropic,POST /v1/messages,tools[].max_uses,body,integer,false,,1,,,,,DOCUMENTED|LIVE_VERIFIED,web_search / web_fetch / advisor: per-request cap; exceeding yields a *_tool_result_error max_uses_exceeded.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool
anthropic,POST /v1/messages,tools[].allowed_domains,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,web_search / web_fetch: bare domains (+path for search). Mutually exclusive with blocked_domains (400). Must be within org allowlist.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/server-tools
anthropic,POST /v1/messages,tools[].blocked_domains,body,array,false,,,,,,,DOCUMENTED,web_search / web_fetch: domains to exclude.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/server-tools
anthropic,POST /v1/messages,tools[].user_location,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"web_search: {type: approximate, city?, region?, country? (2-letter, else 400), timezone?}.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool
anthropic,POST /v1/messages,tools[].response_inclusion,body,string,false,full,,,full|excluded,,,DOCUMENTED|LIVE_VERIFIED,web_search_20260318 / web_fetch_20260318: drop result blocks consumed by completed code execution.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool
anthropic,POST /v1/messages,tools[].citations,body,object,false,"{""enabled"": false}",,,,,,DOCUMENTED|LIVE_VERIFIED,web_fetch: {enabled: bool}; off by default (web search citations are always on).,https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool
anthropic,POST /v1/messages,tools[].max_content_tokens,body,integer,false,,1,,,,,DOCUMENTED|LIVE_VERIFIED,web_fetch: approximate cap on fetched text tokens (not PDFs). Must be > 0 (live 400).,https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool
anthropic,POST /v1/messages,tools[].url_sources,body,object,false,,,,,,,DOCUMENTED,"web_fetch: {user_input: all|none, client_tool_results: all|none|only{tools}|except{tools}, server_tool_results: …}.",https://platform.claude.com/docs/en/api/messages/create
anthropic,POST /v1/messages,tools[].use_cache,body,boolean,false,true,,,,,,DOCUMENTED|LIVE_VERIFIED,web_fetch_20260309+: false bypasses the fetch cache.,https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool
anthropic,POST /v1/messages,tools[].model,body,string,true,,,,,,advisor-tool-2026-03-01,DOCUMENTED|LIVE_VERIFIED,advisor_20260301: advisor model id (must form a valid executor/advisor pair).,https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool
anthropic,POST /v1/messages,tools[].max_tokens,body,integer,false,,1024,,,,advisor-tool-2026-03-01,DOCUMENTED,advisor_20260301: cap on advisor output per call (min 1024).,https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool
anthropic,POST /v1/messages,tools[].caching,body,object,false,,,,,,advisor-tool-2026-03-01,DOCUMENTED,"advisor_20260301: {type: ephemeral, ttl: 5m|1h} advisor-side caching switch.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/advisor-tool
anthropic,POST /v1/messages,tools[].mcp_server_name,body,string,true,,,,,,mcp-client-2025-11-20,DOCUMENTED|LIVE_VERIFIED,mcp_toolset: must match one mcp_servers[].name.,https://platform.claude.com/docs/en/agents-and-tools/mcp-connector
anthropic,POST /v1/messages,tools[].default_config,body,object,false,,,,,,mcp-client-2025-11-20,DOCUMENTED|LIVE_VERIFIED,"mcp_toolset: {enabled? (default true), defer_loading? (default false)} applied to all tools of the server.",https://platform.claude.com/docs/en/agents-and-tools/mcp-connector
anthropic,POST /v1/messages,container,body,string|object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Code execution container to reuse: id string, or {id?, skills?: [{type: anthropic|custom, skill_id, version}]} (max 20 skills). Required while a programmatic tool call is pending.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool
anthropic,POST /v1/messages,container.skills[],body,object,false,,,,,,,DOCUMENTED,"{type: anthropic|custom, skill_id (1-64 chars), version ('latest' or version id)} — loads a Skill into the code execution container (Skills API CRUD is documented separately).",https://platform.claude.com/docs/en/build-with-claude/skills-guide
anthropic,POST /v1/messages,mcp_servers,body,array,false,,,,,,mcp-client-2025-11-20,DOCUMENTED|LIVE_VERIFIED,"Remote MCP servers (max 20): [{type: url, url (https), name, authorization_token?}]. Requires beta header.",https://platform.claude.com/docs/en/agents-and-tools/mcp-connector
anthropic,POST /v1/messages,mcp_servers[].tool_configuration,body,object,false,,,,,,mcp-client-2025-04-04,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"DEPRECATED (mcp-client-2025-04-04): {enabled?, allowed_tools?[]}. Still accepted live with the old header.",https://platform.claude.com/docs/en/agents-and-tools/mcp-connector
anthropic,POST /v1/messages,messages[].content[] (tool_result),body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"{type: tool_result, tool_use_id, content?: string|[text|image|document|search_result|tool_reference], is_error?: bool, toolset_name?: computer|browser (required echo for toolset members), cache_control?}. Must come first in the user message following the tool_use; only tool_result blocks when a server/programmatic call is pending.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/handle-tool-calls
anthropic,POST /v1/messages,messages[].content[] (container_upload),body,object,false,,,,,,,DOCUMENTED,"{type: container_upload, file_id} — preload a Files API file into the code execution container.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool
anthropic,POST /v1/messages,"messages[].content[] (tool_use, replayed)",body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Assistant tool_use blocks (and server_tool_use / *_tool_result / mcp_tool_use / mcp_tool_result / tool_search_tool_result) must be replayed verbatim, including encrypted_content / encrypted_index.",https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-search-tool
anthropic,POST /v1/messages/count_tokens,tools,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Client tools and the advisor tool are counted; server tools (web_search, web_fetch, code_execution, tool_search) and mcp_servers -> 400 'Server tools are not supported in the count_tokens endpoint'.",https://platform.claude.com/docs/en/build-with-claude/token-counting
anthropic,POST /v1/messages,anthropic-beta,header,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Header. Tool-related values: computer-use-2025-11-24, computer-use-2025-01-24, computer-use-2024-10-22 (legacy), advisor-tool-2026-03-01, mcp-client-2025-11-20, mcp-client-2025-04-04 (deprecated), fine-grained-tool-streaming-2025-05-14 (legacy no-op-ish), code-execution-2025-05-22 / code-execution-2025-08-25 (legacy opt-ins), web-fetch-2025-09-10 (legacy), advanced-tool-use-2025-11-20 (legacy for tool search/PTC), context-management-2025-06-27 (memory era).",https://platform.claude.com/docs/en/api/beta-headers
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.displayName,body,string,true,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,User-defined batch name (required).,https://ai.google.dev/api/batch-api
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.inputConfig.requests.requests[],body,array<InlinedRequest>,false,,,,,,,DOCUMENTED,"Inline requests (total request < 20 MB). Each item: {request: GenerateContentRequest, metadata: {key: …}}. Output comes back inline (response.inlinedResponses).",https://ai.google.dev/gemini-api/docs/batch-api#inline-requests
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.inputConfig.requests.requests[].request,body,object GenerateContentRequest,true,,,,,,,DOCUMENTED,"Full request (contents, systemInstruction, tools, generationConfig incl. responseModalities for image generation, cachedContent…). The `model` field is taken from the URL.",https://ai.google.dev/api/batch-api#GenerateContentRequest
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.inputConfig.requests.requests[].metadata,body,object (Struct),false,,,,,,,DOCUMENTED,"Free-form metadata echoed in InlinedResponse.metadata; convention {""key"": ""request-1""}.",https://ai.google.dev/api/batch-api#InlinedRequest
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.inputConfig.fileName,body,string,false,,,,,,,DOCUMENTED,"Files API name (files/…) of a JSONL input file (<= 2 GB), one {""key"": …, ""request"": GenerateContentRequest} per line. Output = JSONL responsesFile. Mutually exclusive with requests.",https://ai.google.dev/gemini-api/docs/batch-api#input-file
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.priority,body,string (int64),false,0,,,,,,DOCUMENTED,Higher value processed first; negative allowed.,https://ai.google.dev/api/batch-api
gemini,POST /v1beta/models/{model}:batchGenerateContent,batch.webhookConfig.uris[],body,array<string>,false,,,,,,,DOCUMENTED,Per-batch webhook URIs (override registered webhooks; events batch.succeeded / batch.failed).,https://ai.google.dev/api/batch-api
gemini,POST /v1beta/models/{model}:asyncBatchEmbedContent,batch.inputConfig.requests.requests[].request,body,object EmbedContentRequest,true,,,,,gemini-embedding-2|gemini-embedding-2-preview|gemini-embedding-001,,DOCUMENTED|ACCOUNT_RESTRICTED,"{content:{parts:[{text}]}, embedContentConfig:{taskType, outputDimensionality, title}} (top-level taskType/title/outputDimensionality are deprecated).",https://ai.google.dev/api/batch-api#EmbedContentRequest
gemini,POST /v1beta/models/{model}:asyncBatchEmbedContent,batch.displayName,body,string,true,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|ACCOUNT_RESTRICTED,Required user-defined batch name.,https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:asyncBatchEmbedContent,batch.inputConfig.fileName,body,string,false,,,,,,,DOCUMENTED,JSONL file of EmbedContentRequest lines (Files API).,https://ai.google.dev/gemini-api/docs/batch-api#batch-embedding-support
gemini,GET /v1beta/batches,pageSize,query,integer,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Standard list page size.,https://ai.google.dev/api/batch-api
gemini,GET /v1beta/batches,pageToken,query,string,false,,,,,,,DOCUMENTED,Standard list page token.,https://ai.google.dev/api/batch-api
gemini,GET /v1beta/batches,filter,query,string,false,,,,,,,DOCUMENTED,Standard list filter.,https://ai.google.dev/api/batch-api
gemini,GET /v1beta/batches,returnPartialSuccess,query,boolean,false,,,,,,,DOCUMENTED,Return reachable operations and list unreachable ones; UNIMPLEMENTED unless documented.,https://ai.google.dev/api/batch-api
gemini,PATCH /v1beta/{name=batches/*}:updateGenerateContentBatch,updateMask,query,string (FieldMask),false,,,,,,,DOCUMENTED,Comma-separated fields to update (e.g. priority).,https://ai.google.dev/api/batch-api#method:-batches.updategeneratecontentbatch
gemini,PATCH /v1beta/{name=batches/*}:updateEmbedContentBatch,updateMask,query,string (FieldMask),false,,,,,,,DOCUMENTED,Comma-separated fields to update.,https://ai.google.dev/api/batch-api
gemini,GET /v1beta/{name=batches/*},name,path,string,true,,,,,,,DOCUMENTED,"batches/{batchId} returned by create. Response is an Operation: metadata.state, done, response.inlinedResponses | response.responsesFile.",https://ai.google.dev/api/batch-api
gemini,JSONL input line,key,body,string,true,,,,,,,DOCUMENTED,User key echoed in the output line.,https://ai.google.dev/gemini-api/docs/batch-api#input-file
gemini,JSONL input line,request,body,object GenerateContentRequest | EmbedContentRequest,true,,,,,,,DOCUMENTED,Request body per line (camelCase generationConfig or snake_case generation_config both appear in official samples).,https://ai.google.dev/gemini-api/docs/batch-api#input-file
gemini,POST /v1beta/cachedContents,model,body,string,true,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-flash-lite|gemini-3.1-pro-preview|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED|LIVE_VERIFIED,"Required, immutable, format `models/{id}` (else 400 ""Model name must be in the format 'models/{model_name}'""). Caches are usable only with this exact model.",https://ai.google.dev/api/caching
gemini,POST /v1beta/cachedContents,contents,body,array<Content>,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Input-only, immutable content to cache (text, inlineData, fileData — e.g. a video File). Total tokens must reach the model minimum: live min_total_token_count=1024 on gemini-3.5-flash-lite (400 'Cached content is too small. total_token_count=6, min_total_token_count=1024'); gemini-3.5-flash: 400 'Cached content is too small. total_token_count=6, min_total_token_count=1024' (gemini-3.5-flash — the size check runs before the quota check; docs table says 4,096 for implicit caching). Docs (implicit caching table): 4,096 for 3.x Flash/Pro, 2,048 for 2.5.",https://ai.google.dev/gemini-api/docs/generate-content/caching
gemini,POST /v1beta/cachedContents,systemInstruction,body,Content,false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,"Input-only, immutable system instruction stored with the cache (text only). A request that uses the cache must not send its own systemInstruction (not executed — cache creation blocked on this free-tier key (429 RESOURCE_EXHAUSTED 'TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-3.5-flash: limit=0, requested=5301'; same for gemini-3.5-flash-lite)).",https://ai.google.dev/api/caching
gemini,POST /v1beta/cachedContents,tools / toolConfig,body,array<Tool> / ToolConfig,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,"Input-only, immutable tool declarations cached with the content (shape — tools domain).",https://ai.google.dev/api/caching
gemini,POST /v1beta/cachedContents,displayName,body,string,false,,,128,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,"Optional immutable label, ≤ 128 Unicode chars (PATCH of displayName: not executed — cache creation blocked on this free-tier key (429 RESOURCE_EXHAUSTED 'TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-3.5-flash: limit=0, requested=5301'; same for gemini-3.5-flash-lite)).",https://ai.google.dev/api/caching
gemini,POST /v1beta/cachedContents,ttl,body,"string (Duration, e.g. '120s', '3600s')",false,3600s,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,Input-only time-to-live; default 1 h when neither ttl nor expireTime is given; no documented min/max. Response always returns `expireTime` instead. Storage billed per token-hour ($1.00 / 1M tokens / h on 3.5 models).,https://ai.google.dev/gemini-api/docs/generate-content/caching
gemini,POST /v1beta/cachedContents,expireTime,body,string (RFC 3339 UTC timestamp),false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,Absolute expiry; mutually exclusive with ttl (union `expiration`). Always present on output.,https://ai.google.dev/api/caching
gemini,PATCH /v1beta/cachedContents/{id},ttl | expireTime,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,"Only expiration fields are updatable. ttl '300s' → not executed — cache creation blocked on this free-tier key (429 RESOURCE_EXHAUSTED 'TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-3.5-flash: limit=0, requested=5301'; same for gemini-3.5-flash-lite); with ?updateMask=ttl → not executed — cache creation blocked on this free-tier key (429 RESOURCE_EXHAUSTED 'TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-3.5-flash: limit=0, requested=5301'; same for gemini-3.5-flash-lite).",https://ai.google.dev/api/caching
gemini,PATCH /v1beta/cachedContents/{id},updateMask,query,string (FieldMask),false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,Optional list of fields to update (e.g. `ttl`).,https://ai.google.dev/api/caching
gemini,GET /v1beta/cachedContents,pageSize,query,integer,false,,,1000,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,Max items per page; server default when unset; max 1000 (coerced).,https://ai.google.dev/api/caching
gemini,GET /v1beta/cachedContents,pageToken,query,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,nextPageToken from a previous list call.,https://ai.google.dev/api/caching
gemini,GET /v1beta/cachedContents/{id},name,path,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,"`cachedContents/{id}` (ids are opaque, e.g. cachedContents/abc123…). After delete: not executed — cache creation blocked on this free-tier key (429 RESOURCE_EXHAUSTED 'TotalCachedContentStorageTokensPerModelFreeTier limit exceeded for model gemini-3.5-flash: limit=0, requested=5301'; same for gemini-3.5-flash-lite).",https://ai.google.dev/api/caching
gemini,POST /v1beta/models/{model}:countTokens,model,path,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Model whose tokenizer is used. Different models may count differently (gemini-3.5-flash-lite and gemini-3.5-flash both gave 11 for the fox sentence).,https://ai.google.dev/api/tokens
gemini,POST /v1beta/models/{model}:countTokens,contents,body,array<Content>,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Prompt turns to count (same Content/Part shapes as generateContent incl. inlineData, fileData, YouTube URLs). Ignored when generateContentRequest is set (verified: both sent → wrapper counted).",https://ai.google.dev/api/tokens
gemini,POST /v1beta/models/{model}:countTokens,generateContentRequest,body,GenerateContentRequest,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Whole request to count — includes systemInstruction, tools, toolConfig, cachedContent, generationConfig (e.g. mediaResolution changes the count: PNG 1089 → 256 with LOW). Its `model` field is required in format models/{id} but the path model wins (mismatch accepted). Either contents or generateContentRequest must be present: empty body → 400 'CountTokens requires generate_content_request or contents to be set.'",https://ai.google.dev/api/tokens
gemini,POST /v1beta/models/{model}:countTokens,systemInstruction (top level),body,—,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"NOT accepted at top level (400 'Invalid JSON payload received. Unknown name ""systemInstruction"": Cannot find field.'); nest it inside generateContentRequest. SDK caveat: google-genai 2.24 and @google/genai 2.23 REFUSE system_instruction/tools in count_tokens for the Developer API ('system_instruction parameter is only supported in Gemini Enterprise Agent Platform mode, not in Gemini Developer API mode') — send the wrapper via raw REST.",https://ai.google.dev/api/tokens
gemini,POST /v1beta/models/{model}:countTokens,generateContentRequest.cachedContent,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,Counting with a cache reports cachedContentTokenCount + cacheTokensDetails.,https://ai.google.dev/api/tokens
gemini,POST /v1beta/models/{model}:embedContent,model,path,string,true,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,"Embedding model: gemini-embedding-001 (text, 2,048 input tokens), gemini-embedding-2 / -2-preview (multimodal, 8,192 tokens). Generation models → 404 'not supported for embedContent'.",https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:embedContent,model (body),body,string,false,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,Optional repeat of `models/{id}` in the body (required in batch requests).,https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:embedContent,content,body,Content,true,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,"Content to embed. embedding-001: text parts only. embedding-2: text, inlineData/fileData image (≤ 6 PNG/JPEG), audio (≤ 180 s MP3/WAV), video (≤ 120 s MP4/MOV, 32 frames), PDF (1 file ≤ 6 pages, 258 tokens/page + text) — all parts of ONE content are aggregated into ONE vector; use several contents (batchEmbedContents) for separate vectors.",https://ai.google.dev/gemini-api/docs/embeddings
gemini,POST /v1beta/models/{model}:embedContent,taskType,body,string,false,,,,TASK_TYPE_UNSPECIFIED|RETRIEVAL_QUERY|RETRIEVAL_DOCUMENT|SEMANTIC_SIMILARITY|CLASSIFICATION|CLUSTERING|QUESTION_ANSWERING|FACT_VERIFICATION|CODE_RETRIEVAL_QUERY,gemini-embedding-001,,DOCUMENTED|LIVE_VERIFIED,Task hint (top-level form deprecated in favour of embedContentConfig.taskType). Documented for gemini-embedding-001 only; docs say unsupported on gemini-embedding-2 (use `task: … | query: …` prompt prefixes) — live it was silently accepted (200) on embedding-2.,https://ai.google.dev/gemini-api/docs/embeddings
gemini,POST /v1beta/models/{model}:embedContent,title,body,string,false,,,,,gemini-embedding-001,,DOCUMENTED|LIVE_VERIFIED,"Document title, only with taskType RETRIEVAL_DOCUMENT (improves retrieval quality). Deprecated top-level form; also embedContentConfig.title.",https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:embedContent,outputDimensionality,body,integer,false,3072,128,3072,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,"Matryoshka truncation of the 3072-d vector; supported 128–3072 (recommended 768, 1536, 3072). embedding-2 renormalizes truncated vectors; embedding-001 does not (normalize client-side). 4096 → 400 'Unable to submit request because the outputDimensionality value is not supported.' Deprecated top-level form; also embedContentConfig.outputDimensionality.",https://ai.google.dev/gemini-api/docs/embeddings
gemini,POST /v1beta/models/{model}:embedContent,embedContentConfig,body,"EmbedContentConfig {taskType, title, outputDimensionality, autoTruncate, documentOcr, audioTrackExtraction}",false,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,Preferred nested config (SDK EmbedContentConfig). autoTruncate silently truncates over-long input; documentOcr / audioTrackExtraction toggle PDF OCR and video audio-track processing (Gemini Developer API always OCRs PDFs; docs say documentOcr is Vertex-only).,https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:batchEmbedContents,requests[],body,array<EmbedContentRequest>,true,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|LIVE_VERIFIED,Batch of embed requests; each MUST include `model: models/{id}` matching the path (missing → 400 '* BatchEmbedContentsRequest.requests[0].model: model is not specified'). Response embeddings[] keeps order.,https://ai.google.dev/api/embeddings
gemini,POST /v1beta/models/{model}:asyncBatchEmbedContent,batch.inputConfig.requests.requests[].request / .metadata,body,InlinedEmbedContentRequest[],false,,,,,gemini-embedding-001|gemini-embedding-2|gemini-embedding-2-preview,,DOCUMENTED|ACCOUNT_RESTRICTED,"Inline requests (EmbedContentRequest + free-form metadata echoed in inlinedResponses) — or `batch.inputConfig.fileName` pointing to a JSONL File. Optional `batch.priority` (int64, higher first).",https://ai.google.dev/api/embeddings
gemini,POST /v1beta/fileSearchStores,displayName,body,string,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. The human-readable display name for the `FileSearchStore`. The display name must be no more than 512 characters in length, including spaces. Example: ""Docs on Semantic Retriever""",https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /v1beta/fileSearchStores,embeddingModel,body,string,false,default text embedding model,,,,,,DOCUMENTED|PREVIEW,"Optional. The embedding model to use for the `FileSearchStore`. The model's resource name. This serves as an ID for the Model to use. Format: `models/{model}`. If not specified, the default embedding model will be used. Use models/gemini-embedding-2 for multimodal (image) File Search; default is text-only (gemini-embedding-001 per guide).",https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores,pageSize,query,integer,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. The maximum number of `FileSearchStores` to return (per page). The service may return fewer `FileSearchStores`. If unspecified, at most 10 `FileSearchStores` will be returned. The maximum size limit is 20 `FileSearchStores` per page.",https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores,pageToken,query,string,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. A page token, received from a previous `ListFileSearchStores` call. Provide the `next_page_token` returned in the response as an argument to the next request to retrieve the next page. When paginating, all other parameters provided to `ListFileSearchStores` must match the call that provided the page token.",https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores/{name},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The name of the `FileSearchStore`. Example: `fileSearchStores/my-file-search-store-123`,https://ai.google.dev/api/file-search/file-search-stores
gemini,DELETE /v1beta/fileSearchStores/{name},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The resource name of the `FileSearchStore`. Example: `fileSearchStores/my-file-search-store-123`,https://ai.google.dev/api/file-search/file-search-stores
gemini,DELETE /v1beta/fileSearchStores/{name},force,query,boolean,false,false,,,,,,DOCUMENTED|PREVIEW,"Optional. If set to true, any `Document`s and objects related to this `FileSearchStore` will also be deleted. If false (the default), a `FAILED_PRECONDITION` error will be returned if `FileSearchStore` contains any `Document`s.",https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,fileSearchStoreName,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. Immutable. The name of the `FileSearchStore` to upload the file into. Example: `fileSearchStores/my-file-search-store-123` Metadata-only URI: POST /v1beta/{fileSearchStoreName}:uploadToFileSearchStore; media upload URI: POST /upload/v1beta/...; resumable: /resumable/upload/v1beta/... (discovery mediaUpload).,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,displayName,body(multipart metadata),string,false,,,,,,,DOCUMENTED|PREVIEW,Optional. Display name of the created document. Visible in citations.,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,mimeType,body(multipart metadata),string,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. MIME type of the data. If not provided, it will be inferred from the uploaded content.",https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,customMetadata,body(multipart metadata),array<CustomMetadata>,false,,,,,,,DOCUMENTED|PREVIEW,Custom metadata to be associated with the data. Max 20 per Document.,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,customMetadata[].key,body(multipart metadata),string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The key of the metadata to store.,https://ai.google.dev/api/file-search/documents#CustomMetadata
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,customMetadata[].stringValue,body(multipart metadata),string,false,,,,,,,DOCUMENTED|PREVIEW,The string value of the metadata to store. One of stringValue | numericValue | stringListValue.,https://ai.google.dev/api/file-search/documents#CustomMetadata
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,customMetadata[].numericValue,body(multipart metadata),number,false,,,,,,,DOCUMENTED|PREVIEW,The numeric value of the metadata to store.,https://ai.google.dev/api/file-search/documents#CustomMetadata
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,customMetadata[].stringListValue.values,body(multipart metadata),array<string>,false,,,,,,,DOCUMENTED|PREVIEW,The string values of the metadata to store. Enables INCLUDES/EXCLUDES filter operators.,https://ai.google.dev/api/file-search/documents#StringList
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,chunkingConfig,body(multipart metadata),object (ChunkingConfig),false,,,,,,,DOCUMENTED|PREVIEW,"Optional. Config for telling the service how to chunk the data. If not provided, the service will use default parameters.",https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,chunkingConfig.whiteSpaceConfig,body(multipart metadata),object (WhiteSpaceConfig),false,,,,,,,DOCUMENTED|PREVIEW,White space chunking configuration. Only chunking algorithm exposed.,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,chunkingConfig.whiteSpaceConfig.maxTokensPerChunk,body(multipart metadata),integer,false,,,512,,,,DOCUMENTED|PREVIEW,"Maximum number of tokens per chunk. Tokens are defined as words for this chunking algorithm. Note: we are defining tokens as words split by whitespace as opposed to the output of a tokenizer. The context window of the latest gemini embedding model as of 2025-04-17 is currently 8192 tokens. We assume that the average word is 5 characters. Therefore, we set the upper limit to 2**9, which is 512 words, or 2560 tokens, assuming worst case a character per token. This is a conservative estimate meant to prevent context window overflow.",https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,chunkingConfig.whiteSpaceConfig.maxOverlapTokens,body(multipart metadata),integer,false,,,,,,,DOCUMENTED|PREVIEW,Maximum number of overlapping tokens between two adjacent chunks.,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,<file bytes>,file,"binary (any MIME, max 100 MB)",true,,,,,,,DOCUMENTED|PREVIEW,"Media payload. Simple/multipart protocol: multipart/related body (JSON metadata part + media part). Resumable protocol: first POST with headers X-Goog-Upload-Protocol: resumable, X-Goog-Upload-Command: start, X-Goog-Upload-Header-Content-Length, X-Goog-Upload-Header-Content-Type and the JSON metadata body; then POST the bytes to the returned X-Goog-Upload-URL with X-Goog-Upload-Offset: 0 and X-Goog-Upload-Command: upload, finalize. Discovery mediaUpload: accept */*, maxSize 104857600.",https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,X-Goog-Upload-Protocol,header,string,false,,,,,,,DOCUMENTED|PREVIEW,'resumable' (or 'multipart') — selects the upload protocol on the /upload/ URI.,https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,X-Goog-Upload-Command,header,string,false,,,,,,,DOCUMENTED|PREVIEW,"'start' on the initiating request; 'upload, finalize' on the byte-transfer request (resumable protocol).",https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,X-Goog-Upload-Header-Content-Length,header,string,false,,,,,,,DOCUMENTED|PREVIEW,Total byte length of the file (initiating request).,https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,X-Goog-Upload-Header-Content-Type,header,string,false,,,,,,,DOCUMENTED|PREVIEW,MIME type of the file (initiating request).,https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /upload/v1beta/fileSearchStores/{name}:uploadToFileSearchStore,X-Goog-Upload-Offset,header,string,false,,,,,,,DOCUMENTED|PREVIEW,Byte offset of the chunk being sent (0 for a single-shot upload).,https://ai.google.dev/gemini-api/docs/file-search
gemini,POST /v1beta/fileSearchStores/{name}:importFile,fileSearchStoreName,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. Immutable. The name of the `FileSearchStore` to import the file into. Example: `fileSearchStores/my-file-search-store-123`,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /v1beta/fileSearchStores/{name}:importFile,fileName,body,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The name of the `File` to import. Example: `files/abc-123` A Files API resource (files/*) uploaded earlier; File objects expire after 48 h but the imported chunks persist.,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /v1beta/fileSearchStores/{name}:importFile,customMetadata,body,array<CustomMetadata>,false,,,,,,,DOCUMENTED|PREVIEW,Custom metadata to be associated with the file. Same shape as uploadToFileSearchStore (key + stringValue|numericValue|stringListValue).,https://ai.google.dev/api/file-search/file-search-stores
gemini,POST /v1beta/fileSearchStores/{name}:importFile,chunkingConfig,body,object (ChunkingConfig),false,,,,,,,DOCUMENTED|PREVIEW,"Optional. Config for telling the service how to chunk the file. If not provided, the service will use default parameters. whiteSpaceConfig.maxTokensPerChunk / maxOverlapTokens.",https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores/{name}/documents,parent,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The name of the `FileSearchStore` containing `Document`s. Example: `fileSearchStores/my-file-search-store-123`,https://ai.google.dev/api/file-search/documents
gemini,GET /v1beta/fileSearchStores/{name}/documents,pageSize,query,integer,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. The maximum number of `Document`s to return (per page). The service may return fewer `Document`s. If unspecified, at most 10 `Document`s will be returned. The maximum size limit is 20 `Document`s per page.",https://ai.google.dev/api/file-search/documents
gemini,GET /v1beta/fileSearchStores/{name}/documents,pageToken,query,string,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. A page token, received from a previous `ListDocuments` call. Provide the `next_page_token` returned in the response as an argument to the next request to retrieve the next page. When paginating, all other parameters provided to `ListDocuments` must match the call that provided the page token.",https://ai.google.dev/api/file-search/documents
gemini,GET /v1beta/fileSearchStores/{name}/documents/{document},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The name of the `Document` to retrieve. Example: `fileSearchStores/my-file-search-store-123/documents/the-doc-abc`,https://ai.google.dev/api/file-search/documents
gemini,DELETE /v1beta/fileSearchStores/{name}/documents/{document},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,Required. The resource name of the `Document` to delete. Example: `fileSearchStores/my-file-search-store-123/documents/the-doc-abc`,https://ai.google.dev/api/file-search/documents
gemini,DELETE /v1beta/fileSearchStores/{name}/documents/{document},force,query,boolean,false,false,,,,,,DOCUMENTED|PREVIEW,"Optional. If set to true, any `Chunk`s and objects related to this `Document` will also be deleted. If false (the default), a `FAILED_PRECONDITION` error will be returned if `Document` contains any `Chunk`s.",https://ai.google.dev/api/file-search/documents
gemini,GET /v1beta/fileSearchStores/{name}/operations/{operation},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,The name of the operation resource. Operation returned by importFile.,https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores/{name}/upload/operations/{operation},name,path,string,true,,,,,,,DOCUMENTED|PREVIEW,The name of the operation resource. Operation returned by uploadToFileSearchStore (name pattern fileSearchStores/*/upload/operations/*).,https://ai.google.dev/api/file-search/file-search-stores
gemini,GET /v1beta/fileSearchStores/{name}/media/{blob},mediaId,path,string,true,,,,,,,DOCUMENTED|PREVIEW|UNVERIFIED,Media blob resource name from groundingChunks[].retrievedContext.mediaId (format fileSearchStores/{store}/media/{blobId}); downloads the cited image chunk. SDK: client.file_search_stores.download_media(media_id=...). Not in the v1beta discovery method list on 2026-09-18 (guide shows a /v1/ URL).,https://ai.google.dev/gemini-api/docs/generate-content/file-search
gemini,POST /v1beta/models/{model}:generateContent,tools[].fileSearch.fileSearchStoreNames,body,array<string>,true,,,,,,,DOCUMENTED|PREVIEW,Required. The names of the file_search_stores to retrieve from. Example: `fileSearchStores/my-file-search-store-123`,https://ai.google.dev/api/generate-content#FileSearch
gemini,POST /v1beta/models/{model}:generateContent,tools[].fileSearch.metadataFilter,body,string,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. Metadata filter to apply to the semantic retrieval documents and chunks. AIP-160 list-filter syntax over customMetadata, e.g. author = ""Robert Graves"", year > 1930.",https://ai.google.dev/api/generate-content#FileSearch
gemini,POST /v1beta/models/{model}:generateContent,tools[].fileSearch.topK,body,integer,false,,,,,,,DOCUMENTED|PREVIEW|UNVERIFIED,Optional. The number of semantic retrieval chunks to retrieve. Only in REST reference/discovery (not in the guide).,https://ai.google.dev/api/generate-content#FileSearch
gemini,POST /upload/v1beta/files,X-Goog-Upload-Protocol,header,string,false,,,,resumable,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Upload protocol selector for the first request: `resumable` (start a session) — the only value used by the docs; simple uploads use the `uploadType` query instead.,https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,X-Goog-Upload-Command,header,string,false,,,,"start|upload|finalize|upload, finalize|query|cancel",all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"`start` on the metadata request; then `upload, finalize` (single shot), or `upload` per chunk followed by `finalize`; `query` and `cancel` are the resumable protocol's control commands (on X-Goog-Upload-Control-URL).",https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,X-Goog-Upload-Header-Content-Length,header,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Total byte size of the media to be uploaded (start request).,https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,X-Goog-Upload-Header-Content-Type,header,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,MIME type of the media (start request); becomes File.mimeType.,https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,X-Goog-Upload-Offset,header,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Byte offset of the chunk being sent (0 for a single-shot upload). Chunks must respect X-Goog-Upload-Chunk-Granularity (8 MiB observed).,https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,X-Goog-Upload-URL (response),header,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Session URL returned by `start` (absolute URL on generativelanguage.googleapis.com with upload_id); POST the bytes there. Also returned: X-Goog-Upload-Control-URL, X-Goog-Upload-Status (active → final), X-Goog-Upload-Chunk-Granularity: 8388608.",https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,uploadType,query,string,false,,,,multipart|media,all generateContent models unless noted,,LIVE_DISCOVERED|LIVE_VERIFIED,"Simple-upload selector: `multipart` (multipart/related body: JSON metadata part + media part) or `media` (raw bytes, Content-Type = media type). Legacy Google upload protocol, verified working.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /upload/v1beta/files,file.displayName,body,string,false,,,512,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Human-readable name ≤ 512 chars (JSON metadata in the start request / multipart metadata part).,https://ai.google.dev/api/files
gemini,POST /upload/v1beta/files,file.name,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,"Optional custom id `files/<id>`: ≤ 40 lowercase alphanumerics or dashes, no leading/trailing dash; server generates one when omitted (observed 12-char ids like files/w7x6n2hlpxrq).",https://ai.google.dev/api/files
gemini,GET /v1beta/files,pageSize,query,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Max Files per page (server default when omitted).,https://ai.google.dev/api/files
gemini,GET /v1beta/files,pageToken,query,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,nextPageToken from a previous call.,https://ai.google.dev/api/files
gemini,GET /v1beta/files/{id},name,path,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,`files/{id}`. Unknown/deleted → 403 PERMISSION_DENIED (not 404).,https://ai.google.dev/api/files
gemini,POST /v1beta/files:register,uris[],body,array<string>,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Google Cloud Storage URIs `gs://bucket/object` to register (not copied). Any failure fails the whole call. Registered files have `source: REGISTERED` and no expirationTime in the response (docs: valid up to 30 days).,https://ai.google.dev/api/files
gemini,GET /v1beta/generatedFiles,pageSize / pageToken,query,integer / string,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Pagination for generated files.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,model,path,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Model id (`gemini-3.5-flash-lite`, `gemini-3.8-flash`, …) or `models/{id}` prefix implied by the route; also `tunedModels/{id}` and `dynamic/{id}` resources. Unknown id → 404 NOT_FOUND 'models/<id> is not found for API version v1beta, or is not supported for generateContent. Call ModelService.ListModels …'. Embedding models → same 404.",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,alt,query,string,false,,,,json|sse|media|proto,all generateContent models unless noted,,LIVE_DISCOVERED|LIVE_VERIFIED,Response format. `sse` turns even the non-streaming method into a one-event `text/event-stream`; `json` (default); `media`/`proto` from the discovery global params.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,key,query,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,API key as a query parameter (discovery global parameter). Works but the key then appears in URLs/logs — this atlas never uses it; send `x-goog-api-key` instead.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,fields|prettyPrint|quotaUser|$.xgafv|callback,query,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,"Standard Google API system parameters (partial response FieldMask, pretty printing, quota attribution, error format 1|2, JSONP).",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,x-goog-api-key,header,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,API key header (recommended auth). Missing/invalid → 400/403 with status INVALID_ARGUMENT/PERMISSION_DENIED.,https://ai.google.dev/gemini-api/docs/api-key
gemini,POST /v1beta/models/{model}:generateContent,contents,body,array<Content>,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Required conversation: ordered `Content` turns. Single-turn = one user Content. Multi-turn = alternating user/model turns; the last turn must be a user turn ('Requests ending with a model turn are not supported.'). A bare Content object instead of an array is accepted (proto JSON leniency, verified). Missing → 400 '* GenerateContentRequest.contents: contents is not specified'.",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,contents[].role,body,string,false,,,,user|model,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Producer of the turn: `user` or `model`. Optional for single-turn (omitted → treated as user, verified). `assistant`/`system` are not valid roles (request with a final `assistant` turn was rejected as a model turn).",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts,body,array<Part>,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Ordered parts of one turn; each Part carries exactly one data field (text | inlineData | fileData | functionCall | functionResponse | executableCode | codeExecutionResult | toolCall | toolResponse) plus optional metadata (thought, thoughtSignature, videoMetadata, mediaResolution, mediaProcessing, partMetadata). Empty parts array → see l_empty_parts.",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].text,body,string,true,,,,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED,"Text to speak, optionally prefixed with natural-language direction (""Say cheerfully: …"", audio profile / scene / director's notes) and inline audio tags like [whispers], [laughs], [very fast]. Context window 32k tokens; input token limit 8,192.",https://ai.google.dev/gemini-api/docs/speech-generation#prompting-guide
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData.mimeType,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"IANA MIME type of the inline bytes. Documented families: image/png, image/jpeg, image/webp, image/heic, image/heif, image/gif, image/avif; audio/*; video/*; text/plain, text/html, text/css, text/javascript, text/x-typescript, text/csv, text/markdown, text/x-python, text/xml, text/rtf; application/pdf, application/json, application/x-javascript, application/x-typescript, application/x-python-code, application/x-ipynb+json, application/rtf. Unsupported type → 400 (see k_inline_bad_mime).",https://ai.google.dev/gemini-api/docs/generate-content/file-input-methods
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData.data,body,string (base64),false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Raw media bytes, base64-encoded. Total inline request payload limit: 20 MB (image/audio/video guides) — larger media must go through the Files API.",https://ai.google.dev/gemini-api/docs/generate-content/image-understanding
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData.displayName,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,Name the model can use to refer to the blob (e.g. from function responses with verbalization_mode REFERENCE_ONLY).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].fileData.fileUri,body,string,true,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"URI of a File (`https://generativelanguage.googleapis.com/v1beta/files/<id>` from the Files API, or a registered GCS file), a public YouTube URL (`https://www.youtube.com/watch?v=…`, preview, ≤ 10 videos/request on 2.5+, free tier ≤ 8 h/day) or a publicly accessible HTTPS/pre-signed media URL (≤ 100 MB, fetched per request). Deleted/unknown file → 403 PERMISSION_DENIED.",https://ai.google.dev/gemini-api/docs/generate-content/file-input-methods
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].fileData.mimeType,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Optional MIME type of the referenced file; omitted → taken from the File resource (verified for text/plain).,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].fileData.displayName,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,Reference name for the file (see inlineData.displayName).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionCall,body,"object {name, args, id}",false,,,,,all generateContent models unless noted,,DOCUMENTED,"Model-emitted tool call (shape only here — see the tools domain). When echoing history back, keep the part's `thoughtSignature`.",https://ai.google.dev/gemini-api/docs/function-calling
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse,body,"object {name, response, id, parts[], scheduling, willContinue}",false,,,,,all generateContent models unless noted,,DOCUMENTED,Client tool result (shape only — tools domain). `response` is a free-form JSON object; multimedia via `parts[].inlineData` referenced with `$ref` = displayName.,https://ai.google.dev/gemini-api/docs/function-calling
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].executableCode / codeExecutionResult,body,object,false,,,,,all generateContent models unless noted,,DOCUMENTED,"Code-execution tool parts (shape only — tools domain): {language: PYTHON, code, id} / {outcome: OUTCOME_OK|OUTCOME_FAILED|OUTCOME_DEADLINE_EXCEEDED, output, id}.",https://ai.google.dev/gemini-api/docs/code-execution
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall / toolResponse,body,object,false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED|DOCUMENTED,Server-side built-in tool invocations (toolType GOOGLE_SEARCH_WEB|GOOGLE_SEARCH_IMAGE|URL_CONTEXT|GOOGLE_MAPS|FILE_SEARCH) surfaced when `toolConfig.includeServerSideToolInvocations` is true; the client echoes them back verbatim (shape only — tools domain).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].thought,body,boolean,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Marks a thought-summary part (output; set only when `thinkingConfig.includeThoughts` is true and the model produced a summary). Sending thought parts back in history is allowed (SDKs do it).,https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].thoughtSignature,body,string (base64 bytes),false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Opaque encrypted reasoning state. Gemini 3 models attach one to the last part of every response (text parts included, even without thinking) and to the first functionCall of each step. Return it unchanged with the part: mandatory for function calls in the current turn on Gemini 3 (400 'Function call <name> in the <n>. content block is missing a thought_signature'), recommended for text. A tampered value → 400 'Corrupted thought signature.'; the documented dummy values `skip_thought_signature_validator` / `context_engineering_is_the_way_to_go` bypass validation (verified).",https://ai.google.dev/gemini-api/docs/generate-content/thought-signatures
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].videoMetadata.startOffset / endOffset,body,"string (Duration, e.g. '40s')",false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Clip the video (inlineData or fileData incl. YouTube) to a window; static processing only. Discovery marks VideoMetadata deprecated in favour of `GenerateContentRequest.processing_options` (not yet exposed).,https://ai.google.dev/gemini-api/docs/generate-content/video-understanding
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].videoMetadata.fps,body,number,false,1.0,0,24,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Frame sampling rate sent to the model; default 1.0; range (0.0, 24.0]. Token usage scales with fps.",https://ai.google.dev/gemini-api/docs/generate-content/video-understanding
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].mediaResolution.level,body,string,false,,,,MEDIA_RESOLUTION_UNSPECIFIED|MEDIA_RESOLUTION_LOW|MEDIA_RESOLUTION_MEDIUM|MEDIA_RESOLUTION_HIGH|MEDIA_RESOLUTION_ULTRA_HIGH,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-flash-lite|gemini-3.1-pro-preview|gemini-3-flash-preview,,DOCUMENTED|LIVE_VERIFIED,"Per-part media resolution (Gemini 3 only) overriding `generationConfig.mediaResolution`; adds ULTRA_HIGH (2240 image tokens, needed for computer use). Observed on gemini-3.5-flash-lite for a 1×1 PNG: default/HIGH 1089, MEDIUM 529 IMAGE tokens, LOW 256 IMAGE tokens (per-part mediaResolution.level MEDIA_RESOLUTION_LOW), ULTRA_HIGH 2209 IMAGE tokens.",https://ai.google.dev/gemini-api/docs/generate-content/media-resolution
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].mediaProcessing,body,string,false,,,,MEDIA_PROCESSING_UNSPECIFIED|STATIC|AGENTIC,all generateContent models unless noted,,DOCUMENTED|UNVERIFIED,"Video-only: STATIC = fixed 1 fps frame extraction (all frames in context); AGENTIC = model-driven navigation loading only needed frames/transcript/audio (up to 88 % fewer tokens; reported in usageMetadata.toolUsePromptTokenCount + thoughtsTokenCount). Supported for agentic: gemini-3.8/3.7/3.6-flash, gemini-3.5-flash-lite.",https://ai.google.dev/gemini-api/docs/generate-content/video-understanding
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].partMetadata,body,object (Struct),false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED|UNVERIFIED,"Free-form client metadata attached to a Part (agent bookkeeping, e.g. source file name). Discovery only.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].audioTranscription,body,"object {text, speakerLabel, words[]}",false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED,Output-only transcription attached to audio parts by transcription models (`generationConfig.audioTranscriptionConfig`) — media domain.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,systemInstruction,body,Content,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Developer instructions applied to the whole request (text parts only). `role` is ignored (`system`, omitted both work). snake_case `system_instruction` also accepted. Counted in promptTokenCount. Cannot be combined with a cache that already has one (see gemini-cached-contents.json).",https://ai.google.dev/gemini-api/docs/generate-content/text-generation
gemini,POST /v1beta/models/{model}:generateContent,tools,body,array<Tool>,false,,,,,,,DOCUMENTED,"Optional. List of Tool objects the model may use. Each Tool object carries one or more of: functionDeclarations[], googleSearch, googleSearchRetrieval (legacy), googleMaps, urlContext, codeExecution, computerUse, fileSearch, mcpServers[]. Several tools may be combined in one request (see tool-combination rules).",https://ai.google.dev/api/generate-content#request-body
gemini,POST /v1beta/models/{model}:generateContent,toolConfig,body,"object {functionCallingConfig:{mode: AUTO|ANY|NONE|VALIDATED, allowedFunctionNames[]}, retrievalConfig:{latLng, languageCode}, includeServerSideToolInvocations}",false,,,,,all generateContent models unless noted,,DOCUMENTED,Tool behaviour (shape only — tools domain).,https://ai.google.dev/gemini-api/docs/function-calling
gemini,POST /v1beta/models/{model}:generateContent,safetySettings,body,array<SafetySetting>,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Per-category block thresholds; at most one entry per category (a duplicate category was silently accepted live). Default when absent: filters OFF for Gemini 2.5/3 (only the built-in child-safety protections apply). Adding any setting makes the response carry `candidates[].safetyRatings` for the 4 adjustable categories (verified: 4 NEGLIGIBLE ratings); with BLOCK_NONE/OFF no ratings are returned.,https://ai.google.dev/gemini-api/docs/safety-settings
gemini,POST /v1beta/models/{model}:generateContent,safetySettings[].category,body,string,true,,,,HARM_CATEGORY_HARASSMENT|HARM_CATEGORY_HATE_SPEECH|HARM_CATEGORY_SEXUALLY_EXPLICIT|HARM_CATEGORY_DANGEROUS_CONTENT|HARM_CATEGORY_CIVIC_INTEGRITY|HARM_CATEGORY_JAILBREAK,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Harm category. Accepted live: HARM_CATEGORY_HARASSMENT, HATE_SPEECH, SEXUALLY_EXPLICIT, DANGEROUS_CONTENT, CIVIC_INTEGRITY (deprecated → use generationConfig.enableEnhancedCivicAnswers; still accepted, ratings for the 4 standard categories only), JAILBREAK (accepted). PaLM categories (DEROGATORY, TOXICITY, VIOLENCE, SEXUAL, MEDICAL, DANGEROUS) → 400 '* GenerateContentRequest.safety_settings[0]: element predicate failed: $.category in (…)'. Unknown → 400 'Invalid value at safety_settings[0].category'.",https://ai.google.dev/gemini-api/docs/safety-settings
gemini,POST /v1beta/models/{model}:generateContent,safetySettings[].threshold,body,string,true,,,,HARM_BLOCK_THRESHOLD_UNSPECIFIED|BLOCK_LOW_AND_ABOVE|BLOCK_MEDIUM_AND_ABOVE|BLOCK_ONLY_HIGH|BLOCK_NONE|OFF,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Block probability threshold: OFF (filter off) · BLOCK_NONE (always show) · BLOCK_ONLY_HIGH · BLOCK_MEDIUM_AND_ABOVE · BLOCK_LOW_AND_ABOVE · HARM_BLOCK_THRESHOLD_UNSPECIFIED (default). Blocking is by probability, not severity.",https://ai.google.dev/gemini-api/docs/safety-settings
gemini,POST /v1beta/models/{model}:generateContent,cachedContent,body,string,false,,,,,all generateContent models unless noted,,DOCUMENTED|ACCOUNT_RESTRICTED,"Name of an explicit cache (`cachedContents/{id}`) used as prompt prefix; the cache's model must equal the request model, and the request must not repeat systemInstruction/tools already in the cache. usageMetadata.cachedContentTokenCount reports the cached part (promptTokenCount still includes it).",https://ai.google.dev/gemini-api/docs/generate-content/caching
gemini,POST /v1beta/models/{model}:generateContent,serviceTier,body,string,false,standard,,,unspecified|standard|flex|priority,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Processing tier: standard (default; echoed as usageMetadata.serviceTier 'standard'), flex (cheaper, best-effort latency — flex-inference guide), priority (2× price, priority-inference guide). Invalid value → 400 'Invalid value at service_tier (…ServiceTier), ""turbo""'.",https://ai.google.dev/gemini-api/docs/flex-inference
gemini,POST /v1beta/models/{model}:generateContent,labels,body,"map<string,string>",false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"User metadata (Cloud label rules: keys start with a letter, ≤ 63 chars, lowercase/digits/_/-). Documented use: `safety_identifier` for aggregators (e.g. {""safety_identifier"": ""user_session_123""}).",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,store,body,boolean,false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED|LIVE_VERIFIED,Per-request override of the project logging (prompt-logging) configuration.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,generationConfig,body,GenerationConfig,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Sampling/output configuration object. Unknown keys → 400 'Invalid JSON payload received. Unknown name ""<k>"" at generation_config: Cannot find field.' (fieldViolations detail).",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.temperature,body,number,false,1.0,0.0,2.0,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Randomness, range [0.0, 2.0] (verified message: 'temperature must be in the range [0.0, 2.0]'). Model defaults from models.get: 1.0 for gemini-3.5-flash-lite (maxTemperature 2). Gemini 3.x: Google strongly recommends leaving it at 1.0 (lower values can loop/degrade).",https://ai.google.dev/gemini-api/docs/generate-content/gemini-3
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.topP,body,number,false,0.95,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Nucleus sampling cumulative probability. Default per model (0.95 for gemini-3.5-flash-lite). Not recommended to change on Gemini 3.x.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.topK,body,integer,false,64,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Top-k sampling; default per model (64 for gemini-3.5-flash-lite; models with no topK in models.get reject the field). Not recommended to change on Gemini 3.x.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.candidateCount,body,integer,false,1,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Number of candidates. Gemini 3.x models do not support values > 1 (docs: 400). Default 1.,https://ai.google.dev/gemini-api/docs/generate-content/gemini-3
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.maxOutputTokens,body,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Hard cap on generated tokens INCLUDING thinking tokens; hitting it → finishReason MAX_TOKENS and possibly empty text (verified: thinkingBudget -1 with maxOutputTokens 64 returned only a thought part, thoughtsTokenCount 60, no candidatesTokenCount). Default = model outputTokenLimit (65536 for 3.x Flash/Pro).",https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.stopSequences,body,array<string>,false,,,5,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Up to 5 stop strings (6 → 400 '* GenerateContentRequest.generation_config.stop_sequences: the number of stop_sequences must not exceed 5.'); the sequence is not included in the output.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.presencePenalty,body,number,false,,,,,all generateContent models unless noted,,DOCUMENTED|FAILED_VERIFICATION,Binary penalty on tokens already present in the response (positive = more vocabulary).,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.frequencyPenalty,body,number,false,,,,,all generateContent models unless noted,,DOCUMENTED|FAILED_VERIFICATION,Penalty proportional to the number of prior uses of a token; large negative values make the model repeat until maxOutputTokens.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.seed,body,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Decoding seed (random when unset). Determinism is best-effort.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseLogprobs,body,boolean,false,,,,,all generateContent models unless noted,,DOCUMENTED|FAILED_VERIFICATION,Return log-probabilities in candidates[].logprobsResult (+ avgLogprobs).,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.logprobs,body,integer,false,,0,20,,all generateContent models unless noted,,DOCUMENTED|FAILED_VERIFICATION,Number of top logprobs per decoding step in logprobsResult.topCandidates (0–20); only valid with responseLogprobs=true.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseMimeType,body,string,false,text/plain,,,text/plain|application/json|application/xml|application/yaml|text/x.enum,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Output MIME type. Allowed (verbatim 400 text): `text/plain` (default), `application/json`, `application/xml`, `application/yaml`, `text/x.enum`. `text/x.enum` returns one enum string from responseSchema. Discovery: deprecated in favour of responseFormat.",https://ai.google.dev/gemini-api/docs/generate-content/structured-output
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseSchema,body,Schema (OpenAPI 3.0 subset),false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Legacy typed schema: {type: STRING|NUMBER|INTEGER|BOOLEAN|ARRAY|OBJECT|NULL, format, description, nullable, enum[], items, properties, required[], propertyOrdering[], minItems, maxItems, minimum, maximum, minLength, maxLength, pattern, anyOf[], title, default, example, minProperties, maxProperties}. Requires a compatible responseMimeType. Non-Schema keywords (e.g. additionalProperties) → 400 'Unknown name ""additionalProperties"" at generation_config.response_schema'. Deprecated in discovery (use responseFormat).",https://ai.google.dev/gemini-api/docs/generate-content/structured-output
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseJsonSchema,body,object (JSON Schema),false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Standard JSON Schema alternative to responseSchema (mutually exclusive; responseMimeType application/json required). Supported keywords: $id, $defs, $ref, $anchor, type (arrays like [""string"",""null""]), format, title, description, enum, items, prefixItems, minItems, maxItems, minimum, maximum, anyOf, oneOf (=anyOf), properties, additionalProperties, required, plus non-standard propertyOrdering. Unsupported keywords are ignored. Cyclic $ref only in non-required properties; a $ref sub-schema may carry only $-keys.",https://ai.google.dev/gemini-api/docs/generate-content/structured-output
gemini,POST /v1beta/models/{model}:generateContent,generationConfig._responseJsonSchema,body,object,false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED,Discovery lists this underscore-prefixed twin with the full description and marks `responseJsonSchema` as 'An internal detail' — a documentation quirk; send `responseJsonSchema`.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseFormat,body,"ResponseFormatConfig {text:{mimeType: APPLICATION_JSON|TEXT_PLAIN, schema}, audio:{mimeType, sampleRate, bitRate, delivery}, image:{mimeType, aspectRatio, imageSize, delivery}}",false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"NEW flat per-modality output format (the current docs use it for structured output: `responseFormat.text.mimeType: ""application/json""` + `schema`). Enum-style values (`APPLICATION_JSON`) are accepted on the wire; live test with the docs' lowercase MIME string: 400 'Invalid value at generation_config.response_format.text.mime_type (…TextResponseFormat.MimeType), ""application/json""' — the wire enum is APPLICATION_JSON | TEXT_PLAIN; the docs' lowercase MIME strings only work through the SDKs. Audio/image branches are media-domain shapes.",https://ai.google.dev/gemini-api/docs/generate-content/structured-output
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseModalities,body,array<string>,false,,,,MODALITY_UNSPECIFIED|TEXT|IMAGE|AUDIO,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Requested output modalities (exact match with a supported combination; empty = TEXT). Text models accept [TEXT]; [TEXT, IMAGE] on a text model: 200 — text model silently returned text only for [TEXT, IMAGE] (no error despite the documented 'exact match' rule).",https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.speechConfig,body,"SpeechConfig {voiceConfig:{prebuiltVoiceConfig:{voiceName}}, multiSpeakerVoiceConfig:{speakerVoiceConfigs[{speaker, voiceConfig}]}, languageCode}",false,,,,,all generateContent models unless noted,,DOCUMENTED,TTS voice selection (shape only — media domain). On a text model: 200 — speechConfig ignored on a text model.,https://ai.google.dev/gemini-api/docs/speech-generation
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.imageConfig,body,"ImageConfig {aspectRatio: 1:1|1:4|4:1|1:8|8:1|2:3|3:2|3:4|4:3|4:5|5:4|9:16|16:9|21:9, imageSize: 512|1K|2K|4K}",false,,,,,all generateContent models unless noted,,DOCUMENTED,Image generation options (shape only — media domain). On a text model: 400 INVALID_ARGUMENT 'Aspect ratio is not enabled for this model' (imageConfig.aspectRatio on gemini-3.5-flash-lite).,https://ai.google.dev/gemini-api/docs/image-generation
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.mediaResolution,body,string,false,,,,MEDIA_RESOLUTION_UNSPECIFIED|MEDIA_RESOLUTION_LOW|MEDIA_RESOLUTION_MEDIUM|MEDIA_RESOLUTION_HIGH,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Global media resolution for all media parts (per-part override via parts[].mediaResolution on Gemini 3). Token budgets (Gemini 3): image LOW 280 / MEDIUM 560 / HIGH & default 1120 / ULTRA_HIGH 2240 (per-part only); video 70 (LOW/MEDIUM) / 280 (HIGH) per frame; PDF 280/560/1120 + native text; audio fixed 25/s. Observed on gemini-3.5-flash-lite 1×1 PNG: default 1089, LOW 256 IMAGE tokens. Ignored on text-only requests (200, ignored on text-only input). ULTRA_HIGH globally: 400 'Invalid value at generation_config.media_resolution (…GenerationConfig.MediaResolution), ""MEDIA_RESOLUTION_ULTRA_HIGH""' — ULTRA_HIGH is per-part only.",https://ai.google.dev/gemini-api/docs/generate-content/media-resolution
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.thinkingConfig,body,"ThinkingConfig {thinkingLevel, thinkingBudget, includeThoughts}",false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Reasoning controls. Error on models without thinking. thinkingLevel and thinkingBudget are mutually exclusive → 400 'You can only set only one of thinking budget and thinking level.',https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.thinkingConfig.thinkingLevel,body,string,false,,,,THINKING_LEVEL_UNSPECIFIED|MINIMAL|LOW|MEDIUM|HIGH,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-flash-lite|gemini-3.1-pro-preview|gemini-3-flash-preview,,DOCUMENTED|LIVE_VERIFIED,"Gemini 3 reasoning depth (case-insensitive: 'low' and 'HIGH' both accepted). Defaults: high (3.1 Pro, 3 Flash), medium (3.5–3.8 Flash), minimal (3.5/3.1 Flash-Lite). Support: minimal NOT on 3.8/3.7 Flash or 3.1 Pro (400 'Thinking level MINIMAL is not supported for this model. Please retry with other thinking level.'); low/medium/high on all Gemini 3 text models; 3.1 Flash-Lite Image: minimal/high only. Not supported on 2.5 (use thinkingBudget). Unknown value → 400 'Invalid value at generation_config.thinking_config.thinking_level'.",https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.thinkingConfig.thinkingBudget,body,integer,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Thinking-token budget (Gemini 2.5 style; accepted by 3.x for compatibility but not recommended). -1 = dynamic; 0 = off where allowed (2.5 Flash/Flash-Lite 0–24576, 2.5 Pro 128–32768 cannot disable). Gemini 3: 3.1 Pro cannot disable (429 RESOURCE_EXHAUSTED on gemini-3.1-pro-preview (free-tier quota for Pro exhausted/zero) — not verifiable on this key; docs: cannot disable thinking); 3.5 Flash-Lite budget 0 → 400 'Request contains an invalid argument.'; 3.5 Flash budget 0: 200 on gemini-3.5-flash with thinkingBudget 0: 'OK.', no thoughtsTokenCount (thinking off accepted, 5.7 s); over-large budget 100000 on Flash-Lite: 400 '* GenerateContentRequest.generation_config.thinking_config.thinking_budget: thinking_budget must be in the range [-1, 65535]'. The model may overflow/underflow the budget.",https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.thinkingConfig.includeThoughts,body,boolean,false,false,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,"Return thought summaries as parts with `thought: true` (before the answer part). Summaries appear only when the model reasoned enough; billing is on full thought tokens (thoughtsTokenCount), not the summary. Without it, thinking still happens and is billed.",https://ai.google.dev/gemini-api/docs/generate-content/thinking
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTimestamp,body,boolean,false,,,,,all generateContent models unless noted,,DOCUMENTED|FAILED_VERIFICATION,"SDK field (GenerateContentConfig.audio_timestamp) for timestamp understanding of audio input. NOT present in the v1beta discovery GenerationConfig: 400 'Invalid JSON payload received. Unknown name ""audioTimestamp"" at generation_config: Cannot find field.' — SDK-only/Vertex field, rejected by the Gemini API.",https://ai.google.dev/gemini-api/docs/generate-content/audio
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.enableEnhancedCivicAnswers,body,boolean,false,,,,,all generateContent models unless noted,,DOCUMENTED|LIVE_VERIFIED,Enhanced answers on civic/election topics (replaces the deprecated HARM_CATEGORY_CIVIC_INTEGRITY setting); may be unavailable on some models.,https://ai.google.dev/api/generate-content
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.enableAffectiveDialog,body,boolean,false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED|UNVERIFIED,Model detects emotions and adapts (Live/native-audio models; discovery lists it on GenerationConfig).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig,body,"AudioTranscriptionConfig {languageCodes[], languageHints, languageAuto, wordTimestamp, diarization, adaptationPhrases[], customVocabulary[], mode: VERBATIM|SMART}",false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED|DOCUMENTED,"Speech-recognition config for transcription models (gemini-3.5-transcribe) — shape only, media domain.",https://ai.google.dev/gemini-api/docs/transcribe
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.translationConfig,body,"TranslationConfig {targetLanguageCode, echoTargetLanguage}",false,,,,,all generateContent models unless noted,,LIVE_DISCOVERED,"Translation config (Live translate model) — shape only, media domain.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.modelSelectionConfig / routingConfig / modelArmorConfig,body,object,false,,,,,all generateContent models unless noted,,UNVERIFIED,Present in the google-genai SDK GenerateContentConfig but NOT in the Gemini API discovery document — Vertex AI-only fields; sending them to generativelanguage.googleapis.com yields 400 Unknown name.,https://github.com/googleapis/python-genai (types.py)
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseModalities,body,array<enum>,true,,,,AUDIO,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED|LIVE_VERIFIED,"TTS models: must be [""AUDIO""] (text-only input, audio-only output).",https://ai.google.dev/gemini-api/docs/speech-generation#single-speaker-tts
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.imageConfig,body,object (ImageConfig),false,,,,,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,Image generation options. The API returns an error if set for models that do not support image generation.,https://ai.google.dev/api/generate-content#ImageConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.imageConfig.aspectRatio,body,string,false,,,,1:1|1:4|4:1|1:8|8:1|2:3|3:2|3:4|4:3|4:5|5:4|9:16|16:9|21:9,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,"Output aspect ratio. Default: matches the input image, otherwise 1:1. 1:4/4:1/1:8/8:1 only on gemini-3.1-flash-image; 2.5-flash-image and 3-pro-image support the 10 classic ratios.",https://ai.google.dev/api/generate-content#ImageConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.imageConfig.imageSize,body,string,false,1K,,,512|1K|2K|4K,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview,,DOCUMENTED,"Output resolution. Uppercase K required (""1k"" is rejected). 512 (0.5K) only on gemini-3.1-flash-image; gemini-3.1-flash-lite-image supports 1K only; gemini-3-pro-image 1K/2K/4K; gemini-2.5-flash-image fixed ~1024px (no imageSize).",https://ai.google.dev/api/generate-content#ImageConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.thinkingConfig.thinkingLevel,body,enum,false,MINIMAL,,,MINIMAL|HIGH,gemini-3.1-flash-image|gemini-3.1-flash-lite-image,,DOCUMENTED,Thinking depth for Gemini 3.1 image models (thinking is always on and billed; cannot be disabled). Default MINIMAL for 3.1 Flash/Flash-Lite Image.,https://ai.google.dev/gemini-api/docs/image-generation#thinking-process
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseFormat.image,body,object (ImageResponseFormat),false,,,,,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED|UNVERIFIED,"Flat per-modality output format (newer alternative to imageConfig): mimeType (IMAGE_JPEG…), delivery (INLINE|URI), aspectRatio, imageSize enums.",https://ai.google.dev/api/generate-content#ResponseFormatConfig
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData,body,"object {mimeType, data(base64)}",false,,,,,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,Reference/input images for editing and composition. Up to 14 reference images on Gemini 3 image models (3.1 Flash Lite: 14 objects; 3.1 Flash: 10 objects + 4 characters; 3 Pro: 6 objects + 5 characters + 3 style). 2.5-flash-image works best with <= 3.,https://ai.google.dev/gemini-api/docs/image-generation#use-up-to-14-reference-images
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].fileData,body,"object {mimeType, fileUri}",false,,,,,gemini-3.1-flash-image|gemini-3.1-flash-lite-image,,DOCUMENTED,Video (Files API URI or public YouTube URL) as multimodal reference for video-to-image generation. Only gemini-3.1-flash-image and gemini-3.1-flash-lite-image. Audio inputs are not supported by image models.,https://ai.google.dev/gemini-api/docs/image-generation#video-to-image
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch,body,object,false,,,,,gemini-3.1-flash-image|gemini-3-pro-image,,DOCUMENTED,Grounding with Google Search for image generation (real-time data). Not supported by gemini-3.1-flash-lite-image. gemini-3.1-flash-image adds Image Search grounding (search_types image_search in the Interactions API); search suggestions must be displayed.,https://ai.google.dev/gemini-api/docs/image-generation#grounding-with-google-search
gemini,POST /v1beta/models/{model}:generateContent,contents (multi-turn),body,array<Content>,false,,,,,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,"Multi-turn editing: resend the conversation (user text, model image part, new user instruction). Recommended way to iterate on an image.",https://ai.google.dev/gemini-api/docs/image-generation#multi-turn-image-editing
gemini,POST /v1beta/interactions,response_format,body,object|array (ResponseFormat),false,,,,,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,"Interactions API: {type:'image', mime_type:'image/jpeg', aspect_ratio, image_size} returns only images; pass an array [{type:'text'},{type:'image'}] for interleaved text+image.",https://ai.google.dev/gemini-api/docs/image-generation#optional-configurations
gemini,POST /v1beta/interactions,response_format.image_size,body,enum,false,,,,512|1K|2K|4K,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview,,DOCUMENTED,Image size (uppercase K).,https://ai.google.dev/api/interactions-api
gemini,POST /v1beta/interactions,response_format.aspect_ratio,body,enum,false,,,,1:1|1:4|4:1|1:8|8:1|2:3|3:2|3:4|4:3|4:5|5:4|9:16|16:9|21:9,gemini-3.1-flash-lite-image|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3-pro-image-preview|nano-banana-pro-preview|gemini-3.1-flash-image-preview|gemini-2.5-flash-image,,DOCUMENTED,Aspect ratio for image output.,https://ai.google.dev/api/interactions-api
gemini,POST /v1beta/interactions,generation_config.thinking_level,body,enum,false,minimal,,,minimal|high,gemini-3.1-flash-image|gemini-3.1-flash-lite-image,,DOCUMENTED,Thinking level for 3.1 image models.,https://ai.google.dev/gemini-api/docs/image-generation#controlling-thinking-levels
gemini,POST /v1beta/interactions,previous_interaction_id,body,string,false,,,,,*,,DOCUMENTED|BETA,Id of the previous interaction to continue server-side state (history only; tools/system_instruction/generation_config are interaction-scoped). 400 if the previous interaction is still in_progress. Agents additionally need `environment`.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[].search_types,body,array<enum>,false,,,,web_search|image_search,gemini-3.1-flash-image,,DOCUMENTED,"Google Search tool: ['web_search','image_search']; image_search only on gemini-3.1-flash-image.",https://ai.google.dev/gemini-api/docs/image-generation#image-search
gemini,POST /v1beta/models/{model}:predict,instances[].prompt,body,string,false,,,,,imagen-4.0-generate-001|imagen-4.0-ultra-generate-001|imagen-4.0-fast-generate-001,,DOCUMENTED|RETIRED,Imagen text prompt (<= 480 tokens). Imagen models are SHUT DOWN in the Gemini API (imagen-4.0-* shut down 2026-08-17); `predict` is kept in the discovery document but no Imagen model is served.,https://ai.google.dev/gemini-api/docs/imagen
gemini,POST /v1beta/models/{model}:predict,parameters.sampleCount,body,integer,false,,1,4,,,,DOCUMENTED|RETIRED,Imagen: number of images (1-4). Historical; retired with Imagen.,https://ai.google.dev/gemini-api/docs/imagen
gemini,POST /v1beta/models/{model}:predict,parameters.aspectRatio,body,string,false,,,,1:1|3:4|4:3|9:16|16:9,,,DOCUMENTED|RETIRED,"Imagen: 1:1, 3:4, 4:3, 9:16, 16:9. Historical; retired with Imagen.",https://ai.google.dev/gemini-api/docs/imagen
gemini,POST /v1beta/models/{model}:predict,parameters.personGeneration,body,string,false,,,,dont_allow|allow_adult|allow_all,,,DOCUMENTED|RETIRED,Imagen: dont_allow | allow_adult | allow_all. Historical; retired with Imagen.,https://ai.google.dev/gemini-api/docs/imagen
gemini,POST /v1beta/models/{model}:predict,parameters.imageSize,body,string,false,,,,1K|2K,,,DOCUMENTED|RETIRED,Imagen 4: 1K | 2K. Historical; retired with Imagen.,https://ai.google.dev/gemini-api/docs/imagen
gemini,POST /v1beta/interactions,x-goog-api-key,header,string,true,,,,,*,,DOCUMENTED|BETA,Gemini API key. Never put the key in the URL.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,Api-Revision,header,string,false,,,,2026-05-20|2026-05-07,*,,DOCUMENTED|BETA,Schema revision header. `2026-05-20` = current steps/response_format schema (default since 2026-05-26); `2026-05-07` = legacy `outputs` schema (removed 2026-06-08).,https://ai.google.dev/gemini-api/docs/interactions-breaking-changes-may-2026
gemini,POST /v1beta/interactions,api_version,path,string,true,,,,v1beta|v1,*,,DOCUMENTED|BETA,Which version of the API to use (path segment: v1beta | v1).,https://ai.google.dev/api/interactions
gemini,GET /v1beta/interactions/{id},x-goog-api-key,header,string,true,,,,,*,,DOCUMENTED|BETA,Gemini API key. Never put the key in the URL.,https://ai.google.dev/api/interactions
gemini,GET /v1beta/interactions/{id},Api-Revision,header,string,false,,,,2026-05-20|2026-05-07,*,,DOCUMENTED|BETA,Schema revision header. `2026-05-20` = current steps/response_format schema (default since 2026-05-26); `2026-05-07` = legacy `outputs` schema (removed 2026-06-08).,https://ai.google.dev/gemini-api/docs/interactions-breaking-changes-may-2026
gemini,GET /v1beta/interactions/{id},api_version,path,string,true,,,,v1beta|v1,*,,DOCUMENTED|BETA,Which version of the API to use (path segment: v1beta | v1).,https://ai.google.dev/api/interactions
gemini,GET /v1beta/interactions/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,"The unique identifier of the interaction (Interaction.id, e.g. `v1_Chd...`).",https://ai.google.dev/api/interactions
gemini,DELETE /v1beta/interactions/{id},x-goog-api-key,header,string,true,,,,,*,,DOCUMENTED|BETA,Gemini API key. Never put the key in the URL.,https://ai.google.dev/api/interactions
gemini,DELETE /v1beta/interactions/{id},Api-Revision,header,string,false,,,,2026-05-20|2026-05-07,*,,DOCUMENTED|BETA,Schema revision header. `2026-05-20` = current steps/response_format schema (default since 2026-05-26); `2026-05-07` = legacy `outputs` schema (removed 2026-06-08).,https://ai.google.dev/gemini-api/docs/interactions-breaking-changes-may-2026
gemini,DELETE /v1beta/interactions/{id},api_version,path,string,true,,,,v1beta|v1,*,,DOCUMENTED|BETA,Which version of the API to use (path segment: v1beta | v1).,https://ai.google.dev/api/interactions
gemini,DELETE /v1beta/interactions/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,"The unique identifier of the interaction (Interaction.id, e.g. `v1_Chd...`).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions/{id}/cancel,x-goog-api-key,header,string,true,,,,,*,,DOCUMENTED|BETA,Gemini API key. Never put the key in the URL.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions/{id}/cancel,Api-Revision,header,string,false,,,,2026-05-20|2026-05-07,*,,DOCUMENTED|BETA,Schema revision header. `2026-05-20` = current steps/response_format schema (default since 2026-05-26); `2026-05-07` = legacy `outputs` schema (removed 2026-06-08).,https://ai.google.dev/gemini-api/docs/interactions-breaking-changes-may-2026
gemini,POST /v1beta/interactions/{id}/cancel,api_version,path,string,true,,,,v1beta|v1,*,,DOCUMENTED|BETA,Which version of the API to use (path segment: v1beta | v1).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions/{id}/cancel,id,path,string,true,,,,,*,,DOCUMENTED|BETA,"The unique identifier of the interaction (Interaction.id, e.g. `v1_Chd...`).",https://ai.google.dev/api/interactions
gemini,GET /v1beta/interactions/{id},stream,query,boolean,false,false,,,,*,,DOCUMENTED|BETA,"If true, the interaction is streamed incrementally as SSE (reconnect to a background interaction).",https://ai.google.dev/api/interactions
gemini,GET /v1beta/interactions/{id},last_event_id,query,string,false,,,,,*,,DOCUMENTED|BETA,Resumes the interaction stream from the next chunk after the event with this event_id. Only with stream=true.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,model,body,string (ModelOption),one of model|agent,,,,gemini-2.5-flash|gemini-2.5-pro|gemma-4-26b-a4b-it|gemma-4-31b-it|gemini-flash-latest|gemini-flash-lite-latest|gemini-pro-latest|gemini-2.5-flash-lite|gemini-2.5-flash-image|gemini-3-flash-preview|gemini-3.1-pro-preview|gemini-3.1-pro-preview-customtools|gemini-3.1-flash-lite|gemini-3-pro-image|nano-banana-pro-preview|gemini-3.1-flash-image|gemini-3.5-flash|gemini-3.6-flash|gemini-3.7-flash|gemini-3.8-flash|lyria-3-clip-preview|lyria-3-pro-preview|gemini-robotics-er-1.6-preview|gemini-robotics-er-2-preview,*,,DOCUMENTED|BETA,Name of the Model generating the interaction. Required if `agent` is not provided. Mutually exclusive with `agent`.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,agent,body,string (AgentOption),one of model|agent,,,,deep-research-pro-preview-12-2025|deep-research-preview-04-2026|deep-research-max-preview-04-2026|antigravity-preview-05-2026|antigravity-preview-09-2026|<custom agent id>,deep-research-pro-preview-12-2025|deep-research-preview-04-2026|deep-research-max-preview-04-2026|antigravity-preview-05-2026|antigravity-preview-09-2026|<custom agent id>,,DOCUMENTED|BETA,Name of the Agent (managed agent id or custom agent id created via POST /v1beta/agents). Required if `model` is not provided.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input,body,string | Content | Content[] | Step[],true,,,,,*,,DOCUMENTED|BETA,"Inputs for the interaction (common to model and agent). A plain string, a single Content block, an array of Content blocks (single user turn) or an array of Steps (full stateless history: user_input/model_output/function_call/function_result...).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[].type,body,string,false,,,,text|image|audio|document|video|user_input|model_output|function_call|function_result|thought|code_execution_call|code_execution_result|google_search_call|google_search_result|google_maps_call|google_maps_result|url_context_call|url_context_result|file_search_call|file_search_result|mcp_server_tool_call|mcp_server_tool_result|processing_call|processing_result,*,,DOCUMENTED|BETA,"Discriminator of each input item. Content types: text, image, audio, document, video. Step types (history/tool results): user_input, model_output, function_call, function_result, thought, *_call/*_result server-tool steps.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](text).text,body,string,true,,,,,*,,DOCUMENTED|BETA,Required. The text content.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](text).annotations[],body,Annotation[],false,,,,,*,,DOCUMENTED|BETA,Citation annotations (file_citation | place_citation | url_citation | word_info). Output-side mostly.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](image).data,body,string (base64),false,,,,,*,,DOCUMENTED|BETA,Inline image bytes (base64). Alternative to `uri`.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](image).uri,body,string,false,,,,,*,,DOCUMENTED|BETA,URI of the image (e.g. Files API URI).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](image).mime_type,body,string,false,,,,image/png|image/jpeg|image/webp|image/heic|image/heif|image/gif|image/bmp|image/tiff,*,,DOCUMENTED|BETA,Image MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](image).resolution,body,string (MediaResolution),false,,,,low|medium|high|ultra_high,*,,DOCUMENTED|BETA,Media resolution hint.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](audio).data,body,string (base64),false,,,,,*,,DOCUMENTED|BETA,Inline audio bytes.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](audio).uri,body,string,false,,,,,*,,DOCUMENTED|BETA,URI of the audio.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](audio).mime_type,body,string,false,,,,audio/wav|audio/mp3|audio/aiff|audio/aac|audio/ogg|audio/flac|audio/mpeg|audio/m4a|audio/l16|audio/opus|audio/alaw|audio/mulaw|audio/webm,*,,DOCUMENTED|BETA,Audio MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](audio).channels,body,integer,false,,,,,*,,DOCUMENTED|BETA,Number of audio channels.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](audio).sample_rate,body,integer,false,,,,,*,,DOCUMENTED|BETA,Sample rate of the audio.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](document).data,body,string (base64),false,,,,,*,,DOCUMENTED|BETA,Inline document bytes.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](document).uri,body,string,false,,,,,*,,DOCUMENTED|BETA,URI of the document.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](document).mime_type,body,string,false,,,,application/pdf|text/csv,*,,DOCUMENTED|BETA,Document MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).data,body,string (base64),false,,,,,*,,DOCUMENTED|BETA,Inline video bytes.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).uri,body,string,false,,,,,*,,DOCUMENTED|BETA,URI of the video (Files API URI or YouTube URL).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).mime_type,body,string,false,,,,video/mp4|video/mpeg|video/mpg|video/mov|video/avi|video/x-flv|video/webm|video/wmv|video/3gpp,*,,DOCUMENTED|BETA,Video MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).name,body,string,false,,,,,*,,DOCUMENTED|BETA,User-defined name for this content block; the model can reference it in the final response.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).resolution,body,string (MediaResolution),false,,,,low|medium|high|ultra_high,*,,DOCUMENTED|BETA,Media resolution.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).processing,body,string | StaticMediaProcessing,false,,,,static|agentic,*,,DOCUMENTED|BETA,"How the model processes the video: `""agentic""` (model-driven dynamic navigation) or `""static""` (fixed-rate frame extraction) or an object {type:'static', fps, start_offset, end_offset}.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).processing.type,body,string,false,,,,static,*,,DOCUMENTED|BETA,Always `static` for the object form.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).processing.fps,body,number,false,,,,,*,,DOCUMENTED|BETA,Optional. Frame-rate sampling density.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).processing.start_offset,body,string,false,,,,,*,,DOCUMENTED|BETA,"Optional segment start, decimal seconds with `s` suffix (e.g. ""10.5s""). Non-negative.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](video).processing.end_offset,body,string,false,,,,,*,,DOCUMENTED|BETA,"Optional segment end, e.g. ""30s""; must be > start_offset.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](user_input).content,body,Content[],false,,,,,*,,DOCUMENTED|BETA,Content blocks of a user turn (stateless history).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](model_output).content,body,Content[],false,,,,,*,,DOCUMENTED|BETA,Content blocks of a previous model turn (stateless history).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_call).id,body,string,true,,,,,*,,DOCUMENTED|BETA,Required. Tool call id (echo back from the model's function_call step).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_call).name,body,string,true,,,,,*,,DOCUMENTED|BETA,Required. Function name.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_call).arguments,body,object,true,,,,,*,,DOCUMENTED|BETA,Required. Arguments object.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_result).call_id,body,string,true,,,,,*,,DOCUMENTED|BETA,Required. Must match the `id` of the function_call step being answered.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_result).name,body,string,false,,,,,*,,DOCUMENTED|BETA,Name of the tool that was called.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_result).result,body,string | object | (TextContent|ImageContent)[],true,,,,,*,,DOCUMENTED|BETA,"Required. Result of the tool call (plain string, JSON object or rich content array).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](function_result).is_error,body,boolean,false,,,,,*,,DOCUMENTED|BETA,Whether the tool call resulted in an error.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](thought).signature,body,string,false,,,,,*,,DOCUMENTED|BETA,Thought signature to replay (stateless multi-turn with thinking models).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,input[](thought).summary,body,(TextContent|ImageContent)[],false,,,,,*,,DOCUMENTED|BETA,Thought summary content.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,system_instruction,body,string,false,,,,,*,,DOCUMENTED|BETA,System instruction for this interaction (interaction-scoped: must be re-sent every turn; additive with AGENTS.md for agents).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools,body,Tool[],false,,,,,*,,DOCUMENTED|BETA,"Tool declarations the model/agent may call. Interaction-scoped (re-send each turn). Antigravity default set: code_execution, google_search, url_context; Deep Research default: google_search, url_context, code_execution.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[].type,body,string,false,,,,function|google_search|google_maps|code_execution|url_context|file_search|mcp_server|computer_use,*,,DOCUMENTED|BETA,Tool discriminator.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](function).name,body,string,false,,,,,*,,DOCUMENTED|BETA,Function name.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](function).description,body,string,false,,,,,*,,DOCUMENTED|BETA,Function description.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](function).parameters,body,object (JSON Schema),false,,,,,*,,DOCUMENTED|BETA,JSON Schema of the function parameters.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](google_search).search_types,body,string[],false,,,,web_search|image_search,*,,DOCUMENTED|BETA,Types of search grounding to enable.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](google_maps).latitude,body,number,false,,,,,*,,DOCUMENTED|BETA,User latitude.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](google_maps).longitude,body,number,false,,,,,*,,DOCUMENTED|BETA,User longitude.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](google_maps).enable_widget,body,boolean,false,,,,,*,,DOCUMENTED|BETA,Return a widget context token in the tool result.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](file_search).file_search_store_names,body,string[],false,,,,,*,,DOCUMENTED|BETA,File search store names (`fileSearchStores/...`).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](file_search).metadata_filter,body,string,false,,,,,*,,DOCUMENTED|BETA,Metadata filter for semantic retrieval.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](file_search).top_k,body,integer,false,,,,,*,,DOCUMENTED|BETA,Number of retrieval chunks.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).name,body,string,false,,,,,*,,DOCUMENTED|BETA,"MCP server name (Antigravity: required, must match ^[a-z0-9_-]+$; uppercase → 400).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).url,body,string,false,,,,,*,,DOCUMENTED|BETA,Full MCP endpoint URL (streamable HTTP; SSE transport unsupported for Antigravity).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).headers,body,object,false,,,,,*,,DOCUMENTED|BETA,Optional auth headers etc. Merged on top of `credential`.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).credential,body,string,false,,,,,*,,DOCUMENTED|PREVIEW,Stored credential id injected by the egress proxy (agent-credentials guide; not in the reference page).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/interactions,tools[](mcp_server).allowed_tools,body,AllowedTools[] | string[],false,,,,,*,,DOCUMENTED|BETA,"Restrict callable tools. Reference: array of {mode, tools[]}; guides: list of tool names.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).allowed_tools[].mode,body,string,false,,,,auto|any|none|validated,*,,DOCUMENTED|BETA,Tool choice mode.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](mcp_server).allowed_tools[].tools,body,string[],false,,,,,*,,DOCUMENTED|BETA,Names of the allowed tools.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](computer_use).environment,body,string,false,,,,browser|mobile|desktop,gemini-2.5-computer-use-preview-10-2025,,DOCUMENTED|BETA,Environment being operated.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](computer_use).disabled_safety_policies,body,string[],false,,,,financial_transactions|sensitive_data_modification|communication_tool|account_creation|data_modification|user_consent_management|legal_terms_and_agreements,gemini-2.5-computer-use-preview-10-2025,,DOCUMENTED|BETA,Disabled safety policies.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](computer_use).enable_prompt_injection_detection,body,boolean,false,,,,,gemini-2.5-computer-use-preview-10-2025,,DOCUMENTED|BETA,Enable prompt-injection detection on computer-use requests.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,tools[](computer_use).excluded_predefined_functions,body,string[],false,,,,,gemini-2.5-computer-use-preview-10-2025,,DOCUMENTED|BETA,Predefined functions excluded from the model call.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format,body,ResponseFormat | ResponseFormat[],false,,,,,*,,DOCUMENTED|BETA,Output format(s). Polymorphic on `type` (text | image | audio | video); pass an array for multiple output modalities. Replaces legacy `response_mime_type` and `generation_config.image_config` (May 2026 breaking change). Not supported by Deep Research / Antigravity agents.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format.type,body,string,false,,,,text|image|audio|video,*,,DOCUMENTED|BETA,Format discriminator.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(text).mime_type,body,string,false,,,,application/json|text/plain,*,,DOCUMENTED|BETA,Text output MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(text).schema,body,object (JSON Schema),false,,,,,*,,DOCUMENTED|BETA,JSON schema the output must conform to (only with mime_type application/json).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(image).aspect_ratio,body,string,false,,,,1:1|2:3|3:2|3:4|4:3|4:5|5:4|9:16|16:9|21:9|1:8|8:1|1:4|4:1,*,,DOCUMENTED|BETA,Image aspect ratio.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(image).image_size,body,string,false,,,,512|1K|2K|4K,*,,DOCUMENTED|BETA,Image size.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(image).mime_type,body,string,false,,,,image/jpeg,*,,DOCUMENTED|BETA,Image output MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(image).delivery,body,string,false,,,,inline|uri,*,,DOCUMENTED|BETA,Inline base64 or URI.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(audio).mime_type,body,string,false,,,,audio/mp3|audio/ogg_opus|audio/l16|audio/wav|audio/alaw|audio/mulaw,*,,DOCUMENTED|BETA,Audio output MIME type.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(audio).sample_rate,body,integer,false,,,,,*,,DOCUMENTED|BETA,Sample rate in Hz.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(audio).bit_rate,body,integer,false,,,,,*,,DOCUMENTED|BETA,"Bit rate (bps), compressed formats only.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(audio).delivery,body,string,false,,,,inline|uri,*,,DOCUMENTED|BETA,Inline or URI.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(video).aspect_ratio,body,string,false,,,,16:9|9:16,*,,DOCUMENTED|BETA,Video aspect ratio.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(video).resolution,body,string,false,720p,,,360p|720p|1080p|4k,*,,DOCUMENTED|BETA,Video resolution. Defaults to 720p.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(video).duration,body,string,false,,,,,*,,DOCUMENTED|BETA,Video duration.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,response_format(video).delivery,body,string,false,,,,inline|uri,*,,DOCUMENTED|BETA,Inline or URI.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,stream,body,boolean,false,false,,,,*,,DOCUMENTED|BETA,Input only. Stream the response as SSE (InteractionSseEvent).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,store,body,boolean,false,true,,,,*,,DOCUMENTED|BETA,"Input only. Store request+response for retrieval / previous_interaction_id. Default true. Retention: paid tier 55 days (configurable 7/14/28/55 in AI Studio), free tier 1 day. `store=false` is incompatible with `background=true` and prevents chaining via previous_interaction_id.",https://ai.google.dev/gemini-api/docs/interactions-overview
gemini,POST /v1beta/interactions,background,body,boolean,false,false,,,,*,,DOCUMENTED|BETA,Input only. Run asynchronously; returns immediately with status queued/in_progress; poll GET or stream GET ?stream=true. Requires store=true. Mandatory in practice for deep-research-* agents.,https://ai.google.dev/gemini-api/docs/background-execution
gemini,POST /v1beta/interactions,labels,body,"object (map<string,string>)",false,,,,,*,,DOCUMENTED|BETA,User-defined metadata labels for the request.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,safety_settings,body,SafetySetting[],false,,,,,*,,DOCUMENTED|BETA|UNVERIFIED,Safety settings. NOTE: the overview says custom safety settings are NOT supported in the Interactions API.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,service_tier,body,string (ServiceTier),false,standard,,,flex|standard|priority,*,,DOCUMENTED|BETA,"Inference tier: flex (50% price, best-effort, 429/503 when full, no fallback), standard, priority (+75-100% price, 0.3x rate limit, graceful downgrade; check `x-gemini-service-tier` response header).",https://ai.google.dev/gemini-api/docs/flex-inference
gemini,POST /v1beta/interactions,webhook_config,body,WebhookConfig,false,,,,,*,,DOCUMENTED|BETA,Dynamic webhook for completion events (interaction.completed/failed/cancelled/requires_action) signed with JWKS asymmetric keys.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1beta/interactions,webhook_config.uris,body,string[],false,,,,,*,,DOCUMENTED|BETA,Webhook URIs used instead of the registered static webhooks.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1beta/interactions,webhook_config.user_metadata,body,object,false,,,,,*,,DOCUMENTED|BETA,Metadata returned on each webhook event emission.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1beta/interactions,generation_config,body,GenerationConfig,false,,,,,*,,DOCUMENTED|BETA,Model configuration. Only when `model` is set (alternative to agent_config). Interaction-scoped.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.max_output_tokens,body,integer,false,,,,,*,,DOCUMENTED|BETA,Maximum tokens in the response. Not supported by Antigravity (400).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.seed,body,integer,false,,,,,*,,DOCUMENTED|BETA,Decoding seed for reproducibility.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.stop_sequences,body,string[],false,,,,,*,,DOCUMENTED|BETA,Stop sequences. Not supported by Antigravity (400).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.temperature,body,number,false,,,,,*,,DOCUMENTED|BETA|UNVERIFIED,Sampling temperature (shown in the text-generation guide `generation_config: {temperature: 1.0}`; ABSENT from the reference page's GenerationConfig list). Unsupported by Antigravity (400).,https://ai.google.dev/gemini-api/docs/text-generation
gemini,POST /v1beta/interactions,generation_config.top_p,body,number,false,,,,,*,,DOCUMENTED|UNVERIFIED,"Nucleus sampling (mentioned in breaking-changes and antigravity limitations, not in the reference page).",https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/interactions,generation_config.top_k,body,integer,false,,,,,*,,DOCUMENTED|UNVERIFIED,Top-k sampling (mentioned in antigravity limitations only).,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/interactions,generation_config.thinking_level,body,string (ThinkingLevel),false,,,,minimal|low|medium|high,*,,DOCUMENTED|BETA,Amount of thinking (Gemini 3 family).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.thinking_summaries,body,string (ThinkingSummaries),false,,,,auto|none,*,,DOCUMENTED|BETA,Include thought summaries (`thought` steps with `summary`).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.tool_choice,body,string | ToolChoiceConfig,false,,,,auto|any|none|validated,*,,DOCUMENTED|BETA,"Tool choice: enum or {allowed_tools:{mode, tools[]}}.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.tool_choice.allowed_tools.mode,body,string,false,,,,auto|any|none|validated,*,,DOCUMENTED|BETA,Mode for the allowed tools.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.tool_choice.allowed_tools.tools,body,string[],false,,,,,*,,DOCUMENTED|BETA,Names of the allowed tools.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.speech_config,body,SpeakerConfig | SpeechConfig[],false,,,,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED|BETA,Speech / multi-speaker TTS configuration.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.speech_config.speakers[].speaker,body,string,false,,,,,*,,DOCUMENTED|BETA,Speaker name matching the prompt.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.speech_config.speakers[].voice,body,string,false,,,,,*,,DOCUMENTED|BETA,Voice name.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.speech_config.speakers[].language,body,string,false,,,,,*,,DOCUMENTED|BETA,Language of the speech.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config,body,TranscriptionConfig,false,,,,,*,,DOCUMENTED|BETA,Speech recognition (ASR) configuration; presence enables transcription.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.custom_vocabulary,body,string[],false,,,,,*,,DOCUMENTED|BETA,Phrases to bias recognition.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.language_codes,body,string[],false,,,,,*,,DOCUMENTED|BETA,BCP-47 hints; empty = auto-detect.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.mode,body,string | SmartTranscriptionMode | VerbatimTranscriptionMode,false,,,,verbatim|smart,*,,DOCUMENTED|BETA,Transcription mode enum or object.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.mode.type,body,string,false,,,,smart|verbatim,*,,DOCUMENTED|BETA,Object-form discriminator.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.mode(verbatim).diarization_mode,body,string,false,,,,speaker,*,,DOCUMENTED|BETA,"Speaker diarization. Supported: ""speaker"".",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.transcription_config.mode(verbatim).timestamp_granularities,body,string[],false,,,,word,*,,DOCUMENTED|BETA,"Timestamp granularity. Supported: ""word"".",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,generation_config.video_config.task,body,string,false,,,,text_to_video|image_to_video|reference_to_video|edit|extend,veo-*,,DOCUMENTED|BETA,Video generation task mode (auto if omitted).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,agent_config,body,AntigravityAgentConfig | DeepResearchAgentConfig | DynamicAgentConfig | CodeMenderAgentConfig,false,,,,,deep-research-pro-preview-12-2025|deep-research-preview-04-2026|deep-research-max-preview-04-2026|antigravity-preview-05-2026|antigravity-preview-09-2026|<custom agent id>,,DOCUMENTED|BETA,Agent configuration. Only when `agent` is set (alternative to generation_config). Discriminator `type`. CodeMenderAgentConfig is named in the union but has no documented fields.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,agent_config.type,body,string,false,,,,antigravity|deep-research|dynamic,deep-research-pro-preview-12-2025|deep-research-preview-04-2026|deep-research-max-preview-04-2026|antigravity-preview-05-2026|antigravity-preview-09-2026|<custom agent id>,,DOCUMENTED|BETA,Agent config discriminator.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,agent_config(antigravity).model,body,string,false,gemini-3.8-flash,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite,antigravity-preview-09-2026|antigravity-preview-05-2026,,DOCUMENTED|BETA,Underlying Gemini model for the Antigravity agent (default gemini-3.8-flash). Locked for named custom agents.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/interactions,agent_config(antigravity).max_total_tokens,body,string | integer,false,,,,,antigravity-preview-09-2026|antigravity-preview-05-2026,,DOCUMENTED|BETA,"Budget: max total (input+output+thinking, excluding cached) tokens for the run; best-effort; exceeding → status `incomplete` (continue with previous_interaction_id + environment). Reference types it as string, guides pass an integer (50000).",https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/interactions,agent_config(deep-research).collaborative_planning,body,boolean,false,,,,,deep-research-preview-04-2026|deep-research-max-preview-04-2026|deep-research-pro-preview-12-2025,,DOCUMENTED|BETA,Human-in-the-loop planning: agent returns a research plan; proceeds only after confirmation in the next turn (set false/omit to approve).,https://ai.google.dev/gemini-api/docs/deep-research
gemini,POST /v1beta/interactions,agent_config(deep-research).thinking_summaries,body,string,false,,,,auto|none,deep-research-preview-04-2026|deep-research-max-preview-04-2026|deep-research-pro-preview-12-2025,,DOCUMENTED|BETA,Include thought summaries.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,agent_config(deep-research).visualization,body,string,false,,,,off|auto,deep-research-preview-04-2026|deep-research-max-preview-04-2026,,DOCUMENTED|BETA,Allow charts/graphs (image deltas) in the report; agent only generates them when prompted.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment,body,string | EnvironmentConfig,false,,,,,antigravity-preview-09-2026|antigravity-preview-05-2026|<custom agent id>,,DOCUMENTED|BETA,"Sandbox for agent interactions: `""remote""` (fresh sandbox), an existing environment id (reuse files/packages), or an EnvironmentConfig object. Returned as `environment_id` on the Interaction. Required alongside previous_interaction_id when chaining managed agents.",https://ai.google.dev/gemini-api/docs/agent-environment
gemini,POST /v1beta/interactions,environment.type,body,string,true,,,,remote,*,,DOCUMENTED|BETA,Always `remote`.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.environment_id,body,string,false,,,,,*,,DOCUMENTED|BETA,Update an existing environment (e.g. refresh network rules) instead of creating one.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[],body,Source[],false,,,,,*,,DOCUMENTED|BETA,"Files to mount at creation. Limits: repository 500 MB, gcs 2 GB, inline 1 MB/file and 2 MB total. Target must not be `/`.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[].type,body,string,false,,,,gcs|inline|repository,*,,DOCUMENTED|BETA,Source kind.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[].source,body,string,false,,,,,*,,DOCUMENTED|BETA,"GCS path (gs://...) or repository URL (github://, https://github.com/...).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[].target,body,string,false,,,,,*,,DOCUMENTED|BETA,"Where the source appears in the sandbox (e.g. /workspace/repo, .agents/AGENTS.md, .agents/skills/<name>/SKILL.md, .agents/hooks.json).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[].content,body,string,false,,,,,*,,DOCUMENTED|BETA,Inline content when type=inline.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.sources[].encoding,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional encoding for inline content (e.g. base64).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.network,body,string | EnvironmentNetworkEgressAllowlist,false,,,,disabled,*,,DOCUMENTED|BETA,"Outbound network: omit = unrestricted; `""disabled""` = no egress; object {allowlist[]} = only listed domains (add {domain:'*'} as catch-all). New rules fully replace previous ones for a reused environment.",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.network.allowlist[].domain,body,string,false,,,,,*,,DOCUMENTED|BETA,Exact host or wildcard (`*.example.com` does NOT match the root; `*` = all).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.network.allowlist[].credential,body,string,false,,,,,*,,DOCUMENTED|BETA,Stored credential id; the egress proxy injects its header on requests to this domain.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.network.allowlist[].transform,body,object | object[],false,,,,,*,,DOCUMENTED|BETA,Headers to inject on matching requests (e.g. {Authorization: 'Bearer ...'}); merged over `credential` (transform wins on conflict).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.env,body,"object (map<string, string | {credential}>)",false,,,,,*,,DOCUMENTED|BETA,"Environment variables in the sandbox: literal strings (plaintext, readable by the agent) or `{""credential"": ""<id>""}` references (placeholder `__GEMINI_CRED_<id>__`, substituted by the proxy only towards the credential's trusted_domains).",https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.env.<NAME>.credential,body,string,false,,,,,*,,DOCUMENTED|BETA,Reference to an environment_variable credential id.,https://ai.google.dev/api/interactions
gemini,POST /v1beta/interactions,environment.env.<NAME>.value,body,string,false,,,,,*,,DOCUMENTED|BETA,Direct string value (reference-page object form).,https://ai.google.dev/api/interactions
gemini,POST /v1beta/credentials,id,body,string,false,,,,,*,,DOCUMENTED|BETA,Unique credential id (UUID generated if omitted; lowercase alphanumerics/hyphens/underscores recommended). 409 `aborted` if it exists.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,type,body,string,true,,,,bearer_token|oauth2|environment_variable,*,,DOCUMENTED|BETA,Credential type (fixed at creation).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,token,body,string,bearer_token,,,,,*,,DOCUMENTED|BETA,bearer_token: the static token. Write-only.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,header_name,body,string,false,Authorization,,,,*,,DOCUMENTED|BETA,bearer_token: header to inject. Default Authorization.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,prefix,body,string,false,Bearer,,,,*,,DOCUMENTED|BETA,"bearer_token: value prefix. Default `Bearer`; `""""` for none.",https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,client_id,body,string,oauth2,,,,,*,,DOCUMENTED|BETA,oauth2: client id.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,client_secret,body,string,oauth2,,,,,*,,DOCUMENTED|BETA,oauth2: client secret. Write-only.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,refresh_token,body,string,oauth2,,,,,*,,DOCUMENTED|BETA,oauth2: refresh token (live exchange at creation). Write-only.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,token_url,body,string,oauth2,,,,,*,,DOCUMENTED|BETA,oauth2: token endpoint URL.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,scopes,body,string[],false,,,,,*,,DOCUMENTED|BETA,oauth2: scopes.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,value,body,string,environment_variable,,,,,*,,DOCUMENTED|BETA,environment_variable: the secret value. Write-only.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,injection_location,body,string | string[],environment_variable,,,,header|query|body,*,,DOCUMENTED|BETA,environment_variable: where the proxy substitutes the placeholder.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/credentials,trusted_domains,body,string[],false,,,,,*,,DOCUMENTED|BETA,environment_variable: domain patterns authorized for substitution.,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,GET /v1beta/credentials,page_size,query,integer,false,50,,1000,,*,,DOCUMENTED|BETA,"Max credentials to return. Default 50, max 1000.",https://ai.google.dev/api/interactions
gemini,GET /v1beta/credentials,page_token,query,string,false,,,,,*,,DOCUMENTED|BETA,Pagination token (next_page_token of previous response).,https://ai.google.dev/api/interactions
gemini,GET /v1beta/credentials/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Credential id.,https://ai.google.dev/api/interactions
gemini,PATCH /v1beta/credentials/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Credential id.,https://ai.google.dev/api/interactions
gemini,DELETE /v1beta/credentials/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Credential id.,https://ai.google.dev/api/interactions
gemini,PATCH /v1beta/credentials/{id},update_mask,query,string,false,,,,,*,,DOCUMENTED|BETA,Optional. List of fields to update.,https://ai.google.dev/api/interactions
gemini,PATCH /v1beta/credentials/{id},type,body,string,true,,,,bearer_token|oauth2|environment_variable,*,,DOCUMENTED|BETA,Must match the existing type (cannot change).,https://ai.google.dev/api/interactions
gemini,PATCH /v1beta/credentials/{id},token,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `token` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},header_name,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `header_name` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},prefix,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `prefix` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},client_id,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `client_id` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},client_secret,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `client_secret` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},refresh_token,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `refresh_token` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},token_url,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `token_url` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},scopes,body,string[],false,,,,,*,,DOCUMENTED|BETA,Optional new value for `scopes` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},value,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional new value for `value` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},injection_location,body,string | string[],false,,,,,*,,DOCUMENTED|BETA,Optional new value for `injection_location` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,PATCH /v1beta/credentials/{id},trusted_domains,body,string[],false,,,,,*,,DOCUMENTED|BETA,Optional new value for `trusted_domains` (same semantics as create).,https://ai.google.dev/gemini-api/docs/agent-credentials
gemini,POST /v1beta/agents,id,body,string,true,,,,,*,,DOCUMENTED|BETA,"Unique agent id in the project; used as `agent` in interactions. Reserved prefixes forbidden (antigravity-, veo-, omni-, lyria-, imagen-, gemma-, gemini-, google-, youtube-, android-, chrome-, pixel-, waze-, fitbit-, nest-, kaggle-).",https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,description,body,string,false,,,,,*,,DOCUMENTED|BETA,Human-readable description.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,base_agent,body,string,true,,,,antigravity-preview-09-2026|antigravity-preview-05-2026,*,,DOCUMENTED|BETA,Base managed agent. Guide: only `antigravity-preview-09-2026`; reference example: `antigravity-preview-05-2026`.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,agent_config,body,AntigravityAgentConfig,false,,,,,*,,DOCUMENTED|BETA,Base agent configuration incl. model. Default gemini-3.8-flash. Cannot be overridden at interaction time for named agents.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,agent_config.type,body,string,false,,,,antigravity,*,,DOCUMENTED|BETA,`antigravity`.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,agent_config.model,body,string,false,gemini-3.8-flash,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite,*,,DOCUMENTED|BETA,Model for the agent.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,system_instruction,body,string,false,,,,,*,,DOCUMENTED|BETA,System prompt / persona (overridable per interaction).,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,tools,body,Tool[],false,,,,,*,,DOCUMENTED|BETA,"Default tools (code_execution, google_search, url_context if omitted); supports mcp_server and function (overridable per interaction).",https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/agents,base_environment,body,string | EnvironmentConfig,false,,,,,*,,DOCUMENTED|BETA,"`""remote""`, an environment_id (fork), or {type:'remote', sources[], network}. Each invocation forks it.",https://ai.google.dev/gemini-api/docs/custom-agents
gemini,GET /v1beta/agents,page_size,query,integer,false,,,,,*,,DOCUMENTED|UNVERIFIED,Page size (SDK ListAgentsParams).,https://ai.google.dev/api/interactions
gemini,GET /v1beta/agents,page_token,query,string,false,,,,,*,,DOCUMENTED|UNVERIFIED,Page token (SDK ListAgentsParams).,https://ai.google.dev/api/interactions
gemini,GET /v1beta/agents/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Agent id.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,DELETE /v1beta/agents/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Agent id.,https://ai.google.dev/gemini-api/docs/custom-agents
gemini,POST /v1beta/triggers,schedule,body,string (cron),true,,,,,*,,DOCUMENTED|BETA,"Cron expression, e.g. `0 9 * * 1-5`.",https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers,time_zone,body,string (IANA),true,,,,,*,,DOCUMENTED|BETA,IANA time zone.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers,display_name,body,string,false,,,,,*,,DOCUMENTED|BETA,Human-readable name.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers,max_consecutive_failures,body,integer,false,5,,,,*,,DOCUMENTED|BETA,Failures before auto-pause. Default 5.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers,execution_timeout_seconds,body,integer,false,600,,,,*,,DOCUMENTED|BETA,Timeout per execution. Default 600.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers,interaction,body,CreateInteractionRequest,true,,,,,*,,DOCUMENTED|BETA,"The agent interaction to run (agent, input, tools, environment ...).",https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,PATCH /v1beta/triggers/{id},status,body,string,false,,,,paused|active,*,,DOCUMENTED|BETA,Pause (`paused`) or resume (`active`).,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,GET /v1beta/triggers/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Trigger id.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,PATCH /v1beta/triggers/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Trigger id.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,DELETE /v1beta/triggers/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Trigger id.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1beta/triggers/{id}/executions,id,path,string,true,,,,,*,,DOCUMENTED|BETA,Trigger id.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,GET /v1beta/triggers/{id}/executions,id,path,string,true,,,,,*,,DOCUMENTED|BETA,Trigger id.,https://ai.google.dev/gemini-api/docs/antigravity-agent
gemini,POST /v1/webhooks,name,body,string,true,,,,,*,,DOCUMENTED|BETA,Webhook display name.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1/webhooks,uri,body,string,true,,,,,*,,DOCUMENTED|BETA,HTTPS listener URL.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1/webhooks,subscribed_events,body,string[],true,,,,interaction.requires_action|interaction.completed|interaction.failed|interaction.cancelled|batch.succeeded|batch.cancelled|batch.expired|batch.failed|video.generated,*,,DOCUMENTED|BETA,Events to deliver.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,PATCH /v1/webhooks/{id},subscribed_events,body,string[],false,,,,,*,,DOCUMENTED|BETA,Replace subscribed events.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1/webhooks/{id}/rotate_secret,revocation_behavior,body,string,false,,,,REVOKE_PREVIOUS_SECRETS_AFTER_H24,*,,DOCUMENTED|BETA,When to revoke previous secrets.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,GET /v1/webhooks/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Webhook id.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,PATCH /v1/webhooks/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Webhook id.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,DELETE /v1/webhooks/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Webhook id.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1/webhooks/{id}/rotate_secret,id,path,string,true,,,,,*,,DOCUMENTED|BETA,Webhook id.,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1beta/environments,sources[],body,Source[],false,,,,,*,,DOCUMENTED|BETA,Sources to mount (REST page snake_case; discovery `sources`).,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,sources[].type,body,string,false,,,,gcs|inline|repository,*,,DOCUMENTED|BETA,Source kind.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,sources[].source,body,string,false,,,,,*,,DOCUMENTED|BETA,GCS path or repository URL.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,sources[].target,body,string,false,,,,,*,,DOCUMENTED|BETA,Mount target inside the environment.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,sources[].content,body,string,false,,,,,*,,DOCUMENTED|BETA,Inline content (type=inline).,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,sources[].encoding,body,string,false,,,,,*,,DOCUMENTED|BETA,Optional inline encoding (base64).,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,network,body,string | EnvironmentNetworkEgressAllowlist,false,,,,disabled,*,,DOCUMENTED|BETA,REST page: `disabled` or {allowlist[]}. Discovery doc splits this into `networkMode` (NETWORK_MODE_UNSPECIFIED | DISABLED) and `networkAllowlist`.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,network.allowlist[].domain,body,string,false,,,,,*,,DOCUMENTED|BETA,"Allowed domain (wildcards, `*`).",https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,network.allowlist[].credential,body,string,false,,,,,*,,DOCUMENTED|BETA,Stored credential id.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,network.allowlist[].transform,body,object | object[],false,,,,,*,,DOCUMENTED|BETA,Headers to inject.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,from_environment,body,string,false,,,,,*,,DOCUMENTED|BETA,Fork/copy from `environments/{id}` or `{id}`; sources/env must be empty. Discovery: `fromEnvironment`.,https://ai.google.dev/api/environments
gemini,POST /v1beta/environments,networkMode,body,string,false,,,,NETWORK_MODE_UNSPECIFIED|DISABLED,*,,DOCUMENTED|UNVERIFIED,Discovery-doc camelCase field: network egress mode.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/environments,networkAllowlist,body,EnvironmentNetworkEgressAllowlist,false,,,,,*,,DOCUMENTED|UNVERIFIED,Discovery-doc camelCase field: {allowlist: EgressRule[]}.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/environments,fromEnvironment,body,string,false,,,,,*,,DOCUMENTED|UNVERIFIED,Discovery-doc camelCase alias of from_environment.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,GET /v1beta/environments,page_size,query,integer,false,50,,1000,,*,,DOCUMENTED|BETA,"Max environments (default 50, max 1000). Discovery/guide: `pageSize`.",https://ai.google.dev/api/environments
gemini,GET /v1beta/environments,page_token,query,string,false,,,,,*,,DOCUMENTED|BETA,Pagination token. Discovery: `pageToken`.,https://ai.google.dev/api/environments
gemini,GET /v1beta/environments/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Environment id (discovery: resource ID segment of `environments/{id}`).,https://ai.google.dev/api/environments
gemini,DELETE /v1beta/environments/{id},id,path,string,true,,,,,*,,DOCUMENTED|BETA,Environment id (discovery: resource ID segment of `environments/{id}`).,https://ai.google.dev/api/environments
gemini,GET /v1beta/environments/{environment}/files/{path},environment,path,string,true,,,,,*,,DOCUMENTED|BETA,Environment id (`parent` = environments/{environment_id} in discovery).,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,GET /v1beta/environments/{environment}/files/{path},path,path,string,false,,,,,*,,DOCUMENTED|BETA,File or directory path relative to the workspace; empty = root.,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,GET /v1beta/environments/{environment}/files/{path},alt,query,string,false,,,,media,*,,DOCUMENTED|BETA,`media` → raw file bytes (file) or POSIX tar (directory).,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,GET /v1beta/environments/{environment}/files/{path},recursive,query,boolean,false,false,,,,*,,DOCUMENTED|BETA,Recurse into subdirectories (listing or tar).,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,GET /v1beta/environments/{environment}/files/{path},page_size,query,integer,false,100,,1000,,*,,DOCUMENTED|BETA,"Directory listing page size. Default 100, max 1000.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,GET /v1beta/environments/{environment}/files/{path},page_token,query,string,false,,,,,*,,DOCUMENTED|BETA,Directory listing page token.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},environment,path,string,true,,,,,*,,DOCUMENTED|BETA,Environment id.,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},path,path,string,true,,,,,*,,DOCUMENTED|BETA,Destination relative path (e.g. workspace/data/file.txt; directory with trailing `/` for extract).,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},extract,query,boolean,false,false,,,,*,,DOCUMENTED|BETA,Treat the body as tar/tar.gz and unpack into `path`.,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},overwrite,query,boolean,false,false,,,,*,,DOCUMENTED|BETA,Overwrite existing file(s); otherwise 409 Conflict.,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},Content-Type,header,string,true,,,,,*,,DOCUMENTED|BETA,MIME type of the uploaded bytes.,https://ai.google.dev/gemini-api/docs/agent-environment
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},<body>,file,bytes,true,,,2147483648 bytes,,*,,DOCUMENTED|BETA,Raw file or archive bytes (`--data-binary @file`). Max 2 GiB (discovery mediaUpload.maxSize).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},X-Goog-Upload-Protocol,header,string,false,,,,resumable,*,,DOCUMENTED|BETA,Resumable protocol selector (`resumable`).,https://ai.google.dev/api/environments
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},X-Goog-Upload-Command,header,string,false,,,,start,*,,DOCUMENTED|BETA,`start` to open a resumable session (URL returned in X-Goog-Upload-URL).,https://ai.google.dev/api/environments
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},X-Goog-Upload-Header-Content-Length,header,integer,false,,,,,*,,DOCUMENTED|BETA,Total bytes to be uploaded to the session URL.,https://ai.google.dev/api/environments
gemini,PUT /upload/v1beta/environments/{environment}/files/{path},X-Goog-Upload-Header-Content-Type,header,string,false,,,,,*,,DOCUMENTED|BETA,MIME type of the bytes uploaded to the session URL.,https://ai.google.dev/api/environments
gemini,GET /v1beta/files/environment-{environment}:download,alt,query,string,true,,,,media,*,,DOCUMENTED|LEGACY,`media` (required to get the tar).,https://ai.google.dev/gemini-api/docs/managed-agents-quickstart
gemini,POST /v1beta/dynamic/{dynamicId}:generateContent,model,path,string,true,,,,,*,,DOCUMENTED|UNVERIFIED|DOCUMENTATION_INCOMPLETE,Resource name `dynamic/{dynamicId}` (pattern ^dynamic/[^/]+$).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/dynamic/{dynamicId}:generateContent,<body>,body,GenerateContentRequest,false,,,,,*,,DOCUMENTED|UNVERIFIED|DOCUMENTATION_INCOMPLETE,"Same body as models/*:generateContent (contents, tools, generationConfig, systemInstruction, safetySettings, cachedContent, toolConfig). Semantics of `dynamic/*` undocumented.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/dynamic/{dynamicId}:streamGenerateContent,model,path,string,true,,,,,*,,DOCUMENTED|UNVERIFIED|DOCUMENTATION_INCOMPLETE,Resource name `dynamic/{dynamicId}` (pattern ^dynamic/[^/]+$).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/dynamic/{dynamicId}:streamGenerateContent,<body>,body,GenerateContentRequest,false,,,,,*,,DOCUMENTED|UNVERIFIED|DOCUMENTATION_INCOMPLETE,"Same body as models/*:generateContent (contents, tools, generationConfig, systemInstruction, safetySettings, cachedContent, toolConfig). Semantics of `dynamic/*` undocumented.",https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/dynamic/{dynamicId}:streamGenerateContent,alt,query,string,false,,,,sse,*,,DOCUMENTED|UNVERIFIED,`sse` for server-sent-events framing (as for models).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,WSS BidiGenerateContent,key,query,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard API key as WebSocket URL query parameter (`?key=YOUR_API_KEY`) — the documented raw-WebSocket auth. Server-side only; never ship in clients. (The project convention prefers the x-goog-api-key header for REST; for WSS the docs show the query param.),"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/get-started-websocket""}"
gemini,WSS BidiGenerateContent,access_token,query,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Ephemeral token (AuthToken.name) as query parameter on the v1beta `BidiGenerateContentConstrained` WebSocket URL (`?access_token=<token>`). Alternative: HTTP header `Authorization: Token <token>`. Ephemeral tokens work only with the Live API, v1beta.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,WSS BidiGenerateContent,Authorization,header,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,`Authorization: Token <ephemeral token>` — header alternative to ?access_token for ephemeral tokens (RFC 7235 auth-scheme `Token`).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup,body,object (BidiGenerateContentSetup),true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,First and only-first client message. Wait for setupComplete before sending more.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.model,body,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,`models/{model}` resource name.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig,body,object (GenerationConfig subset),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Unsupported in Live: responseLogprobs, responseMimeType, logprobs, responseSchema, responseJsonSchema, stopSequence(s), skipResponseCache, routingConfig, audioTimestamp.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.responseModalities[],body,enum[],false,"[""AUDIO""] (SDK)",,,TEXT|AUDIO,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Output modalities. Native audio agent + translate models: [""AUDIO""] only (text via outputAudioTranscription); gemini-3.5-transcribe-live: [""TEXT""]. SDK default AUDIO.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig,body,object (SpeechConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Voice/language for audio output.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig.voiceConfig,body,object (VoiceConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Single-voice configuration.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig.voiceConfig.prebuiltVoiceConfig,body,object (PrebuiltVoiceConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Prebuilt voice selection.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig.voiceConfig.prebuiltVoiceConfig.voiceName,body,string,false,,,,Zephyr|Puck|Charon|Kore|Fenrir|Leda|Orus|Aoede|Callirrhoe|Autonoe|Enceladus|Iapetus|Umbriel|Algieba|Despina|Erinome|Algenib|Rasalgethi|Laomedeia|Achernar|Alnilam|Schedar|Gacrux|Pulcherrima|Achird|Zubenelgenubi|Vindemiatrix|Sadachbia|Sadaltager|Sulafat,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"Any Gemini TTS voice (30): Zephyr, Puck, Charon, Kore, Fenrir, Leda, Orus, Aoede, Callirrhoe, Autonoe, Enceladus, Iapetus, Umbriel, Algieba, Despina, Erinome, Algenib, Rasalgethi, Laomedeia, Achernar, Alnilam, Schedar, Gacrux, Pulcherrima, Achird, Zubenelgenubi, Vindemiatrix, Sadachbia, Sadaltager, Sulafat.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig.languageCode,body,string,false,,,,,,,DOCUMENTED|UNVERIFIED,"BCP-47 output language (de-DE, en-AU, en-GB, en-IN, en-US, es-US, fr-FR, hi-IN, pt-BR, ar-XA, es-ES, fr-CA, id-ID, it-IT, ja-JP, tr-TR, vi-VN, bn-IN, gu-IN, kn-IN, ml-IN, mr-IN, ta-IN, te-IN, nl-NL, ko-KR, cmn-CN, pl-PL, ru-RU, th-TH). Native audio output models pick the language automatically and do not support setting it — effect on current models UNVERIFIED (steer via system instructions).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.speechConfig.multiSpeakerVoiceConfig,body,object (MultiSpeakerVoiceConfig),false,,,,,,,UNVERIFIED,"Multi-speaker TTS config (speakerVoiceConfigs[]{speaker, voiceConfig}); mutually exclusive with voiceConfig. Not documented for Live.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.mediaResolution,body,enum,false,,,,MEDIA_RESOLUTION_UNSPECIFIED|MEDIA_RESOLUTION_LOW|MEDIA_RESOLUTION_MEDIUM|MEDIA_RESOLUTION_HIGH,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"Visual input token allocation per frame (LOW 64, MEDIUM 256, HIGH zoomed 256). Audio tokenization unaffected.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.thinkingConfig,body,object (ThinkingConfig),false,,,,,gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Thinking configuration. gemini-3.8-live: not supported (omit). Error on models without thinking.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/thinking""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.thinkingConfig.thinkingLevel,body,enum,false,MINIMAL (3.1); model-dependent,,,THINKING_LEVEL_UNSPECIFIED|MINIMAL|LOW|MEDIUM|HIGH,gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview,,DOCUMENTED,gemini-3.1-flash-live-preview: MINIMAL (default) | LOW | MEDIUM | HIGH. gemini-3.8-live-extended-thinking: LOW | MEDIUM | HIGH (MINIMAL unsupported). gemini-3.8-live: unsupported.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.thinkingConfig.thinkingBudget,body,integer,false,,,,,gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED|LEGACY,"Thinking token budget (0 disabled, -1 automatic) — Gemini 2.5 native-audio thinking; replaced by thinkingLevel on 3.1+.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-live-preview""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.thinkingConfig.includeThoughts,body,boolean,false,false,,,,gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Include thought summaries in modelTurn parts.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.enableAffectiveDialog,body,boolean,false,false,,,,gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Adapt style/tone to the user's expression. Gemini 2.5 native audio (v1beta) only; not supported on 3.1; removed on 3.8 (remove from code).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.translationConfig,body,object (TranslationConfig),false,,,,,gemini-3.5-live-translate-preview,,DOCUMENTED,Live translation settings (translate model only).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-translate""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.translationConfig.targetLanguageCode,body,string,true,en,,,,gemini-3.5-live-translate-preview,,DOCUMENTED,"Required for translation: BCP-47 target language (""en"", ""es"", ""pl"" ...). Guide default ""en"".","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-translate""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.translationConfig.echoTargetLanguage,body,boolean,false,false,,,,gemini-3.5-live-translate-preview,,DOCUMENTED,true: parrot input already in the target language; false: stay silent for it.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-translate""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.temperature,body,number,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.topP,body,number,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.topK,body,integer,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.maxOutputTokens,body,integer,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.candidateCount,body,integer,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.presencePenalty,body,number,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.frequencyPenalty,body,number,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.generationConfig.seed,body,integer,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard GenerationConfig field listed in the Live session-configuration example (seed: SDK LiveConnectConfig).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://github.com/googleapis/python-genai/blob/main/google/genai/types.py""}"
gemini,WSS BidiGenerateContent,setup.systemInstruction,body,object (Content),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"System instructions; text parts only, each part a separate paragraph. Recommended to specify the language here.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.systemInstruction.parts[].text,body,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Instruction text.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[],body,object[] (Tool),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"Tools: functionDeclarations, googleSearch. codeExecution, urlContext, googleMaps: not supported on current Live models. Translate/transcribe models: no tools.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[],body,object[] (FunctionDeclaration),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"Client-executed functions (name, description, parameters | parametersJsonSchema, behavior, response | responseJsonSchema).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].name,body,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,a-z A-Z 0-9 _ - : . ; max 128.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].description,body,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Function description.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].parameters,body,object (Schema),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"OpenAPI-style parameters schema (type OBJECT, properties, required).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].parametersJsonSchema,body,object,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,JSON Schema alternative (mutually exclusive with parameters).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].behavior,body,enum,false,model-dependent (NON_BLOCKING on 3.8),,,UNSPECIFIED|BLOCKING|NON_BLOCKING,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Live-only. BLOCKING waits for the response; NON_BLOCKING lets the conversation continue. Defaults: gemini-3.8-live NON_BLOCKING (BLOCKING allowed); gemini-3.8-live-extended-thinking NON_BLOCKING only (BLOCKING = hard error); gemini-3.1-flash-live-preview synchronous only; Gemini 2.5 both.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.tools[].functionDeclarations[].response,body,object (Schema),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Response schema (optional).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.tools[].googleSearch,body,object {},false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"Grounding with Google Search; results in serverContent.groundingMetadata. 5,000 free searches/month then $14 / 1,000 (3.x).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,setup.tools[].codeExecution,body,object {},false,,,,,,,DOCUMENTED|UNVERIFIED,Listed as NOT supported for Live models.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,setup.tools[].urlContext,body,object {},false,,,,,,,DOCUMENTED|UNVERIFIED,Listed as NOT supported for Live models.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig,body,object (RealtimeInputConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Realtime input behaviour (VAD, interruption, turn coverage).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection,body,object (AutomaticActivityDetection),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Server-side VAD (enabled by default when absent).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection.disabled,body,boolean,false,false,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,true = manual VAD: client sends realtimeInput.activityStart / activityEnd; audioStreamEnd not used.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection.startOfSpeechSensitivity,body,enum,false,START_SENSITIVITY_HIGH,,,START_SENSITIVITY_UNSPECIFIED|START_SENSITIVITY_HIGH|START_SENSITIVITY_LOW,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Start-of-speech sensitivity.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection.endOfSpeechSensitivity,body,enum,false,END_SENSITIVITY_HIGH,,,END_SENSITIVITY_UNSPECIFIED|END_SENSITIVITY_HIGH|END_SENSITIVITY_LOW,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,End-of-speech sensitivity.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection.prefixPaddingMs,body,integer (int32),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Speech duration before start-of-speech is committed (look-back); 0 may clip word onsets.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.automaticActivityDetection.silenceDurationMs,body,integer (int32),false,≈800 (server internal),,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Silence before end-of-speech is committed; server default ≈ 800 ms; recommended 500-800; too low fragments utterances, too high adds latency.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.activityHandling,body,enum,false,START_OF_ACTIVITY_INTERRUPTS,,,ACTIVITY_HANDLING_UNSPECIFIED|START_OF_ACTIVITY_INTERRUPTS|NO_INTERRUPTION,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"START_OF_ACTIVITY_INTERRUPTS (default, barge-in) or NO_INTERRUPTION.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.realtimeInputConfig.turnCoverage,body,enum,false,model-dependent,,,TURN_COVERAGE_UNSPECIFIED|TURN_INCLUDES_ONLY_ACTIVITY|TURN_INCLUDES_ALL_INPUT|TURN_INCLUDES_AUDIO_ACTIVITY_AND_ALL_VIDEO,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Input included in the user turn. Default: Gemini 2.5 TURN_INCLUDES_ONLY_ACTIVITY; Gemini 3.1+ TURN_INCLUDES_AUDIO_ACTIVITY_AND_ALL_VIDEO (all video frames billed).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription,body,object (AudioTranscriptionConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Presence enables user-audio transcription -> serverContent.inputTranscription (+ interimInputTranscription on transcribe model). `{}` = defaults (auto language).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription.languageCodes[],body,string[],false,[] (auto),,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,BCP-47 hints; empty/omitted = auto-detect (multilingual).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-transcribe""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription.customVocabulary[],body,string[],false,,,1000,,gemini-3.5-transcribe-live,,DOCUMENTED,"Up to 1,000 biasing terms (best ≤ 100).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-transcribe""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription.mode,body,enum,false,VERBATIM,,,MODE_UNSPECIFIED|VERBATIM|SMART,gemini-3.5-transcribe-live,,DOCUMENTED,"VERBATIM (default) or SMART (filler removal, formatting, self-corrections). Incompatible with word annotations.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-transcribe""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription.wordTimestamp,body,boolean,false,,,,,,,DOCUMENTED|UNVERIFIED,Word-level timestamps — not supported over the Live API (unary gemini-3.5-transcribe only).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-transcribe""}"
gemini,WSS BidiGenerateContent,setup.inputAudioTranscription.diarization,body,boolean,false,,,,,,,DOCUMENTED|UNVERIFIED,Speaker diarization — not supported over the Live API.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-transcribe""}"
gemini,WSS BidiGenerateContent,setup.outputAudioTranscription,body,object (AudioTranscriptionConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview,,DOCUMENTED,Presence enables model-audio transcription -> serverContent.outputTranscription (language inferred from the response). Same sub-fields as inputAudioTranscription. Transcript text tokens billed at text output rate.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.sessionResumption,body,object (SessionResumptionConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Presence makes the server send sessionResumptionUpdate. `{}` for a new resumable session.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/session-management""}"
gemini,WSS BidiGenerateContent,setup.sessionResumption.handle,body,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Handle from a previous sessionResumptionUpdate.newHandle to resume; absent = new session. Valid 2 h after last termination. Config (except model) may change on resume.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/session-management""}"
gemini,WSS BidiGenerateContent,setup.sessionResumption.transparent,body,boolean,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,UNVERIFIED,SDK-only: request lastConsumedClientMessageIndex in updates for transparent reconnects. Not in REST reference.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://github.com/googleapis/python-genai/blob/main/google/genai/types.py""}"
gemini,WSS BidiGenerateContent,setup.contextWindowCompression,body,object (ContextWindowCompressionConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Enable sliding-window compression to lift the 15-min audio / 2-min audio+video session limits (unlimited duration).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/session-management""}"
gemini,WSS BidiGenerateContent,setup.contextWindowCompression.slidingWindow,body,object (SlidingWindow),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Sliding-window mechanism (union compressionMechanism). `{}` accepts defaults.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/session-management""}"
gemini,WSS BidiGenerateContent,setup.contextWindowCompression.slidingWindow.targetTokens,body,integer (int64; JSON string or number),false,triggerTokens/2,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Tokens kept after compression. Default triggerTokens/2; must be < triggerTokens.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.contextWindowCompression.triggerTokens,body,integer (int64; JSON string or number),false,80% of context window,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Context size (before a turn) that triggers compression. Default 80% of the model context window.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.proactivity,body,object (ProactivityConfig),false,,,,,gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.8-live|gemini-3.8-live-extended-thinking,,DOCUMENTED,Proactive audio configuration.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.proactivity.proactiveAudio,body,boolean,false,false (2.5) / always true (3.8),,,,gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.8-live|gemini-3.8-live-extended-thinking,,DOCUMENTED,Model may ignore irrelevant/incomplete speech. Gemini 2.5: opt-in (v1beta). gemini-3.8-live*: permanently enabled; false returns an error. gemini-3.1-flash-live-preview: unsupported. Billing: input tokens charged the whole time the API listens.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,setup.historyConfig,body,object (HistoryConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Initial-history handshake configuration.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.historyConfig.initialHistoryInClientContent,body,boolean,false,false,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"If true, after setupComplete the server ingests clientContent (may end with role model) until turnComplete=true without calling the model; then realtimeInput begins.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,setup.labels,body,"object<string,string>",false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,UNVERIFIED,"Discovery schema only (not on the public reference): user labels, e.g. {""safety_identifier"": ""user_session_123""}; keys/values ≤ 63 chars, lowercase, digits, _ -.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta""}"
gemini,WSS BidiGenerateContent,setup.explicitVadSignal,body,boolean,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,UNVERIFIED,SDK-only (LiveClientSetup.explicit_vad_signal): client sends VAD signals; server emits voiceActivityDetectionSignal ('allowlisted only').,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://github.com/googleapis/python-genai/blob/main/google/genai/types.py""}"
gemini,WSS BidiGenerateContent,setup.safetySettings[],body,object[] (SafetySetting),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,UNVERIFIED,SDK-only on LiveClientSetup/LiveConnectConfig; not on the public Live reference.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://github.com/googleapis/python-genai/blob/main/google/genai/types.py""}"
gemini,WSS BidiGenerateContent,clientContent,body,object (BidiGenerateContentClientContent),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Incremental conversation update; interrupts current generation.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,clientContent.turns[],body,object[] (Content),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Turns appended to history (single or multi-turn). Roles user/model supported throughout the session (3.1+).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,clientContent.turns[].role,body,enum,false,,,,user|model,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,user | model,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,clientContent.turns[].parts[],body,object[] (Part),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,text / inlineData / functionCall / functionResponse parts.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,clientContent.turnComplete,body,boolean,false,false,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,true = generate now (unconditionally interrupts); false/absent = server waits for more messages.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,realtimeInput,body,object (BidiGenerateContentRealtimeInput),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Realtime streams; one sub-field per message.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,realtimeInput.audio,body,object (Blob),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Audio chunk.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.audio.mimeType,body,string,true,audio/pcm;rate=16000,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,`audio/pcm;rate=16000` (or the actual sample rate; server resamples). Raw 16-bit LE PCM mono.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.audio.data,body,string (base64),true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Base64 PCM bytes; send 20-100 ms chunks.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/best-practices""}"
gemini,WSS BidiGenerateContent,realtimeInput.video,body,object (Blob),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Single image frame (≤ 1 FPS).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.video.mimeType,body,string,true,,,,image/jpeg|image/png,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,image/jpeg or image/png.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/get-started-websocket""}"
gemini,WSS BidiGenerateContent,realtimeInput.video.data,body,string (base64),true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Base64 image bytes.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/get-started-websocket""}"
gemini,WSS BidiGenerateContent,realtimeInput.text,body,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Realtime text stream.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/get-started-websocket""}"
gemini,WSS BidiGenerateContent,realtimeInput.activityStart,body,object {},false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Manual VAD: start of user activity (requires automaticActivityDetection.disabled=true).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.activityEnd,body,object {},false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Manual VAD: end of user activity; server finalizes immediately.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.audioStreamEnd,body,boolean,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Automatic-VAD only: mic off / hybrid-VAD end-of-speech -> flush & finalize. Resume by sending audio.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/capabilities""}"
gemini,WSS BidiGenerateContent,realtimeInput.mediaResolution,body,enum,false,,,,MEDIA_RESOLUTION_UNSPECIFIED|MEDIA_RESOLUTION_LOW|MEDIA_RESOLUTION_MEDIUM|MEDIA_RESOLUTION_HIGH,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Per-message override of setup.generationConfig.mediaResolution.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,realtimeInput.mediaChunks[],body,object[] (Blob),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED|DEPRECATED,DEPRECATED; only first chunk used. Use audio/video/text.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse,body,object (BidiGenerateContentToolResponse),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Function results for a toolCall.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[],body,object[] (FunctionResponse),true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"One per functionCall, matched by id.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].id,body,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,= FunctionCall.id.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].name,body,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,= FunctionCall.name.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].response,body,object,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,"JSON result (""output""/""result""/""error"" keys of your choice). Docs examples also place `scheduling` inside `response` for SDK convenience.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].scheduling,body,enum,false,WHEN_IDLE,,,SCHEDULING_UNSPECIFIED|SILENT|WHEN_IDLE|INTERRUPT,gemini-3.8-live|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,NON_BLOCKING only: INTERRUPT | WHEN_IDLE (default) | SILENT. gemini-3.8-live-extended-thinking: scheduling not supported.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/tools""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].willContinue,body,boolean,false,false,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,NON_BLOCKING only: more responses will follow (generator); false ends the call (pair with scheduling=SILENT to avoid triggering generation).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,WSS BidiGenerateContent,toolResponse.functionResponses[].parts[],body,object[] (FunctionResponsePart),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest,,DOCUMENTED,Multimodal response parts (inlineData/fileData).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,x-goog-api-key,header,string,true,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Standard API key header (server side). Body content-type application/json.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,uses,body,integer (int32),false,1,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Number of sessions the token can start; 0 = unlimited; default 1. Resuming does not count.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,expireTime,body,string (RFC 3339),false,now + 30 minutes,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,When messages in sessions using the token stop being accepted (session may be closed). Default now+30 min; must be < 20 h ahead.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,newSessionExpireTime,body,string (RFC 3339),false,now + 60 seconds,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"After this, new sessions with the token are rejected. Default now+60 s; must be < 20 h ahead.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,fieldMask,body,"string (FieldMask, comma-separated paths)",false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Controls merge of bidiGenerateContentSetup with the client's setup: empty & no setup = client's; empty & setup = token's only; non-empty = listed fields overwrite. SDK: lock_additional_fields.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,bidiGenerateContentSetup,body,object (BidiGenerateContentSetup),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Locked Live setup (union `config`), same fields as the WebSocket setup message (model, generationConfig, systemInstruction, tools, ...).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,bidiGenerateContentSetup.model,body,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,`models/{model}` locked model.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/api/live""}"
gemini,POST /v1beta/auth_tokens,liveConnectConstraints,body,"object {model, config}",false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED|UNVERIFIED,"Shape used in official REST curl examples (and SDK live_connect_constraints): model + SDK-style config (responseModalities, sessionResumption, systemInstruction, tools, translationConfig, inputAudioTranscription ...). Not in the discovery AuthToken schema — REST acceptance UNVERIFIED; the SDK converts it to bidiGenerateContentSetup + fieldMask.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,liveConnectConstraints.model,body,string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,Model to lock (`models/gemini-3.8-live` in REST examples; bare id in SDK).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,liveConnectConstraints.config,body,object (LiveConnectConfig),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"Config to lock; e.g. {""sessionResumption"": {}, ""responseModalities"": [""AUDIO""]}. Fields omitted here can be set by the client only if unlocked (lock_additional_fields: []).","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,config.uses,sdk-config,integer,false,1,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,SDK (python client.auth_tokens.create(config=...) / node client.authTokens.create({config})): same as uses.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,config.expire_time,sdk-config,datetime | ISO string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,SDK: expire_time (python) / expireTime (node).,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,config.new_session_expire_time,sdk-config,datetime | ISO string,false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,SDK: new_session_expire_time / newSessionExpireTime.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,config.live_connect_constraints,sdk-config,object (LiveConnectConstraints),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,"SDK: live_connect_constraints / liveConnectConstraints {model, config}.","{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/ephemeral-tokens""}"
gemini,POST /v1beta/auth_tokens,config.lock_additional_fields,sdk-config,string[],false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,SDK: extra LiveConnectParameters fields to lock (translate guide: set [] to leave translationConfig settable client-side). Maps to fieldMask.,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://ai.google.dev/gemini-api/docs/live-api/live-translate""}"
gemini,POST /v1beta/auth_tokens,config.http_options,sdk-config,object (HttpOptions),false,,,,,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-3.1-flash-live-preview|gemini-2.5-flash-native-audio-preview-12-2025|gemini-2.5-flash-native-audio-preview-09-2025|gemini-2.5-flash-native-audio-latest|gemini-3.5-live-translate-preview|gemini-3.5-transcribe-live,,DOCUMENTED,SDK: HTTP overrides (e.g. api_version 'v1beta').,"{""retrieved_at"": ""2026-09-18"", ""url"": ""https://github.com/googleapis/python-genai/blob/main/google/genai/types.py""}"
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData (image),body,"object {mimeType,data}",false,,,10,,lyria-3.5,,DOCUMENTED,Up to 10 images alongside the text prompt; the model composes music inspired by the visuals (Lyria 3.5).,https://ai.google.dev/gemini-api/docs/music-generation#generate-music-from-images
gemini,POST /v1beta/interactions,response_format,body,object {type:'audio'},false,,,,,lyria-3.5,,DOCUMENTED,Interactions API: request WAV output instead of default MP3 (lyria-3.5).,https://ai.google.dev/gemini-api/docs/music-generation#select-output-format
gemini,WS BidiGenerateMusic,setup.model,body,string,true,,,,,lyria-realtime-exp,,DOCUMENTED|BETA|LIVE_VERIFIED,"First (and only first) client message: {""setup"":{""model"":""models/lyria-realtime-exp""}}. Wait for setupComplete before sending anything else.",https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,WS BidiGenerateMusic,clientContent.weightedPrompts[],body,"array<{text, weight}>",true,,,,,lyria-realtime-exp,,DOCUMENTED,"Weighted prompts steering the music; weight != 0, weights normalized, 1.0 is a good start. Can be re-sent any time; model transitions smoothly.",https://ai.google.dev/gemini-api/docs/realtime-music-generation#prompt-lyria-realtime
gemini,WS BidiGenerateMusic,musicGenerationConfig.guidance,body,number,false,4.0,0.0,6.0,,lyria-realtime-exp,,DOCUMENTED,Prompt adherence; higher = closer but more abrupt transitions.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.bpm,body,integer,false,,60,200,,lyria-realtime-exp,,DOCUMENTED,Beats per minute; requires RESET_CONTEXT (or stop/play) to take effect.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.density,body,number,false,,0.0,1.0,,lyria-realtime-exp,,DOCUMENTED,Note/sound density.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.brightness,body,number,false,,0.0,1.0,,lyria-realtime-exp,,DOCUMENTED,Tonal brightness (higher frequencies).,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.scale,body,enum,false,,,,SCALE_UNSPECIFIED|C_MAJOR_A_MINOR|D_FLAT_MAJOR_B_FLAT_MINOR|D_MAJOR_B_MINOR|E_FLAT_MAJOR_C_MINOR|E_MAJOR_D_FLAT_MINOR|F_MAJOR_D_MINOR|G_FLAT_MAJOR_E_FLAT_MINOR|G_MAJOR_E_MINOR|A_FLAT_MAJOR_F_MINOR|A_MAJOR_G_FLAT_MINOR|B_FLAT_MAJOR_G_MINOR|B_MAJOR_A_FLAT_MINOR,lyria-realtime-exp,,DOCUMENTED,Key/mode (relative major/minor pairs); requires RESET_CONTEXT.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#scale-enum-values
gemini,WS BidiGenerateMusic,musicGenerationConfig.muteBass,body,boolean,false,false,,,,lyria-realtime-exp,,DOCUMENTED,Reduce bass.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.muteDrums,body,boolean,false,false,,,,lyria-realtime-exp,,DOCUMENTED,Reduce drums.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.onlyBassAndDrums,body,boolean,false,false,,,,lyria-realtime-exp,,DOCUMENTED,Only bass and drums.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.musicGenerationMode,body,enum,false,QUALITY,,,QUALITY|DIVERSITY|VOCALIZATION,lyria-realtime-exp,,DOCUMENTED,QUALITY (default) | DIVERSITY | VOCALIZATION (vocalizations as an instrument).,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.temperature,body,number,false,1.1,0.0,3.0,,lyria-realtime-exp,,DOCUMENTED,Sampling temperature.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.topK,body,integer,false,40,1,1000,,lyria-realtime-exp,,DOCUMENTED,Top-k sampling.,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,musicGenerationConfig.seed,body,integer,false,,0,2147483647,,lyria-realtime-exp,,DOCUMENTED,Seed (random by default).,https://ai.google.dev/gemini-api/docs/realtime-music-generation#controls
gemini,WS BidiGenerateMusic,playbackControl,body,enum,false,,,,PLAY|PAUSE|STOP|RESET_CONTEXT,lyria-realtime-exp,,DOCUMENTED,PLAY starts generation; PAUSE holds; STOP stops and resets context (prompts kept); RESET_CONTEXT resets without stopping (needed after bpm/scale changes). The whole musicGenerationConfig must be re-sent on update (fields not sent reset to defaults).,https://ai.google.dev/gemini-api/docs/realtime-music-generation#update-the-configuration
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.speechConfig.voiceConfig.prebuiltVoiceConfig.voiceName,body,string,false,,,,Zephyr|Puck|Charon|Kore|Fenrir|Leda|Orus|Aoede|Callirrhoe|Autonoe|Enceladus|Iapetus|Umbriel|Algieba|Despina|Erinome|Algenib|Rasalgethi|Laomedeia|Achernar|Alnilam|Schedar|Gacrux|Pulcherrima|Achird|Zubenelgenubi|Vindemiatrix|Sadachbia|Sadaltager|Sulafat,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED|LIVE_VERIFIED,One of the 30 prebuilt voices (single speaker).,https://ai.google.dev/gemini-api/docs/speech-generation#voice-options
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.speechConfig.multiSpeakerVoiceConfig.speakerVoiceConfigs[],body,"array<{speaker, voiceConfig}>",false,,,2,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED|LIVE_VERIFIED,Multi-speaker TTS: up to 2 speakers; `speaker` must match the names used in the prompt. Mutually exclusive with voiceConfig.,https://ai.google.dev/gemini-api/docs/speech-generation#multi-speaker-tts
gemini,POST /v1beta/models/{model}:streamGenerateContent?alt=sse,generationConfig.responseModalities,body,array<enum>,true,,,,AUDIO,gemini-3.1-flash-tts-preview,,DOCUMENTED|PREVIEW|LIVE_VERIFIED,Streaming TTS (gemini-3.1-flash-tts-preview only): SSE `data:` events each carrying one inlineData audio chunk; the final event has finishReason STOP and full usageMetadata.,https://ai.google.dev/gemini-api/docs/speech-generation#streaming-speech-generation
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.speechConfig.languageCode,body,string,false,,,,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED,Optional BCP-47 code; TTS auto-detects the input language (80+ languages listed).,https://ai.google.dev/api/generate-content#SpeechConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.responseFormat.audio,body,object (AudioResponseFormat),false,,,,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED|UNVERIFIED,"Flat output format: mimeType (AUDIO_MP3|AUDIO_OGG_OPUS|AUDIO_L16|AUDIO_WAV|AUDIO_ALAW|AUDIO_MULAW), delivery (INLINE|URI), sampleRate, bitRate. Default output is raw PCM L16 24 kHz mono.",https://ai.google.dev/api/generate-content#AudioResponseFormat
gemini,POST /v1beta/models/{model}:streamGenerateContent,alt=sse,query,string,false,,,,,gemini-3.1-flash-tts-preview,,DOCUMENTED,Streaming TTS is supported only from gemini-3.1-flash-tts-preview (streamGenerateContent / Interactions stream:true).,https://ai.google.dev/gemini-api/docs/speech-generation#streaming-speech-generation
gemini,POST /v1beta/interactions,generation_config.speech_config[],body,"array<{speaker?, voice, language?}>",false,,,,,gemini-3.1-flash-tts-preview|gemini-2.5-flash-preview-tts|gemini-2.5-pro-preview-tts,,DOCUMENTED,"Interactions API TTS config: [{voice:'Kore'}] or [{speaker:'Joe', voice:'Kore'},{speaker:'Jane', voice:'Puck'}] with response_format {type:'audio'}.",https://ai.google.dev/gemini-api/docs/speech-generation#single-speaker-tts
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].inlineData|fileData (audio),body,object,true,,,,,gemini-3.5-transcribe,,DOCUMENTED|LIVE_VERIFIED,"gemini-3.5-transcribe input: audio (WAV, MP3, AIFF, AAC, OGG, FLAC, MPEG, M4A, L16, Opus, ALAW, MULAW, WebM), up to 1 h (30 min with diarization/timestamps). Use the Files API for anything longer than a few seconds.",https://ai.google.dev/gemini-api/docs/transcribe
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.languageCodes,body,array<string>,false,,,,,gemini-3.5-transcribe,,DOCUMENTED|LIVE_VERIFIED,"BCP-47 hints (85+ locales, e.g. en-US, fr-FR, cmn-Hans-CN). Empty/omitted = automatic detection with code-switching.",https://ai.google.dev/api/generate-content#AudioTranscriptionConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.mode,body,enum,false,VERBATIM,,,VERBATIM|SMART,gemini-3.5-transcribe,,DOCUMENTED,"VERBATIM (default: fillers, repetitions kept) or SMART (disfluency removal, inline self-corrections, structured formatting). SMART is incompatible with timestamps and diarization.",https://ai.google.dev/api/generate-content#AudioTranscriptionConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.customVocabulary,body,array<string>,false,,,1000,,gemini-3.5-transcribe,,DOCUMENTED,"Up to 1,000 bias terms (best <= 100). Incompatible with diarization and word timestamps (request rejected).",https://ai.google.dev/gemini-api/docs/transcribe#custom-vocabulary
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.adaptationPhrases,body,array<string>,false,,,,,gemini-3.5-transcribe,,LIVE_DISCOVERED,Discovery-document alias of speech adaptation phrases (same purpose as customVocabulary).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.diarization,body,boolean,false,,,,,gemini-3.5-transcribe,,DOCUMENTED|LIVE_VERIFIED,Speaker diarization (labels spk_1..spk_8 in the Interactions docs; generateContent returned speakerLabel 'spk:0'; 3+ speakers experimental).,https://ai.google.dev/api/generate-content#AudioTranscriptionConfig
gemini,POST /v1beta/models/{model}:generateContent,generationConfig.audioTranscriptionConfig.wordTimestamp,body,boolean,false,,,,,gemini-3.5-transcribe,,DOCUMENTED|LIVE_VERIFIED,"Word-level start/end offsets (may reduce accuracy). Returned in parts[].audioTranscription.words[] {word, startOffset, endOffset} (LIVE_DISCOVERED shape).",https://ai.google.dev/api/generate-content#AudioTranscriptionConfig
gemini,POST /v1beta/interactions,generation_config.transcription_config,body,object,false,,,,,gemini-3.5-transcribe,,DOCUMENTED,"Interactions API form: {language_codes[], custom_vocabulary[], mode: 'smart' | {type:'verbatim', diarization_mode:'speaker', timestamp_granularities:['word']}}. Word annotations come back as word_info annotations on the text block.",https://ai.google.dev/gemini-api/docs/transcribe#parameter-reference
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations,body,array<FunctionDeclaration>,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. A list of `FunctionDeclarations` available to the model that can be used for function calling. The model or system does not execute the function. Instead the defined function may be returned as a FunctionCall with arguments to the client side for execution. The model may decide to call a subset of these functions by populating FunctionCall in the response. The next conversation turn may contain a FunctionResponse with the Content.role ""function"" generation context for the next model turn. Max 512 declarations (SDK docstring).",https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].name,body,string,true,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Required. The name of the function. Must be a-z, A-Z, 0-9, or contain underscores, colons, dots, and dashes, with a maximum length of 128. (FunctionCall.name is limited to letters, digits, underscores and dashes.)",https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].description,body,string,true,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Required (REST reference). Required. A brief description of the function. The model relies on it to choose the function; be specific.,https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].parameters,body,"object (Schema, OpenAPI 3.03 subset)",false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. Describes the parameters to this function. Reflects the Open API 3.03 Parameter Object string Key: the name of the parameter. Parameter names are case sensitive. Schema Value: the Schema defining the type used for the parameter. Supported Schema fields: type, format, title, description, nullable, enum, items, minItems, maxItems, properties, required, minProperties, maxProperties, minimum, maximum, minLength, maxLength, pattern, anyOf, propertyOrdering, default, example. Mutually exclusive with parametersJsonSchema.",https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].parametersJsonSchema,body,object (JSON Schema),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. Describes the parameters to the function in JSON Schema format. The schema must describe an object where the properties are the parameters to the function. For example: ``` { ""type"": ""object"", ""properties"": { ""name"": { ""type"": ""string"" }, ""age"": { ""type"": ""integer"" } }, ""additionalProperties"": false, ""required"": [""name"", ""age""], ""propertyOrdering"": [""name"", ""age""] } ``` This field is mutually exclusive with `parameters`.",https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].response,body,object (Schema),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Optional. Describes the output from this function in JSON Schema format. Reflects the Open API 3.03 Response Object. The Schema defines the type used for the response value of the function. Mutually exclusive with responseJsonSchema.,https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,POST /v1beta/models/{model}:generateContent,tools[].functionDeclarations[].responseJsonSchema,body,object (JSON Schema),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Optional. Describes the output from this function in JSON Schema format. The value specified by the schema is the response value of the function. This field is mutually exclusive with `response`.,https://ai.google.dev/api/generate-content#FunctionDeclaration
gemini,WSS BidiGenerateContent (setup.tools) — ignored by generateContent,tools[].functionDeclarations[].behavior,body,string,false,BLOCKING,,,UNSPECIFIED|BLOCKING|NON_BLOCKING,gemini-3.8-live|gemini-3.8-live-extended-thinking|gemini-2.5-flash-native-audio-preview-12-2025,,DOCUMENTED,Optional. Specifies the function Behavior. Currently only supported by the BidiGenerateContent method. REST reference: defaults to BLOCKING. Live API: NON_BLOCKING = asynchronous function calling (default on gemini-3.8-live; the only mode on gemini-3.8-live-extended-thinking; unsupported on gemini-3.1-flash-live-preview).,https://ai.google.dev/api/generate-content#Behavior
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch,body,object (GoogleSearch),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite|gemini-3.1-flash-image|gemini-3-pro-image|gemini-3.8-live|gemini-3.1-flash-live-preview,,DOCUMENTED,Optional. GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google. Empty object enables web search grounding.,https://ai.google.dev/api/generate-content#GoogleSearch
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.timeRangeFilter,body,object (Interval),false,,,,,,,DOCUMENTED|UNVERIFIED,"Optional. Filter search results to a specific time range. If customers set a start time, they must set an end time (and vice versa).",https://ai.google.dev/api/generate-content#GoogleSearch
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.timeRangeFilter.startTime,body,string (RFC 3339 timestamp),false,,,,,,,DOCUMENTED|UNVERIFIED,"Optional. Inclusive start of the interval. If specified, a Timestamp matching this interval will have to be the same or after the start.",https://ai.google.dev/api/generate-content#Interval
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.timeRangeFilter.endTime,body,string (RFC 3339 timestamp),false,,,,,,,DOCUMENTED|UNVERIFIED,"Optional. Exclusive end of the interval. If specified, a Timestamp matching this interval will have to be before the end.",https://ai.google.dev/api/generate-content#Interval
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.searchTypes,body,object (SearchTypes),false,web search only,,,,,,DOCUMENTED,"Optional. The set of search types to enable. If not set, web search is enabled by default.",https://ai.google.dev/api/generate-content#SearchTypes
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.searchTypes.webSearch,body,"object (WebSearch, empty)",false,,,,,,,DOCUMENTED,Optional. Enables web search. Only text results are returned.,https://ai.google.dev/api/generate-content#SearchTypes
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearch.searchTypes.imageSearch,body,"object (ImageSearch, empty)",false,,,,,gemini-3.1-flash-image|gemini-3.1-flash-image-preview,,DOCUMENTED,Optional. Enables image search. Image bytes are returned. Documented for image generation (gemini-3.1-flash-image); cannot search for people.,https://ai.google.dev/gemini-api/docs/generate-content/image-generation
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearchRetrieval,body,object (GoogleSearchRetrieval),false,,,,,gemini-2.0-flash (RETIRED),,DOCUMENTED|LEGACY|DEPRECATED,"Optional. Retrieval tool that is powered by Google search. LEGACY: for Gemini 1.5/2.0 models only ('Older models use a google_search_retrieval tool. For all current models, use the google_search tool'). 2.0 models shut down 2026-06-01.",https://ai.google.dev/api/generate-content#GoogleSearchRetrieval
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearchRetrieval.dynamicRetrievalConfig,body,object (DynamicRetrievalConfig),false,,,,,,,DOCUMENTED|LEGACY,Specifies the dynamic retrieval configuration for the given source.,https://ai.google.dev/api/generate-content#DynamicRetrievalConfig
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearchRetrieval.dynamicRetrievalConfig.mode,body,string,false,,,,MODE_UNSPECIFIED|MODE_DYNAMIC,,,DOCUMENTED|LEGACY,The mode of the predictor to be used in dynamic retrieval. MODE_UNSPECIFIED = always trigger retrieval; MODE_DYNAMIC = only when the predictor decides it is necessary.,https://ai.google.dev/api/generate-content#Mode
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleSearchRetrieval.dynamicRetrievalConfig.dynamicThreshold,body,number,false,,0,1,,,,DOCUMENTED|LEGACY,"The threshold to be used in dynamic retrieval. If not set, a system default value is used. Compared with groundingMetadata.retrievalMetadata.googleSearchDynamicRetrievalScore in [0,1].",https://ai.google.dev/api/generate-content#DynamicRetrievalConfig
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleMaps,body,object (GoogleMaps),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Optional. Tool that allows grounding the model's response with geospatial context related to the user's query. Empty object enables Grounding with Google Maps (off by default).,https://ai.google.dev/api/generate-content#GoogleMaps
gemini,POST /v1beta/models/{model}:generateContent,tools[].googleMaps.enableWidget,body,boolean,false,false,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Optional. Whether to return a widget context token in the GroundingMetadata of the response. Developers can use the widget context token to render a Google Maps widget with geospatial context related to the places that the model references in the response.,https://ai.google.dev/api/generate-content#GoogleMaps
gemini,POST /v1beta/models/{model}:generateContent,tools[].urlContext,body,"object (UrlContext, empty)",false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. Tool to support URL context retrieval. No fields. URLs are taken from the prompt text (max 20 per request, 34 MB per URL).",https://ai.google.dev/api/generate-content#UrlContext
gemini,POST /v1beta/models/{model}:generateContent,tools[].codeExecution,body,"object (CodeExecution, empty)",false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. Enables the model to execute code as part of generation. No fields. Python-only sandbox, 30 s runtime.",https://ai.google.dev/api/generate-content#CodeExecution
gemini,POST /v1beta/models/{model}:generateContent,tools[].computerUse,body,object (ComputerUse),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3-flash-preview|gemini-2.5-computer-use-preview-10-2025,,DOCUMENTED|PREVIEW,"Optional. Tool to support the model interacting directly with the computer. If enabled, it automatically populates computer-use specific Function Declarations.",https://ai.google.dev/api/generate-content#ComputerUse
gemini,POST /v1beta/models/{model}:generateContent,tools[].computerUse.environment,body,string,true,ENVIRONMENT_BROWSER (when UNSPECIFIED),,,ENVIRONMENT_UNSPECIFIED|ENVIRONMENT_BROWSER|ENVIRONMENT_MOBILE|ENVIRONMENT_DESKTOP,,,DOCUMENTED|PREVIEW,Required. The environment being operated. ENVIRONMENT_UNSPECIFIED defaults to browser. MOBILE and DESKTOP are Gemini 3.x only.,https://ai.google.dev/api/generate-content#Environment
gemini,POST /v1beta/models/{model}:generateContent,tools[].computerUse.excludedPredefinedFunctions,body,array<string>,false,,,,,,,DOCUMENTED|PREVIEW,"Optional. By default, predefined functions are included in the final model call. Some of them can be explicitly excluded from being automatically included. This can serve two purposes: 1. Using a more restricted / different action space. 2. Improving the definitions / instructions of predefined functions. Values are predefined action names (e.g. click, type, scroll, navigate; legacy: open_web_browser, click_at, type_text_at, scroll_document ...).",https://ai.google.dev/api/generate-content#ComputerUse
gemini,POST /v1beta/models/{model}:generateContent,tools[].computerUse.enablePromptInjectionDetection,body,boolean,false,false,,,,gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.6-flash|gemini-3.7-flash|gemini-3.8-flash,,DOCUMENTED|PREVIEW,Optional. Whether enable the prompt injection detection check on computer-use request. Opt-in; Gemini 3.5 Flash or later. Scans screenshots for hidden adversarial instructions and blocks execution when detected.,https://ai.google.dev/api/generate-content#ComputerUse
gemini,POST /v1beta/models/{model}:generateContent,tools[].computerUse.disabledSafetyPolicies,body,array<string (SafetyPolicy)>,false,,,,SAFETY_POLICY_UNSPECIFIED|FINANCIAL_TRANSACTIONS|SENSITIVE_DATA_MODIFICATION|COMMUNICATION_TOOL|ACCOUNT_CREATION|DATA_MODIFICATION|USER_CONSENT_MANAGEMENT|LEGAL_TERMS_AND_AGREEMENTS,,,DOCUMENTED|PREVIEW,Optional. Disabled safety policies for computer use. Gemini 3.x. Overrides indicate preference; the model may still return safety_decision require_confirmation.,https://ai.google.dev/api/generate-content#SafetyPolicy
gemini,POST /v1beta/models/{model}:generateContent,tools[].fileSearch,body,object (FileSearch),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash-lite|gemini-2.5-flash,,DOCUMENTED|PREVIEW,Optional. FileSearch tool type. Tool to retrieve knowledge from Semantic Retrieval corpora.,https://ai.google.dev/api/generate-content#FileSearch
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers,body,array<McpServer>,false,,,,,,,DOCUMENTED|DOCUMENTATION_INCOMPLETE|UNVERIFIED,Optional. MCP Servers to connect to. Present in discovery/REST reference; not covered by any generateContent guide (server-side MCP is documented for the Interactions API as type mcp_server). UNVERIFIED on the Gemini Developer API.,https://ai.google.dev/api/generate-content#McpServer
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].name,body,string,false,,,,,,,DOCUMENTED|UNVERIFIED,The name of the MCPServer.,https://ai.google.dev/api/generate-content#McpServer
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport,body,object (StreamableHttpTransport),false,,,,,,,DOCUMENTED|UNVERIFIED,A transport that can stream HTTP requests and responses. Only transport in the union.,https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport.url,body,string,false,,,,,,,DOCUMENTED|UNVERIFIED,"The full URL for the MCPServer endpoint. Example: ""https://api.example.com/mcp""",https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport.headers,body,"object (map<string,string>)",false,,,,,,,DOCUMENTED|UNVERIFIED,"Optional: Fields for authentication headers, timeouts, etc., if needed.",https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport.timeout,body,"string (duration, e.g. '3.5s')",false,,,,,,,DOCUMENTED|UNVERIFIED,HTTP timeout for regular operations.,https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport.sseReadTimeout,body,string (duration),false,,,,,,,DOCUMENTED|UNVERIFIED,Timeout for SSE read operations.,https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,tools[].mcpServers[].streamableHttpTransport.terminateOnClose,body,boolean,false,,,,,,,DOCUMENTED|UNVERIFIED,Whether to close the client session when the transport closes.,https://ai.google.dev/api/generate-content#StreamableHttpTransport
gemini,POST /v1beta/models/{model}:generateContent,toolConfig,body,object (ToolConfig),false,,,,,,,DOCUMENTED,"Optional. Tool configuration for any Tool specified in the request: functionCallingConfig, retrievalConfig, includeServerSideToolInvocations.",https://ai.google.dev/api/generate-content#ToolConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.functionCallingConfig,body,object (FunctionCallingConfig),false,,,,,,,DOCUMENTED,Optional. Function calling config.,https://ai.google.dev/api/generate-content#FunctionCallingConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.functionCallingConfig.mode,body,string,false,AUTO (VALIDATED in tool-combination requests),,,MODE_UNSPECIFIED|AUTO|ANY|NONE|VALIDATED,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. Specifies the mode in which function calling should execute. If unspecified, the default value will be set to AUTO. AUTO: model chooses call or text. ANY: always a (schema-constrained) function call. NONE: never call. VALIDATED: call or text with constrained-decoding validation; default (and only allowed mode) when built-in tools/structured outputs are combined or includeServerSideToolInvocations=true. ANY may reject very large/deeply nested schemas.",https://ai.google.dev/api/generate-content#FunctionCallingConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.functionCallingConfig.allowedFunctionNames,body,array<string>,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,"Optional. A set of function names that, when provided, limits the functions the model will call. This should only be set when the Mode is ANY or VALIDATED. Function names should match [FunctionDeclaration.name]. When set, model will predict a function call from only allowed function names.",https://ai.google.dev/api/generate-content#FunctionCallingConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.retrievalConfig,body,object (RetrievalConfig),false,,,,,,,DOCUMENTED,Optional. Retrieval config. Used by Grounding with Google Maps (user location) and search.,https://ai.google.dev/api/generate-content#RetrievalConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.retrievalConfig.latLng,body,object (LatLng),false,,,,,,,DOCUMENTED,Optional. The location of the user. WGS84 degrees.,https://ai.google.dev/api/generate-content#LatLng
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.retrievalConfig.latLng.latitude,body,number,false,,-90,90,,,,DOCUMENTED,"The latitude in degrees. It must be in the range [-90.0, +90.0].",https://ai.google.dev/api/generate-content#LatLng
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.retrievalConfig.latLng.longitude,body,number,false,,-180,180,,,,DOCUMENTED,"The longitude in degrees. It must be in the range [-180.0, +180.0].",https://ai.google.dev/api/generate-content#LatLng
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.retrievalConfig.languageCode,body,string (BCP-47),false,,,,,,,DOCUMENTED,Optional. The language code of the user. Language code for content. Use language tags defined by [BCP47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt).,https://ai.google.dev/api/generate-content#RetrievalConfig
gemini,POST /v1beta/models/{model}:generateContent,toolConfig.includeServerSideToolInvocations,body,boolean,false,false,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Optional. If true, the API response will include the server-side tool calls and responses within the `Content` message. This allows clients to observe the server's tool interactions. Enables Gemini 3 'tool context circulation': toolCall/toolResponse parts are returned and must be echoed back with their id, toolType and thoughtSignature. Forces VALIDATED function-calling mode (AUTO unsupported). Preview, Gemini 3 only.",https://ai.google.dev/gemini-api/docs/generate-content/tool-combination
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionCall,body,object (FunctionCall),false,,,,,,,DOCUMENTED,Model-generated part (echo it back verbatim in history). A predicted `FunctionCall` returned from the model that contains a string representing the `FunctionDeclaration.name` with the arguments and their values.,https://ai.google.dev/api/generate-content#FunctionCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionCall.name,body,string,true,,,,,,,DOCUMENTED,"Required. The name of the function to call. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 128.",https://ai.google.dev/api/generate-content#FunctionCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionCall.args,body,object,false,,,,,,,DOCUMENTED,Optional. The function parameters and values in JSON object format.,https://ai.google.dev/api/generate-content#FunctionCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionCall.id,body,string,false,,,,,,,DOCUMENTED,"Optional. Unique identifier of the function call. If populated, the client to execute the `function_call` and return the response with the matching `id`. Gemini 3 always returns an id; you must echo it in functionResponse.id.",https://ai.google.dev/api/generate-content#FunctionCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse,body,object (FunctionResponse),false,,,,,,,DOCUMENTED,"Client-provided result of a functionCall, sent in the next content (role user or function). The result output from a `FunctionCall` that contains a string representing the `FunctionDeclaration.name` and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a`FunctionCall` made based on model prediction.",https://ai.google.dev/api/generate-content#FunctionResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.name,body,string,true,,,,,,,DOCUMENTED,"Required. The name of the function to call. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 128.",https://ai.google.dev/api/generate-content#FunctionResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.response,body,object,true,,,,,,,DOCUMENTED,"Required. The function response in JSON object format. Callers can use any keys of their choice that fit the function's syntax to return the function output, e.g. ""output"", ""result"", etc. In particular, if the function call failed to execute, the response can have an ""error"" key to return error details to the model. Multimedia can be included by using a subobject containing a single ""$ref"" key whose value is the `inline_data.display_name` of a `FunctionResponsePart` holding the multimedia. See https://ai.google.dev/gemini-api/docs/function-calling#multimodal.",https://ai.google.dev/api/generate-content#FunctionResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.id,body,string,false,,,,,,,DOCUMENTED,Optional. The identifier of the function call this response is for. Populated by the client to match the corresponding function call `id`. Mandatory mapping on Gemini 3.,https://ai.google.dev/api/generate-content#FunctionResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.parts,body,array<FunctionResponsePart>,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED,"Optional. Ordered `Parts` that constitute a function response. Parts may have different IANA MIME types. Gemini 3 multimodal function responses: image/png, image/jpeg, image/webp, application/pdf, text/plain.",https://ai.google.dev/api/generate-content#FunctionResponsePart
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.parts[].inlineData,body,object (FunctionResponseBlob),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED,Inline media bytes. Only member of the data union.,https://ai.google.dev/api/generate-content#FunctionResponseBlob
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.parts[].inlineData.mimeType,body,string,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED,"The IANA standard MIME type of the source data. Examples: - image/png - image/jpeg If an unsupported MIME type is provided, an error will be returned. For a complete list of supported types, see [Supported file formats](https://ai.google.dev/gemini-api/docs/prompting_with_media#supported_file_formats).",https://ai.google.dev/api/generate-content#FunctionResponseBlob
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].functionResponse.parts[].inlineData.data,body,string (base64),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED,Raw bytes for media formats.,https://ai.google.dev/api/generate-content#FunctionResponseBlob
gemini,WSS BidiGenerateContent (toolResponse.functionResponses[]) — ignored by generateContent,contents[].parts[].functionResponse.scheduling,body,string,false,WHEN_IDLE,,,SCHEDULING_UNSPECIFIED|SILENT|WHEN_IDLE|INTERRUPT,gemini-3.8-live|gemini-2.5-flash-native-audio-preview-12-2025,,DOCUMENTED,"Optional. Specifies how the response should be scheduled in the conversation. Only applicable to NON_BLOCKING function calls, is ignored otherwise. Defaults to WHEN_IDLE.",https://ai.google.dev/api/generate-content#Scheduling
gemini,WSS BidiGenerateContent (toolResponse.functionResponses[]) — ignored by generateContent,contents[].parts[].functionResponse.willContinue,body,boolean,false,,,,,gemini-3.8-live|gemini-2.5-flash-native-audio-preview-12-2025,,DOCUMENTED,"Optional. Signals that function call continues, and more responses will be returned, turning the function call into a generator. Is only applicable to NON_BLOCKING function calls, is ignored otherwise. If set to false, future responses will not be considered. It is allowed to return empty `response` with `will_continue=False` to signal that the function call is finished. This may still trigger the model generation. To avoid triggering the generation and finish the function call, additionally set `scheduling` to `SILENT`.",https://ai.google.dev/api/generate-content#FunctionResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].thoughtSignature,body,string (base64),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview|gemini-2.5-pro|gemini-2.5-flash|gemini-2.5-flash-lite,,DOCUMENTED,Optional. An opaque signature for the thought so it can be reused in subsequent requests. Gemini 3: mandatory on the first functionCall part of every step of the current turn (400 'Function call X in the N. content block is missing a thought_signature' otherwise); recommended on other parts. Gemini 2.5: optional. Dummy values 'skip_thought_signature_validator' / 'context_engineering_is_the_way_to_go' bypass validation for injected history. Do not merge/split parts carrying signatures.,https://ai.google.dev/gemini-api/docs/generate-content/thought-signatures
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].executableCode,body,object (ExecutableCode),false,,,,,,,DOCUMENTED,"Model-generated (codeExecution tool) — echo back in multi-turn history with id and thoughtSignature. Code generated by the model that is meant to be executed, and the result returned to the model. Only generated when using the `CodeExecution` tool, in which the code will be automatically executed, and a corresponding `CodeExecutionResult` will also be generated.",https://ai.google.dev/api/generate-content#ExecutableCode
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].executableCode.language,body,string,true,,,,LANGUAGE_UNSPECIFIED|PYTHON,,,DOCUMENTED,Required. Programming language of the `code`.,https://ai.google.dev/api/generate-content#Language
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].executableCode.code,body,string,true,,,,,,,DOCUMENTED,Required. The code to be executed.,https://ai.google.dev/api/generate-content#ExecutableCode
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].executableCode.id,body,string,false,,,,,,,DOCUMENTED,Optional. Unique identifier of the `ExecutableCode` part. The server returns the `CodeExecutionResult` with the matching `id`.,https://ai.google.dev/api/generate-content#ExecutableCode
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].codeExecutionResult,body,object (CodeExecutionResult),false,,,,,,,DOCUMENTED,Server-generated result of executableCode — echo back in history. Result of executing the `ExecutableCode`. Generated only when the `CodeExecution` tool is used.,https://ai.google.dev/api/generate-content#CodeExecutionResult
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].codeExecutionResult.outcome,body,string,true,,,,OUTCOME_UNSPECIFIED|OUTCOME_OK|OUTCOME_FAILED|OUTCOME_DEADLINE_EXCEEDED,,,DOCUMENTED,Required. Outcome of the code execution.,https://ai.google.dev/api/generate-content#Outcome
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].codeExecutionResult.output,body,string,false,,,,,,,DOCUMENTED,"Optional. Contains stdout when code execution is successful, stderr or other description otherwise.",https://ai.google.dev/api/generate-content#CodeExecutionResult
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].codeExecutionResult.id,body,string,false,,,,,,,DOCUMENTED,Optional. The identifier of the `ExecutableCode` part this result is for. Only populated if the corresponding `ExecutableCode` has an id.,https://ai.google.dev/api/generate-content#CodeExecutionResult
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall,body,object (ToolCall),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Server-side tool call returned when toolConfig.includeServerSideToolInvocations=true (Gemini 3 tool combination). A predicted server-side `ToolCall` returned from the model. This message contains information about a tool that the model wants to invoke. The client is NOT expected to execute this `ToolCall`. Instead, the client should pass this `ToolCall` back to the API in a subsequent turn within a `Content` message, along with the corresponding `ToolResponse`.",https://ai.google.dev/api/generate-content#ToolCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall.toolType,body,string,true,,,,TOOL_TYPE_UNSPECIFIED|GOOGLE_SEARCH_WEB|GOOGLE_SEARCH_IMAGE|URL_CONTEXT|GOOGLE_MAPS|FILE_SEARCH,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,Required. The type of tool that was called.,https://ai.google.dev/api/generate-content#ToolType
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall.args,body,object,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Optional. The tool call arguments. Example: {""arg1"" : ""value1"", ""arg2"" : ""value2"" , ...} User-visible args: GOOGLE_SEARCH queries[], GOOGLE_MAPS queries[], URL_CONTEXT urls[], FILE_SEARCH none.",https://ai.google.dev/api/generate-content#ToolCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall.id,body,string,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,Optional. Unique identifier of the tool call. The server returns the tool response with the matching `id`.,https://ai.google.dev/api/generate-content#ToolCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolCall.toolName,body,string,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,Optional. The name of the tool that was called.,https://ai.google.dev/api/generate-content#ToolCall
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolResponse,body,object (ToolResponse),false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Server-side tool result echoed with the toolCall. The output from a server-side `ToolCall` execution. This message contains the results of a tool invocation that was initiated by a `ToolCall` from the model. The client should pass this `ToolResponse` back to the API in a subsequent turn within a `Content` message, along with the corresponding `ToolCall`.",https://ai.google.dev/api/generate-content#ToolResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolResponse.toolType,body,string,true,,,,TOOL_TYPE_UNSPECIFIED|GOOGLE_SEARCH_WEB|GOOGLE_SEARCH_IMAGE|URL_CONTEXT|GOOGLE_MAPS|FILE_SEARCH,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Required. The type of tool that was called, matching the `tool_type` in the corresponding `ToolCall`.",https://ai.google.dev/api/generate-content#ToolType
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolResponse.response,body,object,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,"Optional. The tool response. User-visible: GOOGLE_SEARCH search_suggestions; GOOGLE_MAPS places, google_maps_widget_context_token; URL_CONTEXT urls_metadata[{retrieved_url,url_retrieval_status}].",https://ai.google.dev/api/generate-content#ToolResponse
gemini,POST /v1beta/models/{model}:generateContent,contents[].parts[].toolResponse.id,body,string,false,,,,,gemini-3.8-flash|gemini-3.7-flash|gemini-3.6-flash|gemini-3.5-flash|gemini-3.5-flash-lite|gemini-3.1-pro-preview|gemini-3.1-flash-lite|gemini-3-flash-preview,,DOCUMENTED|PREVIEW,Optional. The identifier of the tool call this response is for.,https://ai.google.dev/api/generate-content#ToolResponse
gemini,POST /v1beta/tunedModels,tunedModelId,query,string,false,,,,,,,DOCUMENTED|LEGACY,"Optional id, <= 40 chars, regex [a-z]([a-z0-9-]{0,38}[a-z0-9])?",https://ai.google.dev/api/tuning#method:-tunedmodels.create
gemini,POST /v1beta/tunedModels,displayName,body,string,false,,,40,,,,DOCUMENTED|LEGACY,<= 40 chars; used to derive the resource id (tunedModels/sentence-translator-u3b7m).,https://ai.google.dev/api/tuning
gemini,POST /v1beta/tunedModels,description,body,string,false,,,,,,,DOCUMENTED|LEGACY,Short description.,https://ai.google.dev/api/tuning
gemini,POST /v1beta/tunedModels,baseModel,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"Immutable. Model to tune, e.g. models/gemini-1.5-flash-001. No base model currently supports tuning in the Gemini API (Gemini 1.5 Flash-001 deprecated May 2025).",https://ai.google.dev/gemini-api/docs/model-tuning
gemini,POST /v1beta/tunedModels,tunedModelSource.tunedModel,body,string,false,,,,,,,DOCUMENTED|LEGACY,Immutable. Existing tunedModels/… to start from (alternative to baseModel).,https://ai.google.dev/api/tuning#TunedModelSource
gemini,POST /v1beta/tunedModels,tuningTask.trainingData.examples.examples[],body,"array<{textInput, output}>",true,,,,,,,DOCUMENTED|LEGACY,Inline text examples (required); all examples in a set must be the same type.,https://ai.google.dev/api/tuning#TuningExample
gemini,POST /v1beta/tunedModels,tuningTask.hyperparameters.epochCount,body,integer,false,5,,,,,,DOCUMENTED|LEGACY,Immutable; default 5.,https://ai.google.dev/api/tuning#Hyperparameters
gemini,POST /v1beta/tunedModels,tuningTask.hyperparameters.batchSize,body,integer,false,,,,,,,DOCUMENTED|LEGACY,Immutable; default 4 or 16 depending on dataset size.,https://ai.google.dev/api/tuning#Hyperparameters
gemini,POST /v1beta/tunedModels,tuningTask.hyperparameters.learningRate,body,number,false,,,,,,,DOCUMENTED|LEGACY,Immutable; default 0.001 or 0.0002 by dataset size. Mutually exclusive with learningRateMultiplier.,https://ai.google.dev/api/tuning#Hyperparameters
gemini,POST /v1beta/tunedModels,tuningTask.hyperparameters.learningRateMultiplier,body,number,false,1.0,,,,,,DOCUMENTED|LEGACY,Immutable; actual lr = multiplier x default (default 1.0).,https://ai.google.dev/api/tuning#Hyperparameters
gemini,POST /v1beta/tunedModels,temperature,body,number,false,,0.0,1.0,,,,DOCUMENTED|LEGACY,Default sampling temperature of the tuned model.,https://ai.google.dev/api/tuning
gemini,POST /v1beta/tunedModels,topP,body,number,false,,,,,,,DOCUMENTED|LEGACY,Default nucleus sampling.,https://ai.google.dev/api/tuning
gemini,POST /v1beta/tunedModels,topK,body,integer,false,,,,,,,DOCUMENTED|LEGACY,Default top-k.,https://ai.google.dev/api/tuning
gemini,POST /v1beta/tunedModels,readerProjectNumbers[],body,array<string (int64)>,false,,,,,,,DOCUMENTED|LEGACY,Projects with read access.,https://ai.google.dev/api/tuning
gemini,GET /v1beta/tunedModels,pageSize,query,integer,false,10,,1000,,,,DOCUMENTED|LEGACY|RETIRED,"Default 10, max 1000 per page.",https://ai.google.dev/api/tuning#method:-tunedmodels.list
gemini,GET /v1beta/tunedModels,pageToken,query,string,false,,,,,,,DOCUMENTED|LEGACY,Pagination token.,https://ai.google.dev/api/tuning
gemini,GET /v1beta/tunedModels,filter,query,string,false,,,,,,,DOCUMENTED|LEGACY,"Full-text search over description/displayName; operators owner:me, writers:me, readers:me, readers:everyone.",https://ai.google.dev/api/tuning
gemini,PATCH /v1beta/{tunedModel.name=tunedModels/*},updateMask,query,string (FieldMask),false,,,,,,,DOCUMENTED|LEGACY,"Fields to update (displayName, description, …).",https://ai.google.dev/api/tuning#method:-tunedmodels.patch
gemini,POST /v1beta/{name=tunedModels/*}:transferOwnership,emailAddress,body,string,true,,,,,,,DOCUMENTED|LEGACY,New owner's e-mail; only way to change the owner.,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/{parent=tunedModels/*}/permissions,granteeType,body,enum,false,,,,USER|GROUP|EVERYONE,,,DOCUMENTED|LEGACY,USER | GROUP | EVERYONE (immutable).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/{parent=tunedModels/*}/permissions,emailAddress,body,string,false,,,,,,,DOCUMENTED|LEGACY,User/group e-mail (not set for EVERYONE).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/{parent=tunedModels/*}/permissions,role,body,enum,true,,,,OWNER|WRITER|READER,,,DOCUMENTED|LEGACY,OWNER | WRITER | READER (required).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,POST /v1beta/{model=tunedModels/*}:generateContent,contents[],body,array<Content>,true,,,,,,,DOCUMENTED|LEGACY,Same body as models.generateContent; the tuned model is addressed as tunedModels/{id}.,https://ai.google.dev/api/tuning#method:-tunedmodels.generatecontent
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].prompt,body,string,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview|veo-3.0-generate-001|veo-3.0-fast-generate-001|veo-2.0-generate-001,,DOCUMENTED,"Text description (<= 1,024 tokens). Supports audio cues: quoted dialogue, SFX, ambient noise. Optional if image or video is provided (SDK).",https://ai.google.dev/gemini-api/docs/veo#veo-api-parameters-and-specifications
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].image,body,"object Image {inlineData{mimeType,data}}",false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview|veo-3.0-generate-001|veo-3.0-fast-generate-001|veo-2.0-generate-001,,DOCUMENTED,Initial frame for image-to-video. Not allowed together with `video`.,https://ai.google.dev/gemini-api/docs/veo#image-to-video-generation
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].lastFrame,body,object Image,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,Final frame for interpolation; must be combined with `image`. Veo 3.1 family.,https://ai.google.dev/gemini-api/docs/veo#using-first-and-last-frames
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].referenceImages[],body,"array<{image: Image, referenceType}>",false,,,3,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview,,DOCUMENTED,Up to 3 reference images (subject/product/character consistency). Not on Veo 3.1 Lite. durationSeconds must be 8. personGeneration allow_adult only.,https://ai.google.dev/gemini-api/docs/veo#using-reference-images
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].referenceImages[].referenceType,body,enum,false,,,,asset|style,veo-3.1-generate-preview|veo-3.1-fast-generate-preview,,DOCUMENTED,asset = object/character/scene to include; style = aesthetics (SDK enum ASSET|STYLE). Docs samples use lowercase 'asset'.,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,instances[].video,body,"object Video {inlineData{mimeType:'video/mp4',data}} or {uri}",false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview,,DOCUMENTED,"Veo-generated video to extend (+7 s per call, up to 20 times, input <= 141 s, output <= 148 s, 720p, 9:16 or 16:9, generated/referenced within the last 2 days). Veo 3.1 and 3.1 Fast only.",https://ai.google.dev/gemini-api/docs/veo#extending-veo-videos
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.aspectRatio,body,string,false,16:9,,,16:9|9:16,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview|veo-3.0-generate-001|veo-3.0-fast-generate-001|veo-2.0-generate-001,,DOCUMENTED,Landscape or portrait.,https://ai.google.dev/gemini-api/docs/veo#control-the-aspect-ratio
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.durationSeconds,body,string|integer,false,8,,,4|6|8,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,"Clip length. Must be ""8"" with extension, reference images, 1080p or 4k.",https://ai.google.dev/gemini-api/docs/veo#veo-api-parameters-and-specifications
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.resolution,body,string,false,720p,,,720p|1080p|4k,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,Output resolution. 1080p and 4k require 8 s; 4k not on Veo 3.1 Lite; extension is 720p only. Higher resolution = higher latency and price.,https://ai.google.dev/gemini-api/docs/veo#control-the-resolution
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.personGeneration,body,string,false,,,,allow_all|allow_adult|dont_allow,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview|veo-3.0-generate-001|veo-3.0-fast-generate-001|veo-2.0-generate-001,,DOCUMENTED,"Text-to-video and extension: allow_all only; image-to-video, interpolation, reference images: allow_adult only. EU/UK/CH/MENA: allow_adult only.",https://ai.google.dev/gemini-api/docs/veo#limitations
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.negativePrompt,body,string,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,What to exclude from the video (SDK GenerateVideosConfig.negative_prompt).,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.seed,body,integer,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview|veo-3.0-generate-001|veo-3.0-fast-generate-001|veo-2.0-generate-001,,DOCUMENTED,RNG seed; improves but does not guarantee determinism.,https://ai.google.dev/gemini-api/docs/veo#veo-api-parameters-and-specifications
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.numberOfVideos,body,integer,false,1,1,1,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,Videos per request (Gemini API: 1).,https://ai.google.dev/gemini-api/docs/veo#extending-veo-videos
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.generateAudio,body,boolean,false,true,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,SDK field; Veo 3.x always generates audio (always on).,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.enhancePrompt,body,boolean,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,SDK field: prompt rewriting.,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.fps,body,integer,false,24,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,SDK field; output is 24 fps.,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.compressionQuality,body,enum,false,,,,OPTIMIZED|LOSSLESS,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,SDK VideoCompressionQuality (OPTIMIZED|LOSSLESS).,https://github.com/googleapis/python-genai/blob/main/google/genai/types.py
gemini,POST /v1beta/models/{model}:predictLongRunning,parameters.webhookConfig.uris[],body,array<string>,false,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED|UNVERIFIED,SDK: webhook URIs notified when the operation completes (Webhooks feature).,https://ai.google.dev/gemini-api/docs/webhooks
gemini,POST /v1beta/models/{model}:predictLongRunning,labels,body,"object<string,string>",false,,,,,,,DOCUMENTED,User-defined labels for billing tracking (PredictLongRunningRequest.labels).,https://generativelanguage.googleapis.com/$discovery/rest?version=v1beta
gemini,GET /v1beta/{name=models/*/operations/*},name,path,string,true,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,"Operation name returned by predictLongRunning, e.g. models/veo-3.1-generate-preview/operations/abc123. Poll until done=true (latency 11 s to 6 min).",https://ai.google.dev/gemini-api/docs/veo#handling-asynchronous-operations
gemini,GET {response.generateVideoResponse.generatedSamples[].video.uri},x-goog-api-key,header,string,true,,,,,veo-3.1-generate-preview|veo-3.1-fast-generate-preview|veo-3.1-lite-generate-preview,,DOCUMENTED,Download the generated MP4: GET the returned uri (https://generativelanguage.googleapis.com/v1beta/files/<id>:download?alt=media) with the API key header and follow redirects (-L). Stored 2 days.,https://ai.google.dev/gemini-api/docs/veo#text-to-video-generation
openai,GET /v1/organization/admin_api_keys,after,query,string,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/admin_api_keys,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/admin_api_keys,limit,query,integer,false,20,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/admin_api_keys,name,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/admin_api_keys,expires_in_seconds,body,integer,false,,1,31536000,,,,DOCUMENTED,The number of seconds until the API key expires. Omit this field for a key that does not expire.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/admin_api_keys/{key_id},key_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/admin_api_keys/{key_id},key_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/audit_logs,effective_at,query,object,false,,,,,,,DOCUMENTED,Return only events whose `effective_at` (Unix seconds) is in this range.,https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,project_ids[],query,array<string>,false,,,,,,,DOCUMENTED,Return only events for these projects.,https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,event_types[],query,array<string>,false,,,,api_key.created|api_key.updated|api_key.deleted|certificate.created|certificate.updated|certificate.deleted|certificates.activated|certificates.deactivated|checkpoint.permission.created|checkpoint.permission.deleted|external_key.registered|external_key.removed|external_storage.registered|external_storage.removed|group.created|group.updated|group.deleted|invite.sent|invite.accepted|invite.deleted|ip_allowlist.created|ip_allowlist.updated|ip_allowlist.deleted|ip_allowlist.config.activated|ip_allowlist.config.deactivated|login.succeeded|login.failed|logout.succeeded|logout.failed|organization.updated|project.created|project.updated|project.archived|project.deleted|rate_limit.updated|rate_limit.deleted|resource.deleted|tunnel.created|tunnel.updated|tunnel.deleted|workload_identity_provider.created|workload_identity_provider.updated|workload_identity_provider.deleted|workload_identity_provider_mapping.created|workload_identity_provider_mapping.updated|workload_identity_provider_mapping.deleted|role.created|role.updated|role.deleted|role.assignment.created|role.assignment.deleted|role.bound_to_resource|role.unbound_from_resource|scim.enabled|scim.disabled|service_account.created|service_account.updated|service_account.deleted|user.added|user.updated|user.deleted|tenant.metadata.updated|tenant.microsoft_entra_mapping.upserted|tenant.microsoft_entra_mapping.deleted|tenant.workload_identity.provider.created|tenant.workload_identity.provider.updated|tenant.workload_identity.provider.archived|tenant.workload_identity.mapping.created|tenant.workload_identity.mapping.updated|tenant.workload_identity.mapping.archived|tenant.workload_identity.binding.created|tenant.workload_identity.principal.provisioned|tenant.workload_identity.access_token.issued|tenant.admin_api_key.created|tenant.admin_api_key.updated|tenant.admin_api_key.deleted|tenant.project_api_key.created|tenant.trusted_access.business_verification.started|tenant.trusted_access.application.submitted|tenant.chatgpt_access_token.revoked|tenant.migration.completed|tenant.sso.migrated|tenant.domains.migrated|tenant.sso_connection.created|tenant.sso_connection.updated|tenant.sso_connection.deleted|tenant.sso_connection.setup.started|tenant.policy.created|tenant.policy.updated|tenant.policy.deleted|tenant.policy.attached|tenant.policy.detached|tenant.principal_authentication_policy.resolved|tenant.scim.setup.started|tenant.scim.deletion.requested|tenant.scim.directory.created|tenant.product_access_policy.updated|tenant.resource_share_grant.created|tenant.resource_share_grant.updated|tenant.resource_share_grant.accepted|tenant.resource_share_grant.declined|tenant.resource_share_grant.revoked|tenant.resource_share_grant.deleted|tenant.service_account.updated|tenant.service_account.deleted|tenant.service_account.token.revoked|tenant.billing.overage_limit.updated|tenant.billing.alerts.updated|tenant.billing.info.updated|tenant.usage_limit.workspace.updated|tenant.usage_limit.group.updated|tenant.usage_limit.user.updated|tenant.usage_limit.increase_request.updated|tenant.usage_limit.increase_request.resolved|tenant.group.created|tenant.group.updated|tenant.group.deleted|tenant.group.member.added|tenant.group.member.removed|tenant.migration_rollout.status.updated|tenant.migration_rollout.tier.updated|tenant.role.metadata.updated|tenant.custom_role.created|tenant.custom_role.updated|tenant.custom_role.deleted|tenant.role_assignment.created|tenant.role_assignment.deleted|tenant.resource_role_assignment.created|tenant.resource_role_assignment.deleted|tenant.resource_access.updated|tenant.resource_access.deleted|tenant.ads_account.onboarding.redemption|tenant.session_policy.created|tenant.session_policy.updated|tenant.session_policy.deleted|tenant.session_revocation.started|tenant.third_party_app_policy.updated|tenant.user.added|tenant.user.updated|tenant.user.removed|tenant.user.looked_up|tenant.user.invited|tenant.membership.revoked|tenant.api_organization_invite.upserted|tenant.api_organization_invite.deleted|tenant.chatgpt_workspace_invite.upserted|tenant.membership.accepted|tenant.membership.declined|tenant.workspace_invite_email_settings.updated,,,DOCUMENTED,"Return only events with a `type` in one of these values. For example, `project.created`. For all options, see the documentation for the [audit log object](https://developers.openai.com/api/reference/resources/admin/subresources/organization/subresources/audit_logs).",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,actor_ids[],query,array<string>,false,,,,,,,DOCUMENTED,"Return only events performed by these actors. Can be a user ID, a service account ID, or an api key tracking ID.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,actor_emails[],query,array<string>,false,,,,,,,DOCUMENTED,Return only events performed by users with these emails.,https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,resource_ids[],query,array<string>,false,,,,,,,DOCUMENTED,"Return only events performed on these targets. For example, a project ID updated. For ChatGPT connector role events, use the workspace connector resource ID shown in `details.id`, such as `<workspace_id>__<connector_id>`.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,tenant_only,query,boolean,false,false,,,,,,DOCUMENTED,"Return only tenant-scoped events associated with this organization. Required for tenant-scoped events such as `role.bound_to_resource` and `role.unbound_from_resource`. When `true`, all supplied event types must be tenant-scoped.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/audit_logs,before,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/audit_logs/methods/list
openai,GET /v1/organization/certificates,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/certificates,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/certificates,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates,name,body,string,false,,,,,,,DOCUMENTED,An optional name for the certificate,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates,certificate,body,string,true,,,,,,,DOCUMENTED,The certificate content in PEM format,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates/activate,certificate_ids,body,array<string>,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates/deactivate,certificate_ids,body,array<string>,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/certificates/{certificate_id},certificate_id,path,string,true,,,,,,,DOCUMENTED,Unique ID of the certificate to retrieve.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/certificates/{certificate_id},include,query,array<string>,false,,,,content,,,DOCUMENTED,A list of additional fields to include in the response. Currently the only supported value is `content` to fetch the PEM content of the certificate.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates/{certificate_id},certificate_id,path,string,true,,,,,,,DOCUMENTED,Unique ID of the certificate to modify.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/certificates/{certificate_id},name,body,string,false,,,,,,,DOCUMENTED,The updated name for the certificate,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/certificates/{certificate_id},certificate_id,path,string,true,,,,,,,DOCUMENTED,Unique ID of the certificate to delete.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,bucket_width,query,string,false,1d,,,1d,,,DOCUMENTED,"Width of each time bucket in response. Currently only `1d` is supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only costs for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only costs for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,line_items,query,array<string>,false,,,,,,,DOCUMENTED,"Return only costs for these exact line item names. Each value must match the complete `line_item` value, for example `gpt-6-astra, input_tokens`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,group_by,query,array<string>,false,,,,project_id|line_item|api_key_id,,,DOCUMENTED,"Group the costs by the specified fields. Support fields include `project_id`, `line_item`, `api_key_id` and any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,limit,query,integer,false,7,,,,,,DOCUMENTED,"A limit on the number of buckets to be returned. Limit can range between 1 and 180, and the default is 7.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/costs,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/data_retention,retention_type,body,string,true,,,,zero_data_retention|modified_abuse_monitoring|enhanced_zero_data_retention|enhanced_modified_abuse_monitoring,,,DOCUMENTED,The desired organization data retention type.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups,limit,query,integer,false,100,0,1000,,,,DOCUMENTED,"A limit on the number of groups to be returned. Limit can range between 0 and 1000, and the default is 100.",https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/list
openai,GET /v1/organization/groups,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is a group ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with group_abc, your subsequent call can include `after=group_abc` in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/list
openai,GET /v1/organization/groups,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Specifies the sort order of the returned groups.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/list
openai,POST /v1/organization/groups,name,body,string,true,,1,255,,,,DOCUMENTED,Human readable name for the group.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/create
openai,GET /v1/organization/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to update.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/update
openai,POST /v1/organization/groups/{group_id},name,body,string,true,,1,255,,,,DOCUMENTED,New display name for the group.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/update
openai,DELETE /v1/organization/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to delete.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/methods/delete
openai,GET /v1/organization/groups/{group_id}/roles,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group whose organization role assignments you want to list.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/roles,limit,query,integer,false,,0,1000,,,,DOCUMENTED,A limit on the number of organization role assignments to return.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing organization roles.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/roles,order,query,string,false,,,,asc|desc,,,DOCUMENTED,Sort order for the returned organization roles.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/groups/{group_id}/roles,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group that should receive the organization role.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/groups/{group_id}/roles,role_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the role to assign.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/roles/{role_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the organization role to retrieve for the group.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/groups/{group_id}/roles/{role_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to modify.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/groups/{group_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the organization role to remove from the group.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/users,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to inspect.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/list
openai,GET /v1/organization/groups/{group_id}/users,limit,query,integer,false,100,0,1000,,,,DOCUMENTED,"A limit on the number of users to be returned. Limit can range between 0 and 1000, and the default is 100.",https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/list
openai,GET /v1/organization/groups/{group_id}/users,after,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Provide the ID of the last user from the previous list response to retrieve the next page.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/list
openai,GET /v1/organization/groups/{group_id}/users,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Specifies the sort order of users in the list.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/list
openai,POST /v1/organization/groups/{group_id}/users,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to update.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/create
openai,POST /v1/organization/groups/{group_id}/users,user_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the user to add to the group.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/create
openai,GET /v1/organization/groups/{group_id}/users/{user_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/groups/{group_id}/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to retrieve from the group.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/groups/{group_id}/users/{user_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to update.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/delete
openai,DELETE /v1/organization/groups/{group_id}/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to remove from the group.,https://developers.openai.com/api/reference/resources/organization/subresources/groups/subresources/users/methods/delete
openai,GET /v1/organization/invites,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/list
openai,GET /v1/organization/invites,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/list
openai,POST /v1/organization/invites,email,body,string,true,,,,,,,DOCUMENTED,Send an email to this address,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/create
openai,POST /v1/organization/invites,role,body,string,true,,,,reader|owner,,,DOCUMENTED,`owner` or `reader`,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/create
openai,POST /v1/organization/invites,projects,body,array<object>,false,,,,,,,DOCUMENTED,"An array of projects to which membership is granted at the same time the org invite is accepted. If omitted, the user will be invited to the default project for compatibility with legacy behavior. If empty list is passed, the user will not be invited to any projects, including the default one.",https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/create
openai,POST /v1/organization/invites,projects[].id,body,string,true,,,,,,,DOCUMENTED,Project's public ID,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/create
openai,POST /v1/organization/invites,projects[].role,body,string,true,,,,member|owner,,,DOCUMENTED,Project membership role,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/create
openai,GET /v1/organization/invites/{invite_id},invite_id,path,string,true,,,,,,,DOCUMENTED,The ID of the invite to retrieve.,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/retrieve
openai,DELETE /v1/organization/invites/{invite_id},invite_id,path,string,true,,,,,,,DOCUMENTED,The ID of the invite to delete.,https://developers.openai.com/api/reference/resources/organization/subresources/invites/methods/delete
openai,GET /v1/organization/projects,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/list
openai,GET /v1/organization/projects,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/list
openai,GET /v1/organization/projects,include_archived,query,boolean,false,false,,,,,,DOCUMENTED,If `true` returns all projects including those that have been `archived`. Archived projects are not included by default.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/list
openai,POST /v1/organization/projects,name,body,string,true,,,,,,,DOCUMENTED,"The friendly name of the project, this name appears in reports.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/create
openai,POST /v1/organization/projects,geography,body,string | null,false,,,,,,,DOCUMENTED,Create the project with the specified data residency region. Your organization must have access to Data residency functionality in order to use. See [data residency controls](https://developers.openai.com/api/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field. Deprecated: use `residency` instead. Do not provide both `geography` and `residency`.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/create
openai,POST /v1/organization/projects,residency,body,string | null,false,,,,GLOBAL|US_STORAGE_PROCESSING|EU_STORAGE_PROCESSING|JP_STORAGE|KR_STORAGE|CA_STORAGE|SG_STORAGE|IN_STORAGE|AU_STORAGE|GB_STORAGE|AE_STORAGE|AE_STORAGE_PROCESSING,,,DOCUMENTED,Create the project with the specified residency configuration. Your organization must have access to the requested residency configuration in order to use it. See [data residency controls](https://developers.openai.com/api/docs/guides/your-data#data-residency-controls) to review the functionality and limitations of setting this field.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/create
openai,POST /v1/organization/projects,external_key_id,body,string | null,false,,,,,,,DOCUMENTED,External key ID to associate with the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/create
openai,GET /v1/organization/projects/{project_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/retrieve
openai,POST /v1/organization/projects/{project_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/update
openai,POST /v1/organization/projects/{project_id},name,body,string | null,false,,,,,,,DOCUMENTED,"The updated name of the project, this name appears in reports.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/update
openai,POST /v1/organization/projects/{project_id},external_key_id,body,string | null,false,,,,,,,DOCUMENTED,External key ID to associate with the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/update
openai,POST /v1/organization/projects/{project_id},geography,body,string | null,false,,,,,,,DOCUMENTED,Geography for the project. Deprecated: use `residency` when creating a project to configure data residency. This field is retained for backward compatibility.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/methods/update
openai,GET /v1/organization/projects/{project_id}/api_keys,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/list
openai,GET /v1/organization/projects/{project_id}/api_keys,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/list
openai,GET /v1/organization/projects/{project_id}/api_keys,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/list
openai,GET /v1/organization/projects/{project_id}/api_keys,owner_project_access,query,string,false,,,,active|inactive|any,,,DOCUMENTED,"Filter API keys by whether the owner currently has effective access to the project. Use `active` for owners with access, `inactive` for owners without access, or `any` for all enabled project API keys. If omitted, the endpoint applies its existing membership-based visibility rules, which may exclude some enabled keys.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/list
openai,GET /v1/organization/projects/{project_id}/api_keys/{api_key_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/retrieve
openai,GET /v1/organization/projects/{project_id}/api_keys/{api_key_id},api_key_id,path,string,true,,,,,,,DOCUMENTED,The ID of the API key.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/retrieve
openai,DELETE /v1/organization/projects/{project_id}/api_keys/{api_key_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/delete
openai,DELETE /v1/organization/projects/{project_id}/api_keys/{api_key_id},api_key_id,path,string,true,,,,,,,DOCUMENTED,The ID of the API key.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/api_keys/methods/delete
openai,POST /v1/organization/projects/{project_id}/archive,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/certificates,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/certificates,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/certificates,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/certificates,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/certificates/activate,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/certificates/activate,certificate_ids,body,array<string>,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/certificates/deactivate,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/certificates/deactivate,certificate_ids,body,array<string>,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/data_retention,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/data_retention,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/data_retention,retention_type,body,string,true,,,,organization_default|none|zero_data_retention|modified_abuse_monitoring|enhanced_zero_data_retention|enhanced_modified_abuse_monitoring,,,DOCUMENTED,The desired project data retention type.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/groups,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/list
openai,GET /v1/organization/projects/{project_id}/groups,limit,query,integer,false,20,0,100,,,,DOCUMENTED,A limit on the number of project groups to return. Defaults to 20.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/list
openai,GET /v1/organization/projects/{project_id}/groups,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the ID of the last group from the previous response to fetch the next page.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/list
openai,GET /v1/organization/projects/{project_id}/groups,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Sort order for the returned groups.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/list
openai,POST /v1/organization/projects/{project_id}/groups,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/groups,group_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the group to add to the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/groups,role,body,string,true,,,,,,,DOCUMENTED,Identifier of the project role to grant to the group.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/groups/{group_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to retrieve.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/groups/{group_id},group_type,query,string,false,group,,,group|tenant_group,,,DOCUMENTED,The type of group to retrieve.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/groups/{group_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/delete
openai,DELETE /v1/organization/projects/{project_id}/groups/{group_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to remove from the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/groups/methods/delete
openai,GET /v1/organization/projects/{project_id}/hosted_tool_permissions,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,file_search,body,object | null,false,,,,,,,DOCUMENTED,The file search permission update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,file_search.enabled,body,boolean,true,,,,,,,DOCUMENTED,Whether to enable the hosted tool for the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,web_search,body,object | null,false,,,,,,,DOCUMENTED,The web search permission update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,web_search.enabled,body,boolean,true,,,,,,,DOCUMENTED,Whether to enable the hosted tool for the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,image_generation,body,object | null,false,,,,,,,DOCUMENTED,The image generation permission update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,image_generation.enabled,body,boolean,true,,,,,,,DOCUMENTED,Whether to enable the hosted tool for the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,mcp,body,object | null,false,,,,,,,DOCUMENTED,The MCP permission update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,mcp.enabled,body,boolean,true,,,,,,,DOCUMENTED,Whether to enable the hosted tool for the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,code_interpreter,body,object | null,false,,,,,,,DOCUMENTED,The code interpreter permission update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/hosted_tool_permissions,code_interpreter.enabled,body,boolean,true,,,,,,,DOCUMENTED,Whether to enable the hosted tool for the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/model_permissions,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/model_permissions,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/model_permissions,mode,body,string,true,,,,allow_list|deny_list,,,DOCUMENTED,The model permissions mode to apply.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/model_permissions,model_ids,body,array<string>,true,,,,,,,DOCUMENTED,The model IDs included in this permissions policy.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/model_permissions,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/rate_limits,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/rate_limits,limit,query,integer,false,100,,,,,,DOCUMENTED,A limit on the number of objects to be returned. The default is 100.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/rate_limits,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/rate_limits,before,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, beginning with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},rate_limit_id,path,string,true,,,,,,,DOCUMENTED,The ID of the rate limit.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},max_requests_per_1_minute,body,integer,false,,,,,,,DOCUMENTED,The maximum requests per minute.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},max_tokens_per_1_minute,body,integer,false,,,,,,,DOCUMENTED,The maximum tokens per minute.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},max_images_per_1_minute,body,integer,false,,,,,,,DOCUMENTED,The maximum images per minute. Only relevant for certain models.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},max_audio_megabytes_per_1_minute,body,integer,false,,,,,,,DOCUMENTED,The maximum audio megabytes per minute. Only relevant for certain models.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},max_requests_per_1_day,body,integer,false,,,,,,,DOCUMENTED,The maximum requests per day. Only relevant for certain models.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/rate_limits/{rate_limit_id},batch_1_day_max_input_tokens,body,integer,false,,,,,,,DOCUMENTED,The maximum batch input tokens per day. Only relevant for certain models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/service_accounts,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/list
openai,GET /v1/organization/projects/{project_id}/service_accounts,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/list
openai,GET /v1/organization/projects/{project_id}/service_accounts,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/list
openai,POST /v1/organization/projects/{project_id}/service_accounts,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/create
openai,POST /v1/organization/projects/{project_id}/service_accounts,name,body,string,true,,,,,,,DOCUMENTED,The name of the service account being created.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/create
openai,POST /v1/organization/projects/{project_id}/service_accounts,create_service_account_only,body,boolean | null,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/create
openai,POST /v1/organization/projects/{project_id}/service_accounts,expires_in_seconds,body,integer | null,false,,,,,,,DOCUMENTED,"Number of seconds until the initial API key expires. If omitted or null, the key does not expire unless the effective organization or project policy requires an expiration. When a policy sets a maximum lifetime, this value must be provided and must not exceed that limit. A non-null value cannot be used when `create_service_account_only` is true.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/create
openai,GET /v1/organization/projects/{project_id}/service_accounts/{service_account_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/retrieve
openai,GET /v1/organization/projects/{project_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,The ID of the service account.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/retrieve
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,The ID of the service account.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id},name,body,string,false,,,,,,,DOCUMENTED,The updated service account name.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id},role,body,string,false,,,,member|owner,,,DOCUMENTED,The updated service account role.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/service_accounts/{service_account_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/delete
openai,DELETE /v1/organization/projects/{project_id}/service_accounts/{service_account_id},service_account_id,path,string,true,,,,,,,DOCUMENTED,The ID of the service account.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/service_accounts/methods/delete
openai,GET /v1/organization/projects/{project_id}/spend_alerts,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts,limit,query,integer,false,,0,100,,,,DOCUMENTED,A limit on the number of spend alerts to return. Defaults to 20.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Sort order for the returned spend alerts.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the ID of the last spend alert from the previous response to fetch the next page.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts,before,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the ID of the first spend alert from the previous response to fetch the previous page.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,threshold_amount,body,integer,true,,0,,,,,DOCUMENTED,"The alert threshold amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,currency,body,string,true,,,,USD,,,DOCUMENTED,The currency for the threshold amount.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,interval,body,string,true,,,,month,,,DOCUMENTED,The time interval for evaluating spend against the threshold.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,notification_channel,body,object,true,,,,,,,DOCUMENTED,Email notification settings for a spend alert.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,notification_channel.type,body,string,true,,,,email,,,DOCUMENTED,The notification channel type. Currently only `email` is supported.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,notification_channel.recipients,body,array<string>,true,,,,,,,DOCUMENTED,Email addresses that receive the spend alert notification.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts,notification_channel.subject_prefix,body,string | null,false,,,,,,,DOCUMENTED,Optional subject prefix for alert emails.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts/{alert_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},threshold_amount,body,integer,true,,0,,,,,DOCUMENTED,"The alert threshold amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},currency,body,string,true,,,,USD,,,DOCUMENTED,The currency for the threshold amount.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},interval,body,string,true,,,,month,,,DOCUMENTED,The time interval for evaluating spend against the threshold.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},notification_channel,body,object,true,,,,,,,DOCUMENTED,Email notification settings for a spend alert.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},notification_channel.type,body,string,true,,,,email,,,DOCUMENTED,The notification channel type. Currently only `email` is supported.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},notification_channel.recipients,body,array<string>,true,,,,,,,DOCUMENTED,Email addresses that receive the spend alert notification.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_alerts/{alert_id},notification_channel.subject_prefix,body,string | null,false,,,,,,,DOCUMENTED,Optional subject prefix for alert emails.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/spend_alerts/{alert_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to delete.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/users,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/list
openai,GET /v1/organization/projects/{project_id}/users,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/list
openai,GET /v1/organization/projects/{project_id}/users,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/list
openai,POST /v1/organization/projects/{project_id}/users,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/create
openai,POST /v1/organization/projects/{project_id}/users,user_id,body,string | null,false,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/create
openai,POST /v1/organization/projects/{project_id}/users,email,body,string | null,false,,,,,,,DOCUMENTED,Email of the user to add.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/create
openai,POST /v1/organization/projects/{project_id}/users,role,body,string,true,,,,,,,DOCUMENTED,`owner` or `member`,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/create
openai,GET /v1/organization/projects/{project_id}/users/{user_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/retrieve
openai,GET /v1/organization/projects/{project_id}/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/retrieve
openai,POST /v1/organization/projects/{project_id}/users/{user_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/update
openai,POST /v1/organization/projects/{project_id}/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/update
openai,POST /v1/organization/projects/{project_id}/users/{user_id},role,body,string | null,false,,,,,,,DOCUMENTED,`owner` or `member`,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/update
openai,DELETE /v1/organization/projects/{project_id}/users/{user_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/delete
openai,DELETE /v1/organization/projects/{project_id}/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/projects/subresources/users/methods/delete
openai,GET /v1/organization/roles,limit,query,integer,false,1000,0,1000,,,,DOCUMENTED,A limit on the number of roles to return. Defaults to 1000.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/list
openai,GET /v1/organization/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing roles.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/list
openai,GET /v1/organization/roles,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Sort order for the returned roles.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/list
openai,POST /v1/organization/roles,role_name,body,string,true,,,,,,,DOCUMENTED,Unique name for the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/create
openai,POST /v1/organization/roles,permissions,body,array<string>,true,,,,,,,DOCUMENTED,Permissions to grant to the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/create
openai,POST /v1/organization/roles,description,body,string | null,false,,,,,,,DOCUMENTED,Optional description of the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/create
openai,GET /v1/organization/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to update.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/update
openai,POST /v1/organization/roles/{role_id},permissions,body,array<string> | null,false,,,,,,,DOCUMENTED,Updated set of permissions for the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/update
openai,POST /v1/organization/roles/{role_id},description,body,string | null,false,,,,,,,DOCUMENTED,New description for the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/update
openai,POST /v1/organization/roles/{role_id},role_name,body,string | null,false,,,,,,,DOCUMENTED,New name for the role.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/update
openai,DELETE /v1/organization/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to delete.,https://developers.openai.com/api/reference/resources/organization/subresources/roles/methods/delete
openai,GET /v1/organization/spend_alerts,limit,query,integer,false,,0,100,,,,DOCUMENTED,A limit on the number of spend alerts to return. Defaults to 20.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/spend_alerts,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Sort order for the returned spend alerts.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/spend_alerts,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the ID of the last spend alert from the previous response to fetch the next page.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/spend_alerts,before,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the ID of the first spend alert from the previous response to fetch the previous page.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,threshold_amount,body,integer,true,,0,,,,,DOCUMENTED,"The alert threshold amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,currency,body,string,true,,,,USD,,,DOCUMENTED,The currency for the threshold amount.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,interval,body,string,true,,,,month,,,DOCUMENTED,The time interval for evaluating spend against the threshold.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,notification_channel,body,object,true,,,,,,,DOCUMENTED,Email notification settings for a spend alert.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,notification_channel.type,body,string,true,,,,email,,,DOCUMENTED,The notification channel type. Currently only `email` is supported.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,notification_channel.recipients,body,array<string>,true,,,,,,,DOCUMENTED,Email addresses that receive the spend alert notification.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts,notification_channel.subject_prefix,body,string | null,false,,,,,,,DOCUMENTED,Optional subject prefix for alert emails.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},threshold_amount,body,integer,true,,0,,,,,DOCUMENTED,"The alert threshold amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},currency,body,string,true,,,,USD,,,DOCUMENTED,The currency for the threshold amount.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},interval,body,string,true,,,,month,,,DOCUMENTED,The time interval for evaluating spend against the threshold.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},notification_channel,body,object,true,,,,,,,DOCUMENTED,Email notification settings for a spend alert.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},notification_channel.type,body,string,true,,,,email,,,DOCUMENTED,The notification channel type. Currently only `email` is supported.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},notification_channel.recipients,body,array<string>,true,,,,,,,DOCUMENTED,Email addresses that receive the spend alert notification.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_alerts/{alert_id},notification_channel.subject_prefix,body,string | null,false,,,,,,,DOCUMENTED,Optional subject prefix for alert emails.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/spend_alerts/{alert_id},alert_id,path,string,true,,,,,,,DOCUMENTED,The ID of the spend alert to delete.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_speeches,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/audio_transcriptions,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,group_by,query,array<string>,false,,,,project_id,,,DOCUMENTED,Group the usage data by the specified fields. Support fields include `project_id`.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/code_interpreter_sessions,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,batch,query,boolean,false,,,,,,,DOCUMENTED,"If `true`, return batch jobs only. If `false`, return non-batch jobs only. By default, return both.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model|batch|service_tier,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model`, `batch`, `service_tier` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/completions,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/embeddings,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,vector_store_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these vector stores.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|vector_store_id,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `vector_store_id` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/file_search_calls,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,sources,query,array<string>,false,,,,image.generation|image.edit|image.variation,,,DOCUMENTED,"Return only usages for these sources. Possible values are `image.generation`, `image.edit`, `image.variation` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,sizes,query,array<string>,false,,,,256x256|512x512|1024x1024|1792x1792|1024x1792,,,DOCUMENTED,"Return only usages for these image sizes. Possible values are `256x256`, `512x512`, `1024x1024`, `1792x1792`, `1024x1792` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model|size|source,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model`, `size`, `source` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/images,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/moderations,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,group_by,query,array<string>,false,,,,project_id,,,DOCUMENTED,Group the usage data by the specified fields. Support fields include `project_id`.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/vector_stores,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,start_time,query,integer,true,,,,,,,DOCUMENTED,"Start time (Unix seconds) of the query time range, inclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,end_time,query,integer,false,,,,,,,DOCUMENTED,"End time (Unix seconds) of the query time range, exclusive.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,bucket_width,query,string,false,1d,,,1m|1h|1d,,,DOCUMENTED,"Width of each time bucket in response. Currently `1m`, `1h` and `1d` are supported, default to `1d`.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,project_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these projects.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,user_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these users.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,api_key_ids,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these API keys.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,models,query,array<string>,false,,,,,,,DOCUMENTED,Return only usage for these models.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,context_levels,query,array<string>,false,,,,low|medium|high,,,DOCUMENTED,Return only web search usage for these context levels.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,group_by,query,array<string>,false,,,,project_id|user_id|api_key_id|model|context_level,,,DOCUMENTED,"Group the usage data by the specified fields. Support fields include `project_id`, `user_id`, `api_key_id`, `model`, `context_level` or any combination of them.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,limit,query,integer,false,,,,,,,DOCUMENTED,"Specifies the number of buckets to return. - `bucket_width=1d`: default: 7, max: 31 - `bucket_width=1h`: default: 24, max: 168 - `bucket_width=1m`: default: 60, max: 1440",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/usage/web_search_calls,page,query,string,false,,,,,,,DOCUMENTED,A cursor for use in pagination. Corresponding to the `next_page` field from the previous response.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/users,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/list
openai,GET /v1/organization/users,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/list
openai,GET /v1/organization/users,emails,query,array<string>,false,,,,,,,DOCUMENTED,Filter by the email address of users.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/list
openai,GET /v1/organization/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/retrieve
openai,POST /v1/organization/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/update
openai,POST /v1/organization/users/{user_id},role,body,string | null,false,,,,,,,DOCUMENTED,`owner` or `reader`,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/update
openai,POST /v1/organization/users/{user_id},role_id,body,string | null,false,,,,,,,DOCUMENTED,Role ID to assign to the user.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/update
openai,POST /v1/organization/users/{user_id},technical_level,body,string | null,false,,,,,,,DOCUMENTED,Technical level metadata.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/update
openai,POST /v1/organization/users/{user_id},developer_persona,body,string | null,false,,,,,,,DOCUMENTED,Developer persona metadata.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/update
openai,DELETE /v1/organization/users/{user_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user.,https://developers.openai.com/api/reference/resources/organization/subresources/users/methods/delete
openai,GET /v1/organization/users/{user_id}/roles,user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to inspect.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/list
openai,GET /v1/organization/users/{user_id}/roles,limit,query,integer,false,,0,1000,,,,DOCUMENTED,A limit on the number of organization role assignments to return.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/list
openai,GET /v1/organization/users/{user_id}/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing organization roles.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/list
openai,GET /v1/organization/users/{user_id}/roles,order,query,string,false,,,,asc|desc,,,DOCUMENTED,Sort order for the returned organization roles.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/list
openai,POST /v1/organization/users/{user_id}/roles,user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user that should receive the organization role.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/create
openai,POST /v1/organization/users/{user_id}/roles,role_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the role to assign.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/create
openai,GET /v1/organization/users/{user_id}/roles/{role_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/users/{user_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the organization role to retrieve for the user.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/users/{user_id}/roles/{role_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to modify.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/delete
openai,DELETE /v1/organization/users/{user_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the organization role to remove from the user.,https://developers.openai.com/api/reference/resources/organization/subresources/users/subresources/roles/methods/delete
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to inspect.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles,limit,query,integer,false,,0,1000,,,,DOCUMENTED,A limit on the number of project role assignments to return.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing project roles.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles,order,query,string,false,,,,asc|desc,,,DOCUMENTED,Sort order for the returned project roles.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/list
openai,POST /v1/projects/{project_id}/groups/{group_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/groups/{group_id}/roles,group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group that should receive the project role.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/groups/{group_id}/roles,role_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the role to assign.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/create
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project role to retrieve for the group.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to modify.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/delete
openai,DELETE /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},group_id,path,string,true,,,,,,,DOCUMENTED,The ID of the group whose project role assignment should be removed.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/delete
openai,DELETE /v1/projects/{project_id}/groups/{group_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project role to remove from the group.,https://developers.openai.com/api/reference/resources/projects/subresources/groups/subresources/roles/methods/delete
openai,GET /v1/projects/{project_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/roles,limit,query,integer,false,1000,0,1000,,,,DOCUMENTED,A limit on the number of roles to return. Defaults to 1000.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing roles.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/roles,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED,Sort order for the returned roles.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/list
openai,POST /v1/projects/{project_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/roles,role_name,body,string,true,,,,,,,DOCUMENTED,Unique name for the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/roles,permissions,body,array<string>,true,,,,,,,DOCUMENTED,Permissions to grant to the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/roles,description,body,string | null,false,,,,,,,DOCUMENTED,Optional description of the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/create
openai,GET /v1/projects/{project_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,GET /v1/projects/{project_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/projects/{project_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/update
openai,POST /v1/projects/{project_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to update.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/update
openai,POST /v1/projects/{project_id}/roles/{role_id},permissions,body,array<string> | null,false,,,,,,,DOCUMENTED,Updated set of permissions for the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/update
openai,POST /v1/projects/{project_id}/roles/{role_id},description,body,string | null,false,,,,,,,DOCUMENTED,New description for the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/update
openai,POST /v1/projects/{project_id}/roles/{role_id},role_name,body,string | null,false,,,,,,,DOCUMENTED,New name for the role.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/update
openai,DELETE /v1/projects/{project_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/delete
openai,DELETE /v1/projects/{project_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the role to delete.,https://developers.openai.com/api/reference/resources/projects/subresources/roles/methods/delete
openai,GET /v1/projects/{project_id}/users/{user_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/users/{user_id}/roles,user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to inspect.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/users/{user_id}/roles,limit,query,integer,false,,0,1000,,,,DOCUMENTED,A limit on the number of project role assignments to return.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/users/{user_id}/roles,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination. Provide the value from the previous response's `next` field to continue listing project roles.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/list
openai,GET /v1/projects/{project_id}/users/{user_id}/roles,order,query,string,false,,,,asc|desc,,,DOCUMENTED,Sort order for the returned project roles.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/list
openai,POST /v1/projects/{project_id}/users/{user_id}/roles,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to update.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/users/{user_id}/roles,user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user that should receive the project role.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/create
openai,POST /v1/projects/{project_id}/users/{user_id}/roles,role_id,body,string,true,,,,,,,DOCUMENTED,Identifier of the role to assign.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/create
openai,GET /v1/projects/{project_id}/users/{user_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/projects/{project_id}/users/{user_id}/roles/{role_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user to inspect.,https://developers.openai.com/api/reference/overview
openai,GET /v1/projects/{project_id}/users/{user_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project role to retrieve for the user.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/projects/{project_id}/users/{user_id}/roles/{role_id},project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project to modify.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/delete
openai,DELETE /v1/projects/{project_id}/users/{user_id}/roles/{role_id},user_id,path,string,true,,,,,,,DOCUMENTED,The ID of the user whose project role assignment should be removed.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/delete
openai,DELETE /v1/projects/{project_id}/users/{user_id}/roles/{role_id},role_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project role to remove from the user.,https://developers.openai.com/api/reference/resources/projects/subresources/users/subresources/roles/methods/delete
openai,GET /v1/organization/external_storage,project_id,query,string | null,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/external_storage,after,query,string | null,false,,,,,,,DOCUMENTED,Return external storage configurations after this ID.,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/external_storage,order,query,string,false,,,,asc|desc,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/external_storage,limit,query,integer,false,,1,100,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,project_id,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider,body,object,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.type,body,string,true,azure,,,azure,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.bucket,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.role_arn,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.tenant_id,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.subscription_id,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.resource_group,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.account_name,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage,provider.container,body,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/external_storage/{external_storage_id},external_storage_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/external_storage/{external_storage_id},external_storage_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/external_storage/{external_storage_id}/validate,external_storage_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_limit,threshold_amount,body,integer,true,,1,,,,,DOCUMENTED,"The hard spend limit amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_limit,currency,body,string,true,,,,USD,,,DOCUMENTED,"The currency for the threshold amount. Currently, only `USD` is supported.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/spend_limit,interval,body,string,true,,,,month,,,DOCUMENTED,"The time interval for evaluating spend against the threshold. Currently, only `month` is supported.",https://developers.openai.com/api/reference/overview
openai,GET /v1/organization/projects/{project_id}/spend_limit,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project whose hard spend limit is being managed.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_limit,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project whose hard spend limit is being managed.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_limit,threshold_amount,body,integer,true,,1,,,,,DOCUMENTED,"The hard spend limit amount, in cents.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_limit,currency,body,string,true,,,,USD,,,DOCUMENTED,"The currency for the threshold amount. Currently, only `USD` is supported.",https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/spend_limit,interval,body,string,true,,,,month,,,DOCUMENTED,"The time interval for evaluating spend against the threshold. Currently, only `month` is supported.",https://developers.openai.com/api/reference/overview
openai,DELETE /v1/organization/projects/{project_id}/spend_limit,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project whose hard spend limit is being managed.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys,project_id,path,string,true,,,,,,,DOCUMENTED,The ID of the project.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys,service_account_id,path,string,true,,,,,,,DOCUMENTED,The ID of the service account.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys,name,body,string,false,,,,,,,DOCUMENTED,API key name.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys,scopes,body,array<string>,false,,,,,,,DOCUMENTED,API key scopes.,https://developers.openai.com/api/reference/overview
openai,POST /v1/organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys,expires_in_seconds,body,integer | null,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,GET /v1/assistants,limit,query,integer,false,20,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/assistants,order,query,string,false,desc,,,asc|desc,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/assistants,after,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/assistants,before,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,model,body,string | AssistantSupportedModels,true,,,,gpt-5|gpt-5-mini|gpt-5-nano|gpt-5-2025-08-07|gpt-5-mini-2025-08-07|gpt-5-nano-2025-08-07|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4.1-2025-04-14|gpt-4.1-mini-2025-04-14|gpt-4.1-nano-2025-04-14|o3-mini|o3-mini-2025-01-31|o1|o1-2024-12-17|gpt-4o|gpt-4o-2024-11-20|gpt-4o-2024-08-06|gpt-4o-2024-05-13|gpt-4o-mini|gpt-4o-mini-2024-07-18|gpt-4.5-preview|gpt-4.5-preview-2025-02-27|gpt-4-turbo|gpt-4-turbo-2024-04-09|gpt-4-0125-preview|gpt-4-turbo-preview|gpt-4-1106-preview|gpt-4-vision-preview|gpt-4|gpt-4-0314|gpt-4-0613|gpt-4-32k|gpt-4-32k-0314|gpt-4-32k-0613|gpt-3.5-turbo|gpt-3.5-turbo-16k|gpt-3.5-turbo-0613|gpt-3.5-turbo-1106|gpt-3.5-turbo-0125|gpt-3.5-turbo-16k-0613,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"ID of the model to use. You can use the [List models](https://developers.openai.com/api/reference/resources/models/methods/list) API to see all of your available models, or see our [Model overview](https://developers.openai.com/api/docs/models) for descriptions of them.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,name,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,description,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,instructions,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,reasoning_effort,body,ReasoningEffort,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,tools,body,array<AssistantToolsCode | AssistantToolsFileSearch | AssistantToolsFunction>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types `code_interpreter`, `file_search`, or `function`.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,tool_resources,body,object | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,temperature,body,number | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,top_p,body,number | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants,response_format,body,AssistantsApiResponseFormatOption | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/assistants/{assistant_id},assistant_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the assistant to retrieve.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},assistant_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the assistant to modify.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},model,body,string | AssistantSupportedModels,false,,,,gpt-5|gpt-5-mini|gpt-5-nano|gpt-5-2025-08-07|gpt-5-mini-2025-08-07|gpt-5-nano-2025-08-07|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4.1-2025-04-14|gpt-4.1-mini-2025-04-14|gpt-4.1-nano-2025-04-14|o3-mini|o3-mini-2025-01-31|o1|o1-2024-12-17|gpt-4o|gpt-4o-2024-11-20|gpt-4o-2024-08-06|gpt-4o-2024-05-13|gpt-4o-mini|gpt-4o-mini-2024-07-18|gpt-4.5-preview|gpt-4.5-preview-2025-02-27|gpt-4-turbo|gpt-4-turbo-2024-04-09|gpt-4-0125-preview|gpt-4-turbo-preview|gpt-4-1106-preview|gpt-4-vision-preview|gpt-4|gpt-4-0314|gpt-4-0613|gpt-4-32k|gpt-4-32k-0314|gpt-4-32k-0613|gpt-3.5-turbo|gpt-3.5-turbo-16k|gpt-3.5-turbo-0613|gpt-3.5-turbo-1106|gpt-3.5-turbo-0125|gpt-3.5-turbo-16k-0613,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"ID of the model to use. You can use the [List models](https://developers.openai.com/api/reference/resources/models/methods/list) API to see all of your available models, or see our [Model overview](https://developers.openai.com/api/docs/models) for descriptions of them.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},reasoning_effort,body,ReasoningEffort,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},name,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},description,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},instructions,body,string | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},tools,body,array<AssistantToolsCode | AssistantToolsFileSearch | AssistantToolsFunction>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A list of tool enabled on the assistant. There can be a maximum of 128 tools per assistant. Tools can be of types `code_interpreter`, `file_search`, or `function`.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},tool_resources,body,object | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},temperature,body,number | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},top_p,body,number | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/assistants/{assistant_id},response_format,body,AssistantsApiResponseFormatOption | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,DELETE /v1/assistants/{assistant_id},assistant_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the assistant to delete.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads,messages,body,array<CreateMessageRequest>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,A list of [messages](https://developers.openai.com/api/docs/assistants/migration) to start the thread with.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads,tool_resources,body,object | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads,metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,assistant_id,body,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [assistant](https://developers.openai.com/api/docs/assistants/migration) to use to execute this run.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,thread,body,CreateThreadRequest,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"Options to create a new thread. If no thread is provided when running a
request, an empty thread will be created.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,model,body,string | string,false,,,,gpt-5|gpt-5-mini|gpt-5-nano|gpt-5-2025-08-07|gpt-5-mini-2025-08-07|gpt-5-nano-2025-08-07|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4.1-2025-04-14|gpt-4.1-mini-2025-04-14|gpt-4.1-nano-2025-04-14|gpt-4o|gpt-4o-2024-11-20|gpt-4o-2024-08-06|gpt-4o-2024-05-13|gpt-4o-mini|gpt-4o-mini-2024-07-18|gpt-4.5-preview|gpt-4.5-preview-2025-02-27|gpt-4-turbo|gpt-4-turbo-2024-04-09|gpt-4-0125-preview|gpt-4-turbo-preview|gpt-4-1106-preview|gpt-4-vision-preview|gpt-4|gpt-4-0314|gpt-4-0613|gpt-4-32k|gpt-4-32k-0314|gpt-4-32k-0613|gpt-3.5-turbo|gpt-3.5-turbo-16k|gpt-3.5-turbo-0613|gpt-3.5-turbo-1106|gpt-3.5-turbo-0125|gpt-3.5-turbo-16k-0613,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The ID of the [Model](https://developers.openai.com/api/reference/resources/models) to be used to execute this run. If a value is provided here, it will override the model associated with the assistant. If not, the model associated with the assistant will be used.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,instructions,body,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Override the default system message of the assistant. This is useful for modifying the behavior on a per-run basis.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,tools,body,array<AssistantToolsCode | AssistantToolsFileSearch | AssistantToolsFunction>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Override the tools the assistant can use for this run. This is useful for modifying the behavior on a per-run basis.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,tool_resources,body,object,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A set of resources that are used by the assistant's tools. The resources are specific to the type of tool. For example, the `code_interpreter` tool requires a list of file IDs, while the `file_search` tool requires a list of vector store IDs.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,temperature,body,number,false,1,0,2,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,top_p,body,number,false,1,0,1,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

We generally recommend altering this or temperature but not both.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,stream,body,boolean,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"If `true`, returns a stream of events that happen during the Run as server-sent events, terminating when the Run enters a terminal state with a `data: [DONE]` message.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,max_prompt_tokens,body,integer,false,,256,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified, the run will end with status `incomplete`. See `incomplete_details` for more info.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,max_completion_tokens,body,integer,false,,256,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of completion tokens specified, the run will end with status `incomplete`. See `incomplete_details` for more info.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,truncation_strategy,body,TruncationObject & any,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,tool_choice,body,AssistantsApiToolChoiceOption & any,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,parallel_tool_calls,body,ParallelToolCalls,false,true,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Whether to enable [parallel function calling](https://developers.openai.com/api/docs/guides/function-calling#parallel-function-calling) during tool use.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/runs,response_format,body,AssistantsApiResponseFormatOption,false,,,,auto,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"Specifies the format that the model must output. Compatible with [GPT-4o](https://developers.openai.com/api/docs/models/gpt-4o), [GPT-4 Turbo](https://developers.openai.com/api/docs/models/gpt-4-turbo), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

Setting to `{ ""type"": ""json_schema"", ""json_schema"": {...} }` enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs).

Setting to `{ ""type"": ""json_object"" }` enables JSON mode, which ensures the message the model generates is valid JSON.

**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly ""stuck"" request. Also note that the message content may be partially cut off if `finish_reason=""length""`, which indicates the generation exceeded `max_tokens` or the conversation exceeded the max context length.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to retrieve.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to modify. Only the `metadata` can be modified.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id},tool_resources,body,object | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id},metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,DELETE /v1/threads/{thread_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to delete.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) the messages belong to.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,limit,query,integer,false,20,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,order,query,string,false,desc,,,asc|desc,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,after,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,before,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages,run_id,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Filter messages by the run ID that generated them.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) to create a message for.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages,role,body,string,true,,,,user|assistant,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The role of the entity that is creating the message. Allowed values include:
- `user`: Indicates the message is sent by an actual user and should be used in most cases to represent user-generated messages.
- `assistant`: Indicates the message is generated by the assistant. Use this value to insert messages from the assistant into the conversation.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages,content,body,string | array<MessageContentImageFileObject | MessageContentImageUrlObject | MessageRequestContentTextObject>,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages,attachments,body,array<object> | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages,metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages/{message_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) to which this message belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/messages/{message_id},message_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the message to retrieve.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages/{message_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to which this message belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages/{message_id},message_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the message to modify.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/messages/{message_id},metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,DELETE /v1/threads/{thread_id}/messages/{message_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to which this message belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,DELETE /v1/threads/{thread_id}/messages/{message_id},message_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the message to delete.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread the run belongs to.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs,limit,query,integer,false,20,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs,order,query,string,false,desc,,,asc|desc,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs,after,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs,before,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to run.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,include[],query,array<string>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A list of additional fields to include in the response. Currently the only supported value is `step_details.tool_calls[*].file_search.results[*].content` to fetch the file search result content.

See the [file search tool documentation](https://developers.openai.com/api/docs/guides/tools-file-search#retrieval-customization) for more information.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,assistant_id,body,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [assistant](https://developers.openai.com/api/docs/assistants/migration) to use to execute this run.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,model,body,string | AssistantSupportedModels,false,,,,gpt-5|gpt-5-mini|gpt-5-nano|gpt-5-2025-08-07|gpt-5-mini-2025-08-07|gpt-5-nano-2025-08-07|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4.1-2025-04-14|gpt-4.1-mini-2025-04-14|gpt-4.1-nano-2025-04-14|o3-mini|o3-mini-2025-01-31|o1|o1-2024-12-17|gpt-4o|gpt-4o-2024-11-20|gpt-4o-2024-08-06|gpt-4o-2024-05-13|gpt-4o-mini|gpt-4o-mini-2024-07-18|gpt-4.5-preview|gpt-4.5-preview-2025-02-27|gpt-4-turbo|gpt-4-turbo-2024-04-09|gpt-4-0125-preview|gpt-4-turbo-preview|gpt-4-1106-preview|gpt-4-vision-preview|gpt-4|gpt-4-0314|gpt-4-0613|gpt-4-32k|gpt-4-32k-0314|gpt-4-32k-0613|gpt-3.5-turbo|gpt-3.5-turbo-16k|gpt-3.5-turbo-0613|gpt-3.5-turbo-1106|gpt-3.5-turbo-0125|gpt-3.5-turbo-16k-0613,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The ID of the [Model](https://developers.openai.com/api/reference/resources/models) to be used to execute this run. If a value is provided here, it will override the model associated with the assistant. If not, the model associated with the assistant will be used.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,reasoning_effort,body,ReasoningEffort,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,instructions,body,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Overrides the [instructions](https://developers.openai.com/api/docs/assistants/migration) of the assistant. This is useful for modifying the behavior on a per-run basis.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,additional_instructions,body,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Appends additional instructions at the end of the instructions for the run. This is useful for modifying the behavior on a per-run basis without overriding other instructions.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,additional_messages,body,array<CreateMessageRequest>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Adds additional messages to the thread before creating the run.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,tools,body,array<AssistantToolsCode | AssistantToolsFileSearch | AssistantToolsFunction>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Override the tools the assistant can use for this run. This is useful for modifying the behavior on a per-run basis.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,temperature,body,number,false,1,0,2,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,top_p,body,number,false,1,0,1,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.

We generally recommend altering this or temperature but not both.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,stream,body,boolean,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"If `true`, returns a stream of events that happen during the Run as server-sent events, terminating when the Run enters a terminal state with a `data: [DONE]` message.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,max_prompt_tokens,body,integer,false,,256,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The maximum number of prompt tokens that may be used over the course of the run. The run will make a best effort to use only the number of prompt tokens specified, across multiple turns of the run. If the run exceeds the number of prompt tokens specified, the run will end with status `incomplete`. See `incomplete_details` for more info.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,max_completion_tokens,body,integer,false,,256,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"The maximum number of completion tokens that may be used over the course of the run. The run will make a best effort to use only the number of completion tokens specified, across multiple turns of the run. If the run exceeds the number of completion tokens specified, the run will end with status `incomplete`. See `incomplete_details` for more info.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,truncation_strategy,body,TruncationObject & any,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,tool_choice,body,AssistantsApiToolChoiceOption & any,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,parallel_tool_calls,body,ParallelToolCalls,false,true,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Whether to enable [parallel function calling](https://developers.openai.com/api/docs/guides/function-calling#parallel-function-calling) during tool use.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs,response_format,body,AssistantsApiResponseFormatOption,false,,,,auto,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"Specifies the format that the model must output. Compatible with [GPT-4o](https://developers.openai.com/api/docs/models/gpt-4o), [GPT-4 Turbo](https://developers.openai.com/api/docs/models/gpt-4-turbo), and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.

Setting to `{ ""type"": ""json_schema"", ""json_schema"": {...} }` enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs).

Setting to `{ ""type"": ""json_object"" }` enables JSON mode, which ensures the message the model generates is valid JSON.

**Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly ""stuck"" request. Also note that the message content may be partially cut off if `finish_reason=""length""`, which indicates the generation exceeded `max_tokens` or the conversation exceeded the max context length.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) that was run.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id},run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run to retrieve.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) that was run.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id},run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run to modify.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id},metadata,body,Metadata,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/cancel,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to which this run belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/cancel,run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run to cancel.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread the run and run steps belong to.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run the run steps belong to.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,limit,query,integer,false,20,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,order,query,string,false,desc,,,asc|desc,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,after,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,before,query,string,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps,include[],query,array<string>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A list of additional fields to include in the response. Currently the only supported value is `step_details.tool_calls[*].file_search.results[*].content` to fetch the file search result content.

See the [file search tool documentation](https://developers.openai.com/api/docs/guides/tools-file-search#retrieval-customization) for more information.",https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps/{step_id},thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the thread to which the run and run step belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps/{step_id},run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run to which the run step belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps/{step_id},step_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run step to retrieve.,https://developers.openai.com/api/docs/assistants/migration
openai,GET /v1/threads/{thread_id}/runs/{run_id}/steps/{step_id},include[],query,array<string>,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,"A list of additional fields to include in the response. Currently the only supported value is `step_details.tool_calls[*].file_search.results[*].content` to fetch the file search result content.

See the [file search tool documentation](https://developers.openai.com/api/docs/guides/tools-file-search#retrieval-customization) for more information.",https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/submit_tool_outputs,thread_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the [thread](https://developers.openai.com/api/docs/assistants/migration) to which this run belongs.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/submit_tool_outputs,run_id,path,string,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,The ID of the run that requires the tool output submission.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/submit_tool_outputs,tool_outputs,body,array<object>,true,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,A list of tools for which the outputs are being submitted.,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/threads/{thread_id}/runs/{run_id}/submit_tool_outputs,stream,body,boolean | null,false,,,,,,OpenAI-Beta: assistants=v2 (historical),RETIRED|DOCUMENTED,,https://developers.openai.com/api/docs/assistants/migration
openai,POST /v1/chatkit/sessions/{session_id}/cancel,session_id,path,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Unique identifier for the ChatKit session to cancel.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow,body,WorkflowParam,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Workflow that powers the session.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow.id,body,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Identifier for the workflow invoked by the session.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow.version,body,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Specific workflow version to run. Defaults to the latest deployed version.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow.state_variables,body,object,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"State variables forwarded to the workflow. Keys may be up to 64 characters, values must be primitive types, and the map defaults to an empty object.",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow.tracing,body,WorkflowTracingParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Optional tracing overrides for the workflow invocation. When omitted, tracing is enabled by default.",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,workflow.tracing.enabled,body,boolean,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Whether tracing is enabled during the session. Defaults to true.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,user,body,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,A free-form string that identifies your end user; ensures this Session can access other objects that have the same `user` scope.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,expires_after,body,ExpiresAfterParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Optional override for session expiration timing in seconds from creation. Defaults to 10 minutes.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,expires_after.anchor,body,string,true,created_at,,,created_at,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Base timestamp used to calculate expiration. Currently fixed to `created_at`.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,expires_after.seconds,body,integer,true,,1,600,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Number of seconds after the anchor when the session expires.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,rate_limits,body,RateLimitsParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Optional override for per-minute request limits. When omitted, defaults to 10.",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,rate_limits.max_requests_per_1_minute,body,integer,false,,1,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Maximum number of requests allowed per minute for the session. Defaults to 10.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration,body,ChatkitConfigurationParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Optional overrides for ChatKit runtime configuration features,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.automatic_thread_titling,body,AutomaticThreadTitlingParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Configuration for automatic thread titling. When omitted, automatic thread titling is enabled by default.",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.automatic_thread_titling.enabled,body,boolean,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Enable automatic thread title generation. Defaults to true.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.file_upload,body,FileUploadParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Configuration for upload enablement and limits. When omitted, uploads are disabled by default (max_files 10, max_file_size 512 MB).",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.file_upload.enabled,body,boolean,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Enable uploads for this session. Defaults to false.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.file_upload.max_file_size,body,integer,false,,1,512,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Maximum size in megabytes for each uploaded file. Defaults to 512 MB, which is the maximum allowable size.",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.file_upload.max_files,body,integer,false,,1,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Maximum number of files that can be uploaded to the session. Defaults to 10.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.history,body,HistoryParam,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,"Configuration for chat history retention. When omitted, history is enabled by default with no limit on recent_threads (null).",https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.history.enabled,body,boolean,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Enables chat users to access previous ChatKit threads. Defaults to true.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,POST /v1/chatkit/sessions,chatkit_configuration.history.recent_threads,body,integer,false,,1,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Number of recent ChatKit threads users have access to. Defaults to unlimited when unset.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id}/items,thread_id,path,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Identifier of the ChatKit thread whose items are requested.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id}/items,limit,query,integer,false,,0,100,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Maximum number of thread items to return. Defaults to 20.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id}/items,order,query,OrderEnum,false,,,,asc|desc,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Sort order for results by creation time. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id}/items,after,query,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,List items created after this thread item ID. Defaults to null for the first page.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id}/items,before,query,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,List items created before this thread item ID. Defaults to null for the newest results.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads/{thread_id},thread_id,path,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Identifier of the ChatKit thread to retrieve.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,DELETE /v1/chatkit/threads/{thread_id},thread_id,path,string,true,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Identifier of the ChatKit thread to delete.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads,limit,query,integer,false,,0,100,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Maximum number of thread items to return. Defaults to 20.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads,order,query,OrderEnum,false,,,,asc|desc,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Sort order for results by creation time. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads,after,query,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,List items created after this thread item ID. Defaults to null for the first page.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads,before,query,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,List items created before this thread item ID. Defaults to null for the newest results.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/chatkit/threads,user,query,string,false,,,,,,OpenAI-Beta: chatkit_beta=v1,DOCUMENTED|BETA,Filter threads that belong to this user identifier. Defaults to null to return all users.,https://developers.openai.com/api/reference/resources/beta/subresources/chatkit
openai,GET /v1/agents/environments/{environment_id},environment_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/{environment_id}/files,environment_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/{environment_id}/files,path,query,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Restrict the listing to this absolute workspace directory.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/{environment_id}/files,limit,query,integer | null,false,,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The maximum number of files to return, between 1 and 100.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/{environment_id}/files,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Sort by case-sensitive path components. Defaults to descending.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/{environment_id}/files,page,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The opaque token from the previous page. Keep the same path, order, and limit.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/{environment_id}/files,environment_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/{environment_id}/files,type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/{environment_id}/files,file_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when type=file_id] The ID of the uploaded file.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/{environment_id}/files,path,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when type=inline] The absolute destination path inside `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/{environment_id}/files,data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when type=inline] The standard-base64-encoded file contents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id},subagent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the subagent in this session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/items,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/items,subagent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the subagent in this session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/items,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/items,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/items,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns,subagent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the subagent in this session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id},subagent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the subagent in this session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id},turn_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of a turn belonging to this subagent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,subagent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the subagent in this session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,turn_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of a turn belonging to this subagent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/subagents/{subagent_id}/turns/{turn_id}/items,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents,limit,query,integer | null,false,,1,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The maximum number of resources to return.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Up to 16 string key-value pairs, with keys up to 64 and values up to 512 characters. Omission or null defaults to an empty map.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,name,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,A human-readable name for the agent. Omission or null leaves the agent unnamed.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,model,body,string,true,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The model to use for the agent. The requested model name is preserved.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,reasoning,body,ReasoningParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for model reasoning. Omission uses the model's default effort.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,reasoning.effort,body,ReasoningEffortParam | null,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The amount of reasoning effort the model should use. Omission lets the model select it.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,reasoning.summary,body,ReasoningSummaryParam | null,false,,,,concise|detailed|auto,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Controls whether the response includes a reasoning summary.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,text,body,TextParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for generated text. Defaults to the `text` format and medium verbosity.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,text.format,body,TextFormatParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The output format. Omission uses ordinary text (`{""type"": ""text""}`).",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,text.verbosity,body,VerbosityParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The amount of text the model should produce. Defaults to `medium`, matching Responses.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,service_tier,body,ServiceTierParam | null,false,,,,auto|default|flex|priority|fast,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The service tier used for model requests. Defaults to `auto`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,instructions,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Additional instructions appended to the agent's default base instructions. Omit or set to null to add no custom instructions.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Tools available to the agent. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].type,body,string,true,web_search,,,web_search,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] The type of the object. Always `web_search`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=function] The name of the function.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=function] A description of what the function does.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].parameters,body,object,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=function] A JSON Schema object describing the function's arguments.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].defer_loading,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=function] Whether this function is deferred and discovered through tool search. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].enabled,body,boolean,false,true,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=programmatic_tool_calling] Whether tools can be called from model-generated code. Defaults to `true`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].server_label,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] A label used to identify the MCP server in tool calls.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].credential_id,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] The vault credential selected for this MCP server. Optional when exactly one attached credential matches the server URL.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport,body,PersistedMcpTransportConfigParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] The credential-free transport used to connect to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.type,body,string,true,stdio,,,stdio,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=stdio] The type of the object. Always `stdio`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.server_url,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=http] The URL of the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.headers,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=http] Non-secret HTTP headers sent to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=stdio] The command used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.args,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=stdio] Arguments passed to the MCP server command.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.cwd,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=stdio] The working directory used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].transport.env_vars,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].transport.type=stdio] Environment variable names to inherit from the selected execution environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].request_metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] Metadata included with requests to this MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].allowed_tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] The MCP tools the agent may call. All server tools are allowed when omitted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].required,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] Whether this MCP server must initialize before the first turn. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].connection_origin,body,McpConnectionOriginParam | null,false,,,,service|environment,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=mcp] Selects where outbound MCP HTTP connections originate.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].mode,body,WebSearchModeParam | null,false,,,,disabled|cached|live,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] The source used for web search results. Defaults to `live`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].context_size,body,WebSearchContextSizeParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] The amount of search context made available to the model. Defaults to `medium`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] Domains the search may include.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].location,body,WebSearchLocationParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] Approximate location used to localize search results.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].location.country,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when tools[].type=web_search] The two-letter ISO country code, such as `US`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].location.region,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] The region or state name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].location.city,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when tools[].type=web_search] The city name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,tools[].location.timezone,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when tools[].type=web_search] The IANA timezone, such as `America/Los_Angeles`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,multi_agent,body,MultiAgentConfigCurrentParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for creating and coordinating subagents. Subagent tools are disabled by default.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,multi_agent.enabled,body,boolean,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Whether subagent tools are enabled.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents,multi_agent.max_concurrent_subagents,body,integer,false,,1,4294967295,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Maximum number of subagents that may run concurrently. Defaults to 6.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/{agent_id},agent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the reusable agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},agent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the reusable agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},model,body,string,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The model to use for the agent. The requested model name is preserved.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},reasoning,body,ReasoningParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Configuration for model reasoning. Omit to keep the current settings; pass `null` to reset to the model's default effort.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},reasoning.effort,body,ReasoningEffortParam | null,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The amount of reasoning effort the model should use. Omission lets the model select it.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},reasoning.summary,body,ReasoningSummaryParam | null,false,,,,concise|detailed|auto,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Controls whether the response includes a reasoning summary.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},text,body,TextParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Configuration for text generated by the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},text.format,body,TextFormatParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The output format. Omission uses ordinary text (`{""type"": ""text""}`).",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},text.verbosity,body,VerbosityParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"The amount of text the model should produce. Defaults to `medium`, matching Responses.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},service_tier,body,ServiceTierParam | null,false,,,,auto|default|flex|priority|fast,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The service tier used for model requests.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},instructions,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Additional instructions appended to the agent's default base instructions. Omit to leave unchanged.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},multi_agent,body,MultiAgentConfigCurrentParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Configuration for creating and coordinating subagents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},multi_agent.enabled,body,boolean,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Whether subagent tools are enabled.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},multi_agent.max_concurrent_subagents,body,integer,false,,1,4294967295,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Maximum number of subagents that may run concurrently. Defaults to 6.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"Replaces all metadata. Omit to leave unchanged, or pass null or {} to clear it. Up to 16 string key-value pairs, with keys up to 64 and values up to 512 characters.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},name,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"A replacement name. Omit to leave unchanged, or pass null to clear it.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Tools available to the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].type,body,string,true,web_search,,,web_search,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] The type of the object. Always `web_search`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=function] The name of the function.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=function] A description of what the function does.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].parameters,body,object,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=function] A JSON Schema object describing the function's arguments.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].defer_loading,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=function] Whether this function is deferred and discovered through tool search. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].enabled,body,boolean,false,true,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=programmatic_tool_calling] Whether tools can be called from model-generated code. Defaults to `true`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].server_label,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] A label used to identify the MCP server in tool calls.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].credential_id,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] The vault credential selected for this MCP server. Optional when exactly one attached credential matches the server URL.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport,body,PersistedMcpTransportConfigParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] The credential-free transport used to connect to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.type,body,string,true,stdio,,,stdio,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=stdio] The type of the object. Always `stdio`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.server_url,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=http] The URL of the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.headers,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=http] Non-secret HTTP headers sent to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=stdio] The command used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.args,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=stdio] Arguments passed to the MCP server command.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.cwd,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=stdio] The working directory used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].transport.env_vars,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].transport.type=stdio] Environment variable names to inherit from the selected execution environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].request_metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] Metadata included with requests to this MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].allowed_tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] The MCP tools the agent may call. All server tools are allowed when omitted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].required,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] Whether this MCP server must initialize before the first turn. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].connection_origin,body,McpConnectionOriginParam | null,false,,,,service|environment,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=mcp] Selects where outbound MCP HTTP connections originate.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].mode,body,WebSearchModeParam | null,false,,,,disabled|cached|live,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] The source used for web search results. Defaults to `live`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].context_size,body,WebSearchContextSizeParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] The amount of search context made available to the model. Defaults to `medium`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] Domains the search may include.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].location,body,WebSearchLocationParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] Approximate location used to localize search results.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].location.country,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when tools[].type=web_search] The two-letter ISO country code, such as `US`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].location.region,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] The region or state name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].location.city,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when tools[].type=web_search] The city name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/{agent_id},tools[].location.timezone,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when tools[].type=web_search] The IANA timezone, such as `America/Los_Angeles`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,DELETE /v1/agents/{agent_id},agent_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the reusable agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/templates,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/templates,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/templates,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,packages,body,EnvironmentPackagesParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Packages to install in the environment. Defaults to empty package lists.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,packages.python,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Python packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,packages.system,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,System packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,packages.npm,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,npm packages to install globally. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,setup_commands,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Ordered, confidential setup commands. Command bodies are never returned.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,setup_commands[].command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The shell command to execute.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,setup_commands[].cwd,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The absolute working directory. Defaults to `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,network,body,NetworkPolicyParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Network access policy for the environment. Defaults to disabled for GA requests and enabled for alpha/beta requests.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,network.access,body,NetworkAccessParam,true,,,,enabled|disabled|restricted,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The environment's network access mode.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,network.allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Domains the environment may access when network access is restricted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,env,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Environment variables made available to the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,capability_directories,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Directories that contain capabilities exposed to the agent. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Skills referenced by ID or provided as inline ZIP archives. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].skill_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=skill_reference] The ID of the skill created through `/v1/skills`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].version,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when skills[].type=skill_reference] The skill version, a positive integer or `latest`; omission selects the default.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The skill name declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The skill description declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when skills[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,skills[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Plugins provided as inline ZIP archives. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The plugin name declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The plugin description declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when plugins[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,plugins[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,files,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Files available before the agent starts. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,files[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,files[].file_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=file_id] The ID of the uploaded file.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,files[].path,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The absolute destination path inside `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,files[].data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The standard-base64-encoded file contents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates,name,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,An optional human-readable display name for the template.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/environments/templates/{environment_template_id},environment_template_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the reusable environment template.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},environment_template_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the reusable environment template.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},name,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"A replacement human-readable display name, or `null` to clear the name.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},packages,body,EnvironmentPackagesParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Packages installed before the runtime network policy applies.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},packages.python,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Python packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},packages.system,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,System packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},packages.npm,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,npm packages to install globally. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},setup_commands,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Replacement confidential setup commands, never included in returned resources.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},setup_commands[].command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The shell command to execute.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},setup_commands[].cwd,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The absolute working directory. Defaults to `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},network,body,NetworkPolicyParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Network access available after setup completes. Omit to preserve the current policy, or pass `null` to reset to disabled for GA requests or enabled for alpha/beta requests.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},network.access,body,NetworkAccessParam,true,,,,enabled|disabled|restricted,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The environment's network access mode.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},network.allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Domains the environment may access when network access is restricted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},env,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Replacement confidential environment values.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},capability_directories,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Directories that expose capabilities to the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Replacement skill configuration installed for each new session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].skill_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=skill_reference] The ID of the skill created through `/v1/skills`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].version,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when skills[].type=skill_reference] The skill version, a positive integer or `latest`; omission selects the default.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The skill name declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The skill description declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when skills[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},skills[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when skills[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Replacement plugin configuration installed for each new session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The plugin name declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The plugin description declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when plugins[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},plugins[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when plugins[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},files,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Replacement file configuration materialized for each new session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},files[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},files[].file_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=file_id] The ID of the uploaded file.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},files[].path,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The absolute destination path inside `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/environments/templates/{environment_template_id},files[].data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when files[].type=inline] The standard-base64-encoded file contents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,DELETE /v1/agents/environments/templates/{environment_template_id},environment_template_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the reusable environment template.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions,limit,query,integer | null,false,,1,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The maximum number of resources to return.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Sort order by the `created_at` timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions,agent_id,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Only return sessions whose root agent has this ID. Omit to return sessions for all agents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Up to 16 string key-value pairs, with keys up to 64 and values up to 512 characters. Omission or null defaults to an empty map.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent,body,SessionAgentConfigParam,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Agent configuration. With `agent_id`, supplied fields override the saved agent for this session. Without `agent_id`, `model` is required.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.model,body,string,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The model to use for the agent. The requested model name is preserved.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.reasoning,body,ReasoningParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for model reasoning. Omit to keep the current settings; pass `null` to reset to the model's default effort.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.reasoning.effort,body,ReasoningEffortParam | null,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The amount of reasoning effort the model should use. Omission lets the model select it.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.reasoning.summary,body,ReasoningSummaryParam | null,false,,,,concise|detailed|auto,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Controls whether the response includes a reasoning summary.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.text,body,TextParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for text generated by the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.text.format,body,TextFormatParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The output format. Omission uses ordinary text (`{""type"": ""text""}`).",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.text.verbosity,body,VerbosityParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The amount of text the model should produce. Defaults to `medium`, matching Responses.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.service_tier,body,ServiceTierParam | null,false,,,,auto|default|flex|priority|fast,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The service tier used for model requests.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.instructions,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Additional instructions appended to the agent's default base instructions. Omit to leave unchanged.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.multi_agent,body,MultiAgentConfigCurrentParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Configuration for creating and coordinating subagents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.multi_agent.enabled,body,boolean,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Whether subagent tools are enabled.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.multi_agent.max_concurrent_subagents,body,integer,false,,1,4294967295,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Maximum number of subagents that may run concurrently. Defaults to 6.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Tools available to the agent. Omit to inherit, or pass null to clear them.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].type,body,string,true,web_search,,,web_search,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] The type of the object. Always `web_search`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=function] The name of the function.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=function] A description of what the function does.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].parameters,body,object,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=function] A JSON Schema object describing the function's arguments.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].defer_loading,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=function] Whether this function is deferred and discovered through tool search. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].enabled,body,boolean,false,true,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=programmatic_tool_calling] Whether tools can be called from model-generated code. Defaults to `true`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].server_label,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] A label used to identify the MCP server in tool calls.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].credential_id,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] The attached vault credential used to authenticate this MCP server. Optional when exactly one attached credential matches the server URL.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport,body,McpTransportConfigParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] The transport used to connect to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.type,body,string,true,stdio,,,stdio,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] The type of the object. Always `stdio`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.server_url,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=http] The URL of the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.authorization,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when agent.tools[].transport.type=http] The authorization value sent to the MCP server, if any.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.headers,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=http] Additional HTTP headers sent to the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] The command used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.args,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] Arguments passed to the MCP server command.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.cwd,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] The working directory used to start the MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.env,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] Environment variables set for the MCP server process.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].transport.env_vars,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].transport.type=stdio] Environment variable names to inherit from the selected execution environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].request_metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] Metadata included with requests to this MCP server.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].allowed_tools,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] The MCP tools the agent may call. All server tools are allowed when omitted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].required,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] Whether this MCP server must initialize before the first turn. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].connection_origin,body,McpConnectionOriginParam | null,false,,,,service|environment,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=mcp] Selects where outbound MCP HTTP connections originate. Omitted or `service` uses the Managed Agents service network; `environment` uses the session's selected environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].mode,body,WebSearchModeParam | null,false,,,,disabled|cached|live,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] The source used for web search results. Defaults to `live`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].context_size,body,WebSearchContextSizeParam | null,false,,,,low|medium|high,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] The amount of search context made available to the model. Defaults to `medium`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] Domains the search may include.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].location,body,WebSearchLocationParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] Approximate location used to localize search results.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].location.country,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when agent.tools[].type=web_search] The two-letter ISO country code, such as `US`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].location.region,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] The region or state name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].location.city,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when agent.tools[].type=web_search] The city name.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent.tools[].location.timezone,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when agent.tools[].type=web_search] The IANA timezone, such as `America/Los_Angeles`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,agent_id,body,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of a saved reusable agent. Omit `agent` to use its configuration unchanged.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment,body,EnvironmentParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,An inline execution environment or a reference to an environment template.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.type,body,string,true,self_hosted,,,self_hosted,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=self_hosted] The type of the object. Always `self_hosted`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.packages,body,EnvironmentPackagesParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Packages to install in the environment. Defaults to empty package lists.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.packages.python,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Python packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.packages.system,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] System packages to install. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.packages.npm,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] npm packages to install globally. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.setup_commands,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when environment.type=openai_hosted] Ordered, confidential setup commands. Command bodies are never returned.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.setup_commands[].command,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] The shell command to execute.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.setup_commands[].cwd,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] The absolute working directory. Defaults to `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.network,body,NetworkPolicyParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Network access policy for the environment. Defaults to disabled for GA requests and enabled for alpha/beta requests.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.network.access,body,NetworkAccessParam,true,,,,enabled|disabled|restricted,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] The environment's network access mode.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.network.allowed_domains,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Domains the environment may access when network access is restricted.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.env,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Environment variables made available to the agent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.capability_directories,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=self_hosted] Directories that contain capabilities exposed to the agent. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Skills referenced by ID or provided as inline ZIP archives. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].skill_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].type=skill_reference] The ID of the skill created through `/v1/skills`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].version,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when environment.skills[].type=skill_reference] The skill version, a positive integer or `latest`; omission selects the default.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].type=inline] The skill name declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].type=inline] The skill description declared in `SKILL.md`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when environment.skills[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.skills[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.skills[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Plugins provided as inline ZIP archives. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].type=inline] The plugin name declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].description,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].type=inline] The plugin description declared in `.codex-plugin/plugin.json`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].source,body,InlineCapabilitySourceParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].type=inline] The inline ZIP archive.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].source.type,body,string,true,base64,,,base64,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].source.type=base64] The type of the object. Always `base64`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].source.media_type,body,string,true,application/zip,,,application/zip,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"[when environment.plugins[].source.type=base64] The archive media type, always `application/zip`.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.plugins[].source.data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.plugins[].source.type=base64] Standard-base64 encoded ZIP archive bytes.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.files,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] Files available before the agent starts. Defaults to an empty list.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.files[].type,body,string,true,inline,,,inline,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.files[].type=inline] The type of the object. Always `inline`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.files[].file_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.files[].type=file_id] The ID of the uploaded file.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.files[].path,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.files[].type=inline] The absolute destination path inside `/workspace`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.files[].data,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.files[].type=inline] The standard-base64-encoded file contents.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.environment_template_id,body,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=openai_hosted] A reusable hosted template applied before inline session configuration. Omitted fields inherit the template; network overrides cannot broaden its policy.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,environment.workspace_directory,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when environment.type=self_hosted] Absolute project directory inside the self-hosted environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,vault_ids,body,array | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The IDs of vaults made available to the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,input,body,CreateSessionInputParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Initial input to submit when the session is created. A string is shorthand for a single user message. Required when `environment.type` is `none`, or when `stream` is `true` for an environment that is not `self_hosted`; optional for self-hosted and non-streaming execution environments.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions,stream,body,boolean,false,false,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Whether to stream session events as server-sent events. Defaults to `false`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},agent,body,UpdateSessionAgentParam,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Model settings for subsequent turns. Omitted fields stay unchanged.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},agent.model,body,string,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna (live-verified),OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The model for subsequent turns. Omit to keep the current model.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},agent.reasoning,body,UpdateSessionReasoningParam,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Reasoning settings to update. Omit to keep the current effort.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},agent.reasoning.effort,body,ReasoningEffortParam | null,false,,,,none|minimal|low|medium|high|xhigh|max,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Omit to keep the current effort. Null selects the model's default effort.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},agent.service_tier,body,ServiceTierParam | null,false,,,,auto|default|flex|priority|fast,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Omit to keep the current tier. Null resets it to auto.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id},metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"Replaces all metadata. Omit to leave unchanged, or pass null or {} to clear it. Up to 16 string key-value pairs, with keys up to 64 and values up to 512 characters.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,DELETE /v1/agents/sessions/{session_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Sort by creation time and ID. Defaults to descending.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts,environment_id,query,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Restrict the listing to artifacts produced by this environment.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts,limit,query,integer | null,false,,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The maximum number of artifacts to return, between 1 and 100.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts,after,query,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return artifacts after this immutable artifact ID.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts/{artifact_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session that owns the artifact.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts/{artifact_id},artifact_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The immutable session artifact ID.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,DELETE /v1/agents/sessions/{session_id}/artifacts/{artifact_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session that owns the artifact.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,DELETE /v1/agents/sessions/{session_id}/artifacts/{artifact_id},artifact_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The immutable session artifact ID.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts/{artifact_id}/content,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the session that owns the artifact.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/artifacts/{artifact_id}/content,artifact_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The immutable session artifact ID.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/events,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,Idempotency-Key,header,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,An optional client-generated key that makes retries of submitted messages idempotent.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events,body,array<SessionInputParam>,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The input events to submit to the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].type,body,string,true,agent.session.input.tool_result,,,agent.session.input.tool_result,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] The type of the object. Always `agent.session.input.tool_result`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input,body,array<InputMessageParam>,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.message] The user messages to add to the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].type,body,string,true,message,,,message,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.message] The type of the input item. Always `message`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].role,body,string,true,user,,,user,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.message] The role of the message author. Always `user`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].content,body,array<InputContentParam>,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.message] The content of the message.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].content[].type,body,string,true,input_image,,,input_image,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].input[].content[].type=input_image] The type of the object. Always `input_image`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].content[].text,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].input[].content[].type=input_text] The text sent to the model.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].input[].content[].image_url,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].input[].content[].type=input_image] The URL of the image sent to the model.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].turn_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] The ID of the turn that requested the function call.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].call_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] The ID of the function call.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].success,body,boolean,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] Whether the function call succeeded.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].output,body,FunctionCallOutputParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] The function result when the call succeeded.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,POST /v1/agents/sessions/{session_id}/events,events[].error,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,[when events[].type=agent.session.input.tool_result] The error message when the call failed.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/items,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/items,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/items,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/items,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns,session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns,limit,query,integer,false,20,1,100,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,"The maximum number of resources to return, between 1 and 100. Defaults to 20.",https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The order in which resources are returned. Defaults to `desc`.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns/{turn_id},session_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the session that owns the turn.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/agents/sessions/{session_id}/turns/{turn_id},turn_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA|LIVE_VERIFIED,The ID of the turn.,https://developers.openai.com/api/reference/resources/beta/subresources/agents
openai,GET /v1/vaults,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Sort order by the `created_at` timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `desc`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults,limit,query,integer | null,false,,0,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The maximum number of resources to return. Defaults to 20. Values are clamped between 1 and 100.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults,status,query,VaultStatusFilterParam,false,,,,active|archived,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"Filter by one status or a list, such as `status=active` or `status[]=active&status[]=archived`. Both statuses are included by default.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults,name,body,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The name is trimmed before storage. It must contain 1 to 256 UTF-8 bytes after trimming.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults,metadata,body,object | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"Key-value pairs to associate with the vault, such as an application or team identifier.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id},vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,DELETE /v1/vaults/{vault_id},vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials,vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials,order,query,ListOrderParam,false,,,,asc|desc,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Sort order by the `created_at` timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `desc`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials,limit,query,integer | null,false,,0,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The maximum number of resources to return. Defaults to 20. Values are clamped between 1 and 100.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials,status,query,VaultStatusFilterParam,false,,,,active|archived,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"Filter by one status or a list, such as `status=active` or `status[]=active&status[]=archived`. Both statuses are included by default.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials,after,query,string,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Return resources after this resource ID in the selected order.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The name is trimmed before storage. It must contain 1 to 256 UTF-8 bytes after trimming.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth,body,CreateVaultCredentialAuthParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The authentication method and write-only secret values to store.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.type,body,string,true,environment_variable,,,environment_variable,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=environment_variable] The type of the object. Always `environment_variable`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.mcp_server_url,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=static_bearer] The HTTPS MCP server URL authorized by this credential.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.access_token,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] A write-only OAuth access token; never returned by credential resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.expires_at,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=mcp_oauth] When the OAuth access token expires, as an RFC 3339 timestamp, if known.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh,body,CreateMcpOauthRefreshParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] Optional refresh configuration for an HTTPS OAuth token endpoint.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] The HTTPS OAuth token endpoint used to exchange the refresh token for a new access token.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.client_id,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] The OAuth client ID used when requesting a new access token.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.resource,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=mcp_oauth] The resource URI to send to the OAuth token endpoint during refresh, if required.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.scope,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=mcp_oauth] Space-separated OAuth scopes to request during refresh, if required.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.refresh_token,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] The refresh token to store. This secret is never returned in credential resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth,body,CreateMcpOauthTokenEndpointAuthParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] How the OAuth client authenticates to the token endpoint.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.type,body,string,true,client_secret_post,,,client_secret_post,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.refresh.token_endpoint_auth.type=client_secret_post] The type of the object. Always `client_secret_post`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.refresh.token_endpoint_auth.client_secret,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.refresh.token_endpoint_auth.type=client_secret_post] The OAuth client secret to store. Never returned in credential resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.token,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=static_bearer] The bearer token to store. This secret is never returned in credential resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.secret_name,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=environment_variable] The environment variable name that receives the placeholder, such as `SERVICE_API_KEY`. Use ASCII letters, digits, and underscores, starting with a letter or underscore. Names starting with `CODEX_` and managed proxy or certificate variable names are reserved.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.secret_value,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=environment_variable] The write-only secret to store. Never returned in credential resources or supplied directly to sandbox code. Must be nonempty and must not contain carriage returns, newlines, or NUL bytes.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.networking,body,VaultCredentialNetworkingParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=environment_variable] The destinations where the proxy can substitute this secret. The environment network policy must also allow them.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.networking.type,body,string,true,limited,,,limited,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.networking.type=limited] The type of the object. Always `limited`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials,auth.networking.allowed_hosts,body,array<string>,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.networking.type=limited] The 1 to 16 distinct allowed hostnames or IPv4 addresses, normalized to lowercase. Entries contain no scheme, path, port, or wildcard. IPv6 addresses are not supported.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,GET /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault credential.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault credential.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth,body,RotateVaultCredentialAuthParam,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,Replacement values for the credential's existing authentication method.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.type,body,string,true,environment_variable,,,environment_variable,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=environment_variable] The type of the object. Always `environment_variable`.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.access_token,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] A write-only replacement OAuth access token.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.expires_at,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=mcp_oauth] The replacement expiry as an RFC 3339 timestamp, or `null` to clear it. Omitting this field preserves the expiry unless a new access token is supplied, in which case the expiry is cleared.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh,body,RotateMcpOauthRefreshParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] Optional write-only refresh-token and client-secret updates.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.refresh_token,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] The replacement refresh token. Omit or pass `null` to keep the stored token. This secret is never returned in resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.scope,body,string | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=mcp_oauth] Replacement space-separated OAuth scopes for refresh requests. Omit to keep the scopes, or pass `null` to stop sending a scope parameter.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.refresh.token_endpoint_auth,body,RotateMcpOauthTokenEndpointAuthParam | null,false,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=mcp_oauth] Client-secret updates for the existing token endpoint authentication method.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.token,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,[when auth.type=static_bearer] The replacement bearer token. This secret is never returned in credential resources.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,POST /v1/vaults/{vault_id}/credentials/{credential_id},auth.secret_value,body,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,"[when auth.type=environment_variable] The write-only replacement secret. Never returned in credential resources or supplied directly to sandbox code. Must be nonempty and must not contain carriage returns, newlines, or NUL bytes.",https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,DELETE /v1/vaults/{vault_id}/credentials/{credential_id},vault_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,DELETE /v1/vaults/{vault_id}/credentials/{credential_id},credential_id,path,string,true,,,,,,OpenAI-Beta: agents=v1,DOCUMENTED|BETA,The ID of the vault credential.,https://developers.openai.com/api/docs/guides/agents-api/tools/vaults
openai,all Agents/Vaults endpoints,OpenAI-Beta,header,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Must be `agents=v1`. Live: missing header -> 400 invalid_beta ""To access the Agents API, set the 'OpenAI-Beta' header to 'agents=v1'.""; `?beta=true` is not accepted.",https://developers.openai.com/api/docs/guides/agents-api/quickstart
openai,all ChatKit endpoints,OpenAI-Beta,header,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Must be `chatkit_beta=v1` (spec examples).,https://developers.openai.com/api/docs/guides/agents-api/quickstart
openai,GET /v1/agents/sessions/{session_id}/events,Accept,header,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,text/event-stream recommended for the SSE stream (with ?stream=true).,https://developers.openai.com/api/docs/guides/agents-api/quickstart
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,id,path,string,true,,,,,,,DOCUMENTED|UNVERIFIED,Public API trigger identifier of the published API channel (agtch_…).,https://developers.openai.com/workspace-agents/trigger-runs
openai,GET https://api.chatgpt.com/v1/workspace_agents/{id}/runs/{run_id},run_id,path,string,true,,,,,,OpenAI-Beta: workspace_agent_runs=v1,DOCUMENTED|UNVERIFIED,Trigger run id (apirun_…) returned when OpenAI-Beta: workspace_agent_runs=v1 is sent.,https://developers.openai.com/workspace-agents/trigger-runs
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,input,body,string,true,,,,,,,DOCUMENTED|UNVERIFIED,Message text passed to the agent as trigger input.,https://developers.openai.com/workspace-agents/trigger-runs
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,conversation_key,body,string,false,,,,,,,DOCUMENTED|UNVERIFIED,Caller-defined stable identifier to continue the same agent conversation across trigger events.,https://developers.openai.com/workspace-agents/trigger-runs
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,Idempotency-Key,header,string,false,,,,,,,DOCUMENTED|UNVERIFIED,Safe retry of the same trigger event; reuse only when retrying the same event.,https://developers.openai.com/workspace-agents/trigger-runs
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,OpenAI-Beta,header,string,false,,,,,,,DOCUMENTED|UNVERIFIED,`workspace_agent_runs=v1` to receive agent_trigger_run_id and enable run polling (beta).,https://developers.openai.com/workspace-agents/trigger-runs
openai,POST https://api.chatgpt.com/v1/workspace_agents/{id}/trigger,Authorization,header,string,true,,,,,,,DOCUMENTED|UNVERIFIED,"Bearer Workspace Agent access token (ChatGPT admin access-token flow), scoped to Workspace Agents API only.",https://developers.openai.com/workspace-agents/trigger-runs
openai,POST /v1/audio/speech,model,body,string | enum,true,,,,tts-1|tts-1-hd|gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED|LIVE_VERIFIED,"One of the available [TTS models](https://developers.openai.com/api/docs/guides/text-to-speech): `tts-1`, `tts-1-hd`, `gpt-4o-mini-tts`, or `gpt-4o-mini-tts-2025-12-15`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,input,body,string,true,,,,,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED|LIVE_VERIFIED,The text to generate audio for. The maximum length is 4096 characters.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,instructions,body,string,false,,,,,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED,Control the voice of your generated audio with additional instructions. Does not work with `tts-1` or `tts-1-hd`.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,voice,body,VoiceIdsShared | object,true,,,,,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED|LIVE_VERIFIED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,voice.id,body,string,false,,,,,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,response_format,body,string,false,mp3,,,mp3|opus|aac|flac|wav|pcm,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED|LIVE_VERIFIED,"The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`, `wav`, and `pcm`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,speed,body,number,false,1,0.25,4,,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED,The speed of the generated audio. Select a value from `0.25` to `4.0`. `1.0` is the default.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/speech,stream_format,body,string,false,audio,,,sse|audio,gpt-4o-mini-tts|gpt-4o-mini-tts-2025-12-15|tts-1|tts-1-hd,,DOCUMENTED|LIVE_VERIFIED,The format to stream the audio in. Supported formats are `sse` and `audio`. `sse` is not supported for `tts-1` or `tts-1-hd`.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/speech/methods/create""}"
openai,POST /v1/audio/transcriptions,file,body,string,true,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,"The audio file object (not file name) to transcribe, in one of these formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
The request must include enough format metadata for the file to be identified. We recommend an extension-bearing filename and an appropriate content type.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,model,body,string | enum,true,,,,whisper-1|gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe-diarize,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,"ID of the model to use. The options are `gpt-transcribe`, `gpt-4o-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `whisper-1` (which is powered by our open source Whisper V2 model), and `gpt-4o-transcribe-diarize`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,language,body,string,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,The language of the input audio. Supplying the input language in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format will improve accuracy and latency.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,languages,body,array<string>,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,keywords,body,array<string>,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,prompt,body,string,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,An optional text to guide the model's style or continue a previous audio segment. The [prompt](https://developers.openai.com/api/docs/guides/speech-to-text#prompting) should match the audio language. This field is not supported when using `gpt-4o-transcribe-diarize`.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,response_format,body,string,false,json,,,json|text|srt|verbose_json|vtt|diarized_json,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,"The format of the output, in one of these options: `json`, `text`, `srt`, `verbose_json`, `vtt`, or `diarized_json`. For `gpt-4o-transcribe` and `gpt-4o-mini-transcribe`, the only supported format is `json`. For `gpt-4o-transcribe-diarize`, the supported formats are `json`, `text`, and `diarized_json`, with `diarized_json` required to receive speaker annotations.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,temperature,body,number,false,0,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use [log probability](https://en.wikipedia.org/wiki/Log_probability) to automatically increase the temperature until certain thresholds are hit.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,include,body,array<TranscriptionInclude>,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,"Additional information to include in the transcription response.
`logprobs` will return the log probabilities of the tokens in the
response to understand the model's confidence in the transcription.
`logprobs` only works with response_format set to `json` and only with
the models `gpt-4o-transcribe`, `gpt-4o-mini-transcribe`, and `gpt-4o-mini-transcribe-2025-12-15`. This field is not supported when using `gpt-4o-transcribe-diarize`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,timestamp_granularities,body,array<string>,false,segment,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,"The timestamp granularities to populate for this transcription. `response_format` must be set `verbose_json` to use timestamp granularities. Either or both of these options are supported: `word`, or `segment`. Note: There is no additional latency for segment timestamps, but generating word timestamps incurs additional latency.
This option is not available for `gpt-4o-transcribe-diarize`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,stream,body,boolean | null,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED|LIVE_VERIFIED,,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,chunking_strategy,body,object | null,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,chunking_strategy.type,body,string,false,,,,server_vad,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,Must be set to `server_vad` to enable manual chunking using server side VAD.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,chunking_strategy.prefix_padding_ms,body,integer,false,300,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"Amount of audio to include before the VAD detected speech (in 
milliseconds).","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,chunking_strategy.silence_duration_ms,body,integer,false,200,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"Duration of silence to detect speech stop (in milliseconds).
With shorter values the model will respond more quickly, 
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,chunking_strategy.threshold,body,number,false,0.5,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"Sensitivity threshold (0.0 to 1.0) for voice activity detection. A 
higher threshold will require louder audio to activate the model, and 
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,known_speaker_names,body,array<string>,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,Optional list of speaker names that correspond to the audio samples provided in `known_speaker_references[]`. Each entry should be a short identifier (for example `customer` or `agent`). Up to 4 speakers are supported.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/transcriptions,known_speaker_references,body,array<string>,false,,,,,gpt-transcribe|gpt-4o-transcribe|gpt-4o-mini-transcribe|gpt-4o-transcribe-diarize|whisper-1,,DOCUMENTED,"Optional list of audio samples (as [data URLs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URLs)) that contain known speaker references matching `known_speaker_names[]`. Each sample must be between 2 and 10 seconds, and can use any of the same input audio formats supported by `file`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create""}"
openai,POST /v1/audio/translations,file,body,string,true,,,,,whisper-1,,DOCUMENTED|LIVE_VERIFIED,"The audio file object (not file name) translate, in one of these formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm. The request must include enough format metadata for the file to be identified. We recommend an extension-bearing filename and an appropriate content type.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/translations/methods/create""}"
openai,POST /v1/audio/translations,model,body,string | enum,true,,,,whisper-1,whisper-1,,DOCUMENTED|LIVE_VERIFIED,ID of the model to use. Only `whisper-1` (which is powered by our open source Whisper V2 model) is currently available.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/translations/methods/create""}"
openai,POST /v1/audio/translations,prompt,body,string,false,,,,,whisper-1,,DOCUMENTED,An optional text to guide the model's style or continue a previous audio segment. The [prompt](https://developers.openai.com/api/docs/guides/speech-to-text#prompting) should be in English.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/translations/methods/create""}"
openai,POST /v1/audio/translations,response_format,body,string,false,json,,,json|text|srt|verbose_json|vtt,whisper-1,,DOCUMENTED|LIVE_VERIFIED,"The format of the output, in one of these options: `json`, `text`, `srt`, `verbose_json`, or `vtt`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/translations/methods/create""}"
openai,POST /v1/audio/translations,temperature,body,number,false,0,,,,whisper-1,,DOCUMENTED,"The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use [log probability](https://en.wikipedia.org/wiki/Log_probability) to automatically increase the temperature until certain thresholds are hit.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/translations/methods/create""}"
openai,POST /v1/audio/voices,name,body,string,true,,,,,,,DOCUMENTED,The name of the new voice.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voices/methods/create""}"
openai,POST /v1/audio/voices,audio_sample,body,string,true,,,,,,,DOCUMENTED,"The sample audio recording file. Maximum size is 10 MiB.

Supported MIME types:
`audio/mpeg`, `audio/wav`, `audio/x-wav`, `audio/ogg`, `audio/aac`, `audio/flac`, `audio/webm`, `audio/mp4`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voices/methods/create""}"
openai,POST /v1/audio/voices,consent,body,string,true,,,,,,,DOCUMENTED,"The consent recording ID (for example, `cons_1234`).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voices/methods/create""}"
openai,POST /v1/audio/voice_consents,name,body,string,true,,,,,,,DOCUMENTED,The label to use for this consent recording.,"{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/create""}"
openai,POST /v1/audio/voice_consents,recording,body,string,true,,,,,,,DOCUMENTED,"The consent audio recording file. Maximum size is 10 MiB.

Supported MIME types:
`audio/mpeg`, `audio/wav`, `audio/x-wav`, `audio/ogg`, `audio/aac`, `audio/flac`, `audio/webm`, `audio/mp4`.","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/create""}"
openai,POST /v1/audio/voice_consents,language,body,string,true,,,,,,,DOCUMENTED,"The BCP 47 language tag for the consent phrase (for example, `en-US`).","{""retrieved_at"": ""2026-09-19T01:31:50Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/create""}"
openai,POST /v1/audio/voice_consents/{consent_id},name,body,string,true,,,,,,,DOCUMENTED,The updated label for this consent recording.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/update""}"
openai,POST /v1/audio/voice_consents/{consent_id},consent_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/update""}"
openai,GET /v1/audio/voice_consents,after,query,string,false,,,,,,,DOCUMENTED,Cursor for pagination (object ID).,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/list""}"
openai,GET /v1/audio/voice_consents,limit,query,integer,false,20,1,100,,,,DOCUMENTED,"1–100, default 20.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/list""}"
openai,GET /v1/audio/voice_consents/{consent_id},consent_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/retrieve""}"
openai,DELETE /v1/audio/voice_consents/{consent_id},consent_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/audio/subresources/voice_consents/methods/delete""}"
openai,POST /v1/chat/completions (audio subset),modalities,body,array<string>,false,,,,text|audio,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED|LIVE_VERIFIED," For audio output use `[""text"",""audio""]`. LIVE 2026-09-18: gpt-audio-mini with `modalities:[""text""]` and text-only input → 400 `This model requires that either input content or output modality contain audio.`","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),audio.voice,body,VoiceIdsShared | object,true,,,,,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),audio.voice.id,body,string,false,,,,,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),audio.format,body,string,true,,,,wav|aac|mp3|flac|opus|pcm16,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`,
`opus`, or `pcm16`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].type,body,string,true,,,,input_audio,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,The type of the content part. Always `input_audio`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].input_audio,body,object,true,,,,,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].input_audio.data,body,string,false,,,,,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,Base64 encoded audio data.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].input_audio.format,body,string,false,,,,wav|mp3,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"The format of the encoded audio data. Currently supports ""wav"" and ""mp3"".","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].prompt_cache_breakpoint,body,object,false,,,,,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/chat/completions (audio subset),messages[].content[].prompt_cache_breakpoint.mode,body,string,false,explicit,,,explicit,gpt-audio-1.5|gpt-audio|gpt-audio-mini|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,The breakpoint mode. Always `explicit`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/audio-chat-completions""}"
openai,POST /v1/batches,input_file_id,body,string,true,,,,,,,DOCUMENTED,"The ID of an uploaded file that contains requests for the new batch.

See [upload file](https://developers.openai.com/api/reference/resources/files/methods/create) for how to upload a file.

Your input file must be formatted as a [JSONL file](https://developers.openai.com/api/docs/guides/batch#1-prepare-your-batch-file), and must be uploaded with the purpose `batch`. The file can contain up to 50,000 requests, and can be up to 200 MB in size.",https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,endpoint,body,string,true,,,,/v1/responses|/v1/chat/completions|/v1/embeddings|/v1/completions|/v1/moderations|/v1/images/generations|/v1/images/edits|/v1/videos,,,DOCUMENTED,"The endpoint to be used for all requests in the batch. Currently `/v1/responses`, `/v1/chat/completions`, `/v1/embeddings`, `/v1/completions`, `/v1/moderations`, `/v1/images/generations`, `/v1/images/edits`, and `/v1/videos` are supported. Note that `/v1/embeddings` batches are also restricted to a maximum of 50,000 embedding inputs across all requests in the batch.",https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,completion_window,body,string,true,,,,24h,,,DOCUMENTED,The time frame within which the batch should be processed. Currently only `24h` is supported.,https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,metadata,body,object | null,false,,,,,,,DOCUMENTED,[schema: Metadata],https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,output_expires_after,body,object,false,,,,,,,DOCUMENTED,The expiration policy for the output and/or error file that are generated for a batch. [schema: BatchFileExpirationAfter],https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,output_expires_after.anchor,body,string,true,,,,created_at,,,DOCUMENTED,"Anchor timestamp after which the expiration policy applies. Supported anchors: `created_at`. Note that the anchor is the file creation time, not the time the batch is created.",https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches,output_expires_after.seconds,body,integer,true,,3600,2592000,,,,DOCUMENTED,The number of seconds after the anchor time that the file will expire. Must be between 3600 (1 hour) and 2592000 (30 days).,https://developers.openai.com/api/reference/resources/batches
openai,GET /v1/batches,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/batches
openai,GET /v1/batches,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/batches
openai,GET /v1/batches/{batch_id},batch_id,path,string,true,,,,,,,DOCUMENTED,The ID of the batch to retrieve.,https://developers.openai.com/api/reference/resources/batches
openai,POST /v1/batches/{batch_id}/cancel,batch_id,path,string,true,,,,,,,DOCUMENTED,The ID of the batch to cancel.,https://developers.openai.com/api/reference/resources/batches
openai,input_file (JSONL line),custom_id,file,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Unique per line; echoed in the output/error lines (output order is not guaranteed).,https://developers.openai.com/api/docs/guides/batch#1-prepare-your-batch-file
openai,input_file (JSONL line),method,file,string,true,,,,POST,,,DOCUMENTED|LIVE_VERIFIED,Always POST.,https://developers.openai.com/api/docs/guides/batch#1-prepare-your-batch-file
openai,input_file (JSONL line),url,file,string,true,,,,/v1/responses|/v1/chat/completions|/v1/embeddings|/v1/completions|/v1/moderations|/v1/images/generations|/v1/images/edits|/v1/videos,,,DOCUMENTED|LIVE_VERIFIED,Must equal the batch `endpoint`; all lines target one endpoint and one model.,https://developers.openai.com/api/docs/guides/batch#1-prepare-your-batch-file
openai,input_file (JSONL line),body,file,object,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Same JSON body as the synchronous endpoint (stream must be false/absent; videos/images must be JSON not multipart, assets referenced by file_id/image_url).",https://developers.openai.com/api/docs/guides/batch#1-prepare-your-batch-file
openai,POST /v1/chat/completions,metadata,body,"map<string,string>",false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard. Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,top_logprobs,body,integer | null,false,,0,20,,*,,DOCUMENTED|LIVE_VERIFIED,"An integer between 0 and 20 specifying the maximum number of most likely tokens to return at each token position, each with an associated log probability. In some cases, the number of returned tokens may be fewer than requested. `logprobs` must be set to `true` if this parameter is used.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,top_p,body,number | null,false,1,0,1,,*,,DOCUMENTED,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,user,body,string,false,,,,,*,,DOCUMENTED|LEGACY|LIVE_VERIFIED,This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use `prompt_cache_key` instead to maintain caching optimizations. A stable identifier for your end-users. Used to boost cache hit rates by better bucketing similar requests and to help OpenAI detect and prevent abuse. [Learn more](https://developers.openai.com/api/docs/guides/safety-best-practices#implement-safety-identifiers).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,safety_identifier,body,string | null,false,,,,,*,,DOCUMENTED,"A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies. The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](https://developers.openai.com/api/docs/guides/safety-best-practices#implement-safety-identifiers).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prompt_cache_key,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Used by OpenAI to cache responses for similar requests to optimize your cache hit rates. Replaces the `user` field. [Learn more](https://developers.openai.com/api/docs/guides/prompt-caching).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prompt_cache_retention,body,string | null,false,,,,in_memory|24h,gpt-5.5|gpt-5.5-pro|gpt-5.4|gpt-5.2|gpt-5.1-codex-max|gpt-5.1|gpt-5.1-codex|gpt-5.1-codex-mini|gpt-5.1-chat-latest|gpt-5|gpt-5-codex|gpt-4.1|(in_memory: all cache-capable models),,DOCUMENTED|LEGACY,"Deprecated. Use `prompt_cache_options.ttl` instead. The retention policy for the prompt cache. Set to `24h` to enable extended prompt caching, which keeps cached prefixes active for longer, up to a maximum of 24 hours. [Learn more](https://developers.openai.com/api/docs/guides/prompt-caching#prompt-cache-retention). This field expresses a maximum retention policy, while `prompt_cache_options.ttl` expresses a minimum cache lifetime. The two fields are independent and do not interact. For `gpt-5.5`, `gpt-5.5-pro`, and future models, only `24h` is supported. For older models that support both `in_memory` and `24h`, the default depends on your organization's data retention policy: - Organizations without ZDR enabled default to `24h`. - Organizations with ZDR enabled default to `in_memory` when `prompt_cache_retention` is not specified.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prompt_cache_options,body,object,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Options for prompt caching. Supported for `gpt-5.6` and later models. By default, OpenAI automatically chooses one implicit cache breakpoint. You can add explicit breakpoints to content blocks with `prompt_cache_breakpoint`. Each request can write up to four breakpoints. For cache matching, OpenAI considers up to the latest 80 breakpoints in the conversation, without a content-block lookback limit. Set `mode` to `explicit` to disable the implicit breakpoint. The `ttl` defaults to `30m`, which is currently the only supported value. See the [prompt caching guide](https://developers.openai.com/api/docs/guides/prompt-caching) for current details.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prompt_cache_options.ttl,body,string,false,,,,30m,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"The minimum lifetime applied to every implicit and explicit cache breakpoint written by the request. Defaults to `30m`, which is currently the only supported value. The backend may retain cache entries for longer.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prompt_cache_options.mode,body,string,false,,,,implicit|explicit,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Controls whether OpenAI automatically creates an implicit cache breakpoint. Defaults to `implicit`. With `implicit`, OpenAI creates one implicit breakpoint and writes up to the latest three explicit breakpoints in the request. With `explicit`, OpenAI does not create an implicit breakpoint and writes up to the latest four explicit breakpoints. If there are no explicit breakpoints, the request does not use prompt caching.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages,body,array<object>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of messages comprising the conversation so far. Depending on the [model](https://developers.openai.com/api/docs/models) you use, different message types (modalities) are supported, like [text](https://developers.openai.com/api/docs/guides/text), [images](https://developers.openai.com/api/docs/guides/images-vision), and [audio](https://developers.openai.com/api/docs/guides/audio).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[],body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer),body,object,false,,,,,*,,DOCUMENTED,"Developer-provided instructions that the model should follow, regardless of messages sent by the user. With o1 models and newer, `developer` messages replace the previous `system` messages.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content,body,string | array<object>,true,,,,,*,,DOCUMENTED,The contents of the developer message.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content[],body,array<object> | object,false,,,,,*,,DOCUMENTED,"An array of content parts with a defined type. For developer messages, only type `text` is supported.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).role,body,string,true,,,,developer,*,,DOCUMENTED,"The role of the messages author, in this case `developer`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).name,body,string,false,,,,,*,,DOCUMENTED,An optional name for the participant. Provides the model information to differentiate between participants of the same role.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Developer-provided instructions that the model should follow, regardless of messages sent by the user. With o1 models and newer, use `developer` messages for this purpose instead.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content,body,string | array<object>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The contents of the system message.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content[],body,array<object> | object,false,,,,,*,,DOCUMENTED,"An array of content parts with a defined type. For system messages, only type `text` is supported.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).role,body,string,true,,,,system,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the messages author, in this case `system`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).name,body,string,false,,,,,*,,DOCUMENTED,An optional name for the participant. Provides the model information to differentiate between participants of the same role.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Messages sent by an end user, containing prompts or additional context information.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content,body,string | array<object<text> | object<image_url> | object<input_audio> | object<file>>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The contents of the user message.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[],body,array<object<text> | object<image_url> | object<input_audio> | object<file>> | object<text> | object<file>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An array of content parts with a defined type. Supported options differ based on the [model](https://developers.openai.com/api/docs/models) being used to generate the response. Can contain text, image, or audio inputs.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).role,body,string,true,,,,user,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the messages author, in this case `user`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).name,body,string,false,,,,,*,,DOCUMENTED,An optional name for the participant. Provides the model information to differentiate between participants of the same role.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant),body,object,false,,,,,*,,DOCUMENTED,Messages sent by the model in response to user messages.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content,body,string | array<object<text> | object<refusal>>,false,,,,,*,,DOCUMENTED,The contents of the assistant message. Required unless `tool_calls` or `function_call` is specified.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[],body,array<object<text> | object<refusal>> | object<text> | object<refusal>,false,,,,,*,,DOCUMENTED,"An array of content parts with a defined type. Can be one or more of type `text`, or exactly one of type `refusal`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).refusal,body,string | null,false,,,,,*,,DOCUMENTED,The refusal message by the assistant.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).role,body,string,true,,,,assistant,*,,DOCUMENTED,"The role of the messages author, in this case `assistant`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).name,body,string,false,,,,,*,,DOCUMENTED,An optional name for the participant. Provides the model information to differentiate between participants of the same role.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).audio,body,object | null,false,,,,,*,,DOCUMENTED,Data about a previous audio response from the model. [Learn more](https://developers.openai.com/api/docs/guides/audio).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).audio.id,body,string,true,,,,,*,,DOCUMENTED,Unique identifier for a previous audio response from the model.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls,body,array<object<function> | object<custom>>,false,,,,,*,,DOCUMENTED,"The tool calls generated by the model, such as function calls.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[],body,object<function> | object<custom>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function),body,object,false,,,,,*,,DOCUMENTED,A call to a function tool created by the model.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).function_call,body,object | null,false,,,,,*,,DOCUMENTED|LEGACY,"Deprecated and replaced by `tool_calls`. The name and arguments of a function that should be called, as generated by the model.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).function_call.arguments,body,string,true,,,,,*,,DOCUMENTED,"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).function_call.name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).role,body,string,true,,,,tool,*,,DOCUMENTED,"The role of the messages author, in this case `tool`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content,body,string | array<object>,true,,,,,*,,DOCUMENTED,The contents of the tool message.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content[],body,array<object> | object,false,,,,,*,,DOCUMENTED,"An array of content parts with a defined type. For tool messages, only type `text` is supported.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).tool_call_id,body,string,true,,,,,*,,DOCUMENTED,Tool call that this message is responding to.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](function),body,object,false,,,,,*,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](function).role,body,string,true,,,,function,*,,DOCUMENTED,"The role of the messages author, in this case `function`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](function).content,body,string | null,true,,,,,*,,DOCUMENTED,The contents of the function message.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](function).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,model,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Model ID used to generate the response, like `gpt-6-astra` or `o3`. OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points. Refer to the [model guide](https://developers.openai.com/api/docs/models) to browse and compare available models.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,service_tier,body,string | null,false,auto,,,auto|default|flex|scale|priority|fast,*,,DOCUMENTED,"Specifies the processing type used for serving the request. - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'. - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model. - If set to '[flex](https://developers.openai.com/api/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier. - To opt-in to [Fast mode](https://developers.openai.com/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request. - When not set, the default behavior is 'auto'",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,modalities,body,array<string>,false,,,,,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"Output types that you would like the model to generate. Most models are capable of generating text, which is the default: `[""text""]` The `gpt-4o-audio-preview` model can also be used to [generate audio](https://developers.openai.com/api/docs/guides/audio). To request that this model generate both text and audio responses, you can use: `[""text"", ""audio""]`",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,modalities[],body,string,false,,,,text|audio,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,verbosity,body,string | null,false,medium,,,low|medium|high,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4*|gpt-5.2*|gpt-5.1*|gpt-5*,,DOCUMENTED,"Constrains the verbosity of the model's response. Lower values will result in more concise responses, while higher values will result in more verbose responses. Currently supported values are `low`, `medium`, and `high`. The default is `medium`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,reasoning_effort,body,string | null,false,medium,,,none|minimal|low|medium|high|xhigh|max,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"Constrains effort on reasoning for reasoning models. Currently supported values are `none`, `minimal`, `low`, `medium`, `high`, `xhigh`, and `max`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Not all reasoning models support every value. See the [reasoning guide](https://developers.openai.com/api/docs/guides/reasoning) for model-specific support.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,max_completion_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and [reasoning tokens](https://developers.openai.com/api/docs/guides/reasoning).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,frequency_penalty,body,number | null,false,0,-2,2,,*,,DOCUMENTED,"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,presence_penalty,body,number | null,false,0,-2,2,,*,,DOCUMENTED,"Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options,body,object,false,,,,,,,DOCUMENTED,"Chat Completions web search (search models gpt-5-search-api, gpt-4o(-mini)-search-preview): {user_location: {type: approximate, approximate: {country, region, city, timezone}}, search_context_size}.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,web_search_options.user_location,body,object | null,false,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,Approximate location parameters for the search.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.type,body,string,true,,,,approximate,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,The type of location approximation. Always `approximate`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.approximate,body,object,true,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,Approximate location parameters for the search.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.approximate.country,body,string,false,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,"The two-letter [ISO country code](https://en.wikipedia.org/wiki/ISO_3166-1) of the user, e.g. `US`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.approximate.region,body,string,false,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,"Free text input for the region of the user, e.g. `California`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.approximate.city,body,string,false,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,"Free text input for the city of the user, e.g. `San Francisco`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.user_location.approximate.timezone,body,string,false,,,,,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,"The [IANA timezone](https://timeapi.io/documentation/iana-timezones) of the user, e.g. `America/Los_Angeles`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,web_search_options.search_context_size,body,string,false,medium,,,low|medium|high,gpt-4o-search-preview|gpt-4o-mini-search-preview|gpt-5* (web search via chat),,DOCUMENTED,"High level guidance for the amount of context window space to use for the search. One of `low`, `medium`, or `high`. `medium` is the default.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format,body,object<text> | object<json_schema> | object<json_object>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An object specifying the format that the model must output. Setting to `{ ""type"": ""json_schema"", ""json_schema"": {...} }` enables Structured Outputs which ensures the model will match your supplied JSON schema. Learn more in the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs). Setting to `{ ""type"": ""json_object"" }` enables the older JSON mode, which ensures the message the model generates is valid JSON. Using `json_schema` is preferred for models that support it.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(text),body,object,false,,,,,*,,DOCUMENTED,Default response format. Used to generate text responses.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of response format being defined. Always `text`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,JSON Schema response format. Used to generate structured JSON responses. Learn more about [Structured Outputs](https://developers.openai.com/api/docs/guides/structured-outputs).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).type,body,string,true,,,,json_schema,*,,DOCUMENTED|LIVE_VERIFIED,The type of response format being defined. Always `json_schema`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).json_schema,body,object,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Structured Outputs configuration options, including a JSON Schema.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).json_schema.description,body,string,false,,,,,*,,DOCUMENTED,"A description of what the response format is for, used by the model to determine how to respond in the format.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).json_schema.name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).json_schema.schema,body,"map<string,any>",false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas [here](https://json-schema.org/).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_schema).json_schema.strict,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the `schema` field. Only a subset of JSON Schema is supported when `strict` is `true`. To learn more, read the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_object),body,object,false,,,,,*,,DOCUMENTED,JSON object response format. An older method of generating JSON responses. Using `json_schema` is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,response_format(json_object).type,body,string,true,,,,json_object,*,,DOCUMENTED,The type of response format being defined. Always `json_object`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,audio,body,object | null,false,,,,,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"Parameters for audio output. Required when audio output is requested with `modalities: [""audio""]`. [Learn more](https://developers.openai.com/api/docs/guides/audio).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,audio.voice,body,string | object,true,,,,,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"The voice the model uses to respond. Supported built-in voices are `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `nova`, `onyx`, `sage`, `shimmer`, `marin`, and `cedar`. You may also provide a custom voice object with an `id`, for example `{ ""id"": ""voice_1234"" }`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,audio.voice(object),body,object,true,,,,,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,Custom voice reference.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,audio.voice(object).id,body,string,true,,,,,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,audio.format,body,string,true,,,,wav|aac|mp3|flac|opus|pcm16,gpt-audio-1.5|gpt-4o-audio-preview|gpt-4o-mini-audio-preview,,DOCUMENTED,"Specifies the output audio format. Must be one of `wav`, `mp3`, `flac`, `opus`, or `pcm16`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,store,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,Whether or not to store the output of this chat completion request for use in our [model distillation](https://developers.openai.com/api/docs/guides/supervised-fine-tuning#distilling-from-a-larger-model) or [evals](https://developers.openai.com/api/docs/guides/evals) products. Supports text and image inputs. Note: image inputs over 8MB will be dropped.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation,body,object | null,false,,,,,*,,DOCUMENTED,Configuration for running moderation on the request input and generated output.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.model,body,string,true,,,,,*,,DOCUMENTED,"The moderation model to use for moderated completions, e.g. 'omni-moderation-latest'.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.policy,body,object | null,false,,,,,*,,DOCUMENTED,The policy to apply to moderated response input and output.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.policy.input,body,object | null,false,,,,,*,,DOCUMENTED,The moderation policy for the response input.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.policy.input.mode,body,string,true,,,,score|block,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.policy.output,body,object | null,false,,,,,*,,DOCUMENTED,The moderation policy for the response output.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,moderation.policy.output.mode,body,string,true,,,,score|block,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stream,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set to true, the model response data will be streamed to the client as it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format). See the [Streaming section below](https://developers.openai.com/api/reference/resources/chat/subresources/completions/streaming-events) for more information, along with the [streaming responses](https://developers.openai.com/api/docs/guides/streaming-responses) guide for more information on how to handle the streaming events.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stop,body,string | null | array<string>,false,,,,,*,,DOCUMENTED,Not supported with latest reasoning models `o3` and `o4-mini`. Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stop[],body,array<string> | string,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,logit_bias,body,"map<string,integer>",false,,,,,*,,DOCUMENTED,"Modify the likelihood of specified tokens appearing in the completion. Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,logprobs,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the `content` of `message`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,max_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|LIVE_VERIFIED,"The maximum number of [tokens](https://platform.openai.com/tokenizer) that can be generated in the chat completion. This value can be used to control [costs](https://openai.com/api/pricing/) for text generated via API. This value is now deprecated in favor of `max_completion_tokens`, and is not compatible with [o-series models](https://developers.openai.com/api/docs/guides/reasoning).",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,n,body,integer | null,false,1,1,128,,*,,DOCUMENTED|LIVE_VERIFIED,How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep `n` as `1` to minimize costs.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction,body,object | null,false,,,,,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,"Configuration for a [Predicted Output](https://developers.openai.com/api/docs/guides/predicted-outputs), which can greatly improve response times when large parts of the model response are known ahead of time. This is most common when you are regenerating a file with only minor changes to most of the content.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.type,body,string,true,,,,content,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,The type of the predicted content you want to provide. This type is currently always `content`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content,body,string | array<object>,true,,,,,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content[],body,array<object> | object,false,,,,,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,An array of content parts with a defined type. Supported options differ based on the [model](https://developers.openai.com/api/docs/models) being used to generate the response. Can contain text inputs.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content[].type,body,string,true,,,,text,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content[].text,body,string,true,,,,,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content[].prompt_cache_breakpoint,body,object,false,,,,,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,prediction.content[].prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,gpt-4o|gpt-4o-mini|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,seed,body,integer | null,false,,-9223372036854776000,9223372036854776000,,*,,DOCUMENTED|LEGACY|LIVE_VERIFIED,"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stream_options,body,object | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Options for streaming response. Only set this when you set `stream: true`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stream_options.include_usage,body,boolean,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value. **NOTE:** If the stream is interrupted, you may not receive the final usage chunk which contains the total token usage for the request.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,stream_options.include_obfuscation,body,boolean,false,,,,,*,,DOCUMENTED,"When true, stream obfuscation will be enabled. Stream obfuscation adds random characters to an `obfuscation` field on streaming delta events to normalize payload sizes as a mitigation to certain side-channel attacks. These obfuscation fields are included by default, but add a small amount of overhead to the data stream. You can set `include_obfuscation` to false to optimize for bandwidth if you trust the network links between your application and the OpenAI API.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tools,body,array<object<function> | object<custom>>,false,,,,,*,,DOCUMENTED,A list of tools the model may call. You can provide either [custom tools](https://developers.openai.com/api/docs/guides/function-calling#custom-tools) or [function tools](https://developers.openai.com/api/docs/guides/function-calling).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tools[],body,object<function> | object<custom>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tools[](function),body,object,false,,,,,*,,DOCUMENTED,A function tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tools[](custom),body,object,false,,,,,*,,DOCUMENTED,A custom tool that processes input using a specified format.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice,body,string | object<allowed_tools> | object<function> | object<custom>,false,,,,,*,,DOCUMENTED,"Controls which (if any) tool is called by the model. `none` means the model will not call any tool and instead generates a message. `auto` means the model can pick between generating a message or calling one or more tools. `required` means the model must call one or more tools. Specifying a particular tool via `{""type"": ""function"", ""function"": {""name"": ""my_function""}}` forces the model to call that tool. `none` is the default when no tools are present. `auto` is the default if tools are present.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools),body,object,false,,,,,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools).type,body,string,true,,,,allowed_tools,*,,DOCUMENTED,Allowed tool configuration type. Always `allowed_tools`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools).allowed_tools,body,object,true,,,,,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools).allowed_tools.mode,body,string,true,,,,auto|required,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set. `auto` allows the model to pick from among the allowed tools and generate a message. `required` requires the model to call one or more of the allowed tools.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools).allowed_tools.tools,body,"array<map<string,any>>",true,,,,,*,,DOCUMENTED,"A list of tool definitions that the model should be allowed to call. For the Chat Completions API, the list of tool definitions might look like: ```json [ { ""type"": ""function"", ""function"": { ""name"": ""get_weather"" } }, { ""type"": ""function"", ""function"": { ""name"": ""get_time"" } } ] ```",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(allowed_tools).allowed_tools.tools[],body,"map<string,any>",false,,,,,*,,DOCUMENTED,A tool definition that the model should be allowed to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(function),body,object,false,,,,,*,,DOCUMENTED,Specifies a tool the model should use. Use to force the model to call a specific function.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(function).type,body,string,true,,,,function,*,,DOCUMENTED,"For function calling, the type is always `function`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(function).function,body,object,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(function).function.name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(custom),body,object,false,,,,,*,,DOCUMENTED,Specifies a tool the model should use. Use to force the model to call a specific custom tool.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(custom).type,body,string,true,,,,custom,*,,DOCUMENTED,"For custom tool calling, the type is always `custom`.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(custom).custom,body,object,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,tool_choice(custom).custom.name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,parallel_tool_calls,body,boolean,false,true,,,,,,DOCUMENTED,Whether to enable [parallel function calling](https://developers.openai.com/api/docs/guides/function-calling#parallel-function-calling) during tool use.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,function_call,body,string | object,false,,,,,*,,DOCUMENTED|LEGACY,"Deprecated in favor of `tool_choice`. Controls which (if any) function is called by the model. `none` means the model will not call a function and instead generates a message. `auto` means the model can pick between generating a message or calling a function. Specifying a particular function via `{""name"": ""my_function""}` forces the model to call that function. `none` is the default when no functions are present. `auto` is the default if functions are present.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,function_call(object),body,object,false,,,,,*,,DOCUMENTED,"Specifying a particular function via `{""name"": ""my_function""}` forces the model to call that function.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,function_call(object).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,functions,body,array<object>,false,,,,,*,,DOCUMENTED|LEGACY,Deprecated in favor of `tools`. A list of functions the model may generate JSON inputs for.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,functions[],body,object,false,,,,,*,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,functions[].description,body,string,false,,,,,*,,DOCUMENTED,"A description of what the function does, used by the model to choose when and how to call the function.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,functions[].name,body,string,true,,,,,*,,DOCUMENTED,"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,functions[].parameters,body,"map<string,any>",false,,,,,*,,DOCUMENTED,"The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://developers.openai.com/api/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format. Omitting `parameters` defines a function with an empty parameter list.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content[](text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content[](text).text,body,string,true,,,,,*,,DOCUMENTED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content[](text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](developer).content[](text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content[](text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content[](text).text,body,string,true,,,,,*,,DOCUMENTED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content[](text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](system).content[](text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](text).type,body,string,true,,,,text,*,,DOCUMENTED|LIVE_VERIFIED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](text).text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).type,body,string,true,,,,image_url,*,,DOCUMENTED|LIVE_VERIFIED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).image_url,body,object,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).image_url.url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Either a URL of the image or the base64 encoded image data.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).image_url.detail,body,string,false,auto,,,auto|low|high,*,,DOCUMENTED|LIVE_VERIFIED,Specifies the detail level of the image. Learn more in the [Vision guide](https://developers.openai.com/api/docs/guides/images-vision#choose-an-image-detail-level).,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](image_url).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).type,body,string,true,,,,input_audio,*,,DOCUMENTED,The type of the content part. Always `input_audio`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).input_audio,body,object,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).input_audio.data,body,string,true,,,,,*,,DOCUMENTED,Base64 encoded audio data.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).input_audio.format,body,string,true,,,,wav|mp3,*,,DOCUMENTED,"The format of the encoded audio data. Currently supports ""wav"" and ""mp3"".",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](input_audio).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).type,body,string,true,,,,file,*,,DOCUMENTED,The type of the content part. Always `file`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).file,body,object,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).file.filename,body,string,false,,,,,*,,DOCUMENTED,"The name of the file, used when passing the file to the model as a string.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).file.file_data,body,string,false,,,,,*,,DOCUMENTED,"The base64 encoded file data, used when passing the file to the model as a string.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).file.file_id,body,string,false,,,,,*,,DOCUMENTED,The ID of an uploaded file to use as input.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](user).content[](file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](text).text,body,string,true,,,,,*,,DOCUMENTED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](refusal).type,body,string,true,,,,refusal,*,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).content[](refusal).refusal,body,string,true,,,,,*,,DOCUMENTED,The refusal message generated by the model.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function).id,body,string,true,,,,,*,,DOCUMENTED,The ID of the tool call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function).type,body,string,true,,,,function,*,,DOCUMENTED,"The type of the tool. Currently, only `function` is supported.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function).function,body,object,true,,,,,*,,DOCUMENTED,The function that the model called.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function).function.name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](function).function.arguments,body,string,true,,,,,*,,DOCUMENTED,"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom).id,body,string,true,,,,,*,,DOCUMENTED,The ID of the tool call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom).type,body,string,true,,,,custom,*,,DOCUMENTED,The type of the tool. Always `custom`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom).custom,body,object,true,,,,,*,,DOCUMENTED,The custom tool that the model called.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom).custom.name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool to call.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](assistant).tool_calls[](custom).custom.input,body,string,true,,,,,*,,DOCUMENTED,The input for the custom tool call generated by the model.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content[](text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of the content part.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content[](text).text,body,string,true,,,,,*,,DOCUMENTED,The text content.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content[](text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,POST /v1/chat/completions,messages[](tool).content[](text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/create
openai,GET /v1/chat/completions,model,query,string,false,,,,,*,,DOCUMENTED,The model used to generate the Chat Completions.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/list
openai,GET /v1/chat/completions,metadata,query,"map<string,string> | null",false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A list of metadata keys to filter the Chat Completions by. Example: `metadata[key1]=value1&metadata[key2]=value2`,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/list
openai,GET /v1/chat/completions,after,query,string,false,,,,,*,,DOCUMENTED,Identifier for the last chat completion from the previous pagination request.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/list
openai,GET /v1/chat/completions,limit,query,integer,false,20,,,,*,,DOCUMENTED|LIVE_VERIFIED,Number of Chat Completions to retrieve.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/list
openai,GET /v1/chat/completions,order,query,string,false,asc,,,asc|desc,*,,DOCUMENTED,Sort order for Chat Completions by timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `asc`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/list
openai,GET /v1/chat/completions/{completion_id}/messages,after,query,string,false,,,,,*,,DOCUMENTED,Identifier for the last message from the previous pagination request.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/subresources/messages/methods/list
openai,GET /v1/chat/completions/{completion_id}/messages,limit,query,integer,false,20,,,,*,,DOCUMENTED|LIVE_VERIFIED,Number of messages to retrieve.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/subresources/messages/methods/list
openai,GET /v1/chat/completions/{completion_id}/messages,order,query,string,false,asc,,,asc|desc,*,,DOCUMENTED,Sort order for messages by timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `asc`.,https://developers.openai.com/api/reference/resources/chat/subresources/completions/subresources/messages/methods/list
openai,POST /v1/chat/completions/{completion_id},metadata,body,"map<string,string>",true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Set of 16 key-value pairs (keys <= 64 chars, values <= 512 chars) replacing the stored completion metadata. Only stored completions (store=true) can be updated.",https://developers.openai.com/api/reference/resources/chat/subresources/completions/methods/update
openai,POST /v1/completions,model,body,string,true,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,"ID of the model to use. You can use the [List models](https://developers.openai.com/api/reference/resources/models/methods/list) API to see all of your available models, or see our [Model overview](https://developers.openai.com/api/docs/models) for descriptions of them.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,prompt,body,string | array<string> | array<integer> | array<array<integer>>,true,<|endoftext|>,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,"The prompt(s) to generate completions for, encoded as a string, array of strings, array of tokens, or array of token arrays. Note that <|endoftext|> is the document separator that the model sees during training, so if a prompt is not specified the model will generate as if from the beginning of a new document.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,prompt[],body,array<string> | string | array<integer> | integer | array<array<integer>>,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,prompt[][],body,integer,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,best_of,body,integer | null,false,1,0,20,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"Generates `best_of` completions server-side and returns the ""best"" (the one with the highest log probability per token). Results cannot be streamed. When used with `n`, `best_of` controls the number of candidate completions and `n` specifies how many to return – `best_of` must be greater than `n`. **Note:** Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for `max_tokens` and `stop`.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,echo,body,boolean | null,false,false,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,Echo back the prompt in addition to the completion,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,frequency_penalty,body,number | null,false,0,-2,2,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim. [See more information about frequency and presence penalties.](https://developers.openai.com/api/docs/guides/text)",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,logit_bias,body,"map<string,integer>",false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"Modify the likelihood of specified tokens appearing in the completion. Accepts a JSON object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias value from -100 to 100. You can use this [tokenizer tool](https://platform.openai.com/tokenizer?view=bpe) to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token. As an example, you can pass `{""50256"": -100}` to prevent the <|endoftext|> token from being generated.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,logprobs,body,integer | null,false,,0,5,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"Include the log probabilities on the `logprobs` most likely output tokens, as well the chosen tokens. For example, if `logprobs` is 5, the API will return a list of the 5 most likely tokens. The API will always return the `logprob` of the sampled token, so there may be up to `logprobs+1` elements in the response. The maximum value for `logprobs` is 5.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,max_tokens,body,integer | null,false,16,0,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,The maximum number of [tokens](https://platform.openai.com/tokenizer) that can be generated in the completion. The token count of your prompt plus `max_tokens` cannot exceed the model's context length. [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken) for counting tokens.,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,n,body,integer | null,false,1,1,128,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"How many completions to generate for each prompt. **Note:** Because this parameter generates many completions, it can quickly consume your token quota. Use carefully and ensure that you have reasonable settings for `max_tokens` and `stop`.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,presence_penalty,body,number | null,false,0,-2,2,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics. [See more information about frequency and presence penalties.](https://developers.openai.com/api/docs/guides/text)",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,seed,body,integer | null,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stop,body,string | null | array<string>,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,Not supported with latest reasoning models `o3` and `o4-mini`. Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stop[],body,array<string> | string,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stream,body,boolean | null,false,false,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,"Whether to stream back partial progress. If set, tokens will be sent as data-only [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format) as they become available, with the stream terminated by a `data: [DONE]` message. [Example Python code](https://cookbook.openai.com/examples/how_to_stream_completions).",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stream_options,body,object | null,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,Options for streaming response. Only set this when you set `stream: true`.,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stream_options.include_usage,body,boolean,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LIVE_VERIFIED|LEGACY,"If set, an additional chunk will be streamed before the `data: [DONE]` message. The `usage` field on this chunk shows the token usage statistics for the entire request, and the `choices` field will always be an empty array. All other chunks will also include a `usage` field, but with a null value. **NOTE:** If the stream is interrupted, you may not receive the final usage chunk which contains the total token usage for the request.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,stream_options.include_obfuscation,body,boolean,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"When true, stream obfuscation will be enabled. Stream obfuscation adds random characters to an `obfuscation` field on streaming delta events to normalize payload sizes as a mitigation to certain side-channel attacks. These obfuscation fields are included by default, but add a small amount of overhead to the data stream. You can set `include_obfuscation` to false to optimize for bandwidth if you trust the network links between your application and the OpenAI API.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,suffix,body,string | null,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,The suffix that comes after a completion of inserted text. This parameter is only supported for `gpt-3.5-turbo-instruct`.,https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,temperature,body,number | null,false,1,0,2,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,top_p,body,number | null,false,1,0,1,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both.",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/completions,user,body,string,false,,,,,gpt-3.5-turbo-instruct|davinci-002|babbage-002,,DOCUMENTED|LEGACY,"A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. [Learn more](https://developers.openai.com/api/docs/guides/safety-best-practices#implement-safety-identifiers).",https://developers.openai.com/api/reference/resources/completions/methods/create
openai,POST /v1/conversations,metadata,body,"map<string,string> | null",false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard. Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items,body,array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Initial items to include in the conversation context. You may add up to 20 items at a time.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[],body,object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An item representing part of the context for the response to be generated by the model. Can contain text, images, and audio inputs, as well as previous assistant responses and tool call outputs.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A message input to the model with a role indicating instruction following hierarchy. Instructions given with the `developer` or `system` role take precedence over instructions given with the `user` role. Messages with the `assistant` role are presumed to have been generated by the model in previous interactions.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).role,body,string,true,,,,user|assistant|system|developer,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the message input. One of `user`, `assistant`, `system`, or `developer`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content,body,string | array<object<input_text> | object<input_image> | object<input_file>> | array<object<output_text> | object<refusal>>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or audio input to the model, used to generate a response. Can also contain previous assistant responses.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[],body,array<object<input_text> | object<input_image> | object<input_file>> | object<input_text> | object<input_file> | object<output_text> | object<refusal>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).phase,body,string | null,false,,,,commentary|final_answer,*,,DOCUMENTED,"Labels an `assistant` message as intermediate commentary (`commentary`) or the final answer (`final_answer`). For models like `gpt-5.3-codex` and beyond, when sending follow-up requests, preserve and resend phase on all assistant messages — dropping it can degrade performance. Not used for user messages.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).type,body,string,true,,,,message,*,,DOCUMENTED|LIVE_VERIFIED,The type of the message input. Always `message`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the output message.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a file search tool call. See the [file search guide](https://developers.openai.com/api/docs/guides/tools-file-search) for more information.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the file search tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).type,body,string,true,,,,file_search_call,*,,DOCUMENTED,The type of the file search tool call. Always `file_search_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).status,body,string,true,,,,in_progress|searching|completed|incomplete|failed,*,,DOCUMENTED,"The status of the file search tool call. One of `in_progress`, `searching`, `incomplete` or `failed`,",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).queries,body,array<string>,true,,,,,*,,DOCUMENTED,The queries used to search for files.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).queries[],body,string,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).results,body,array<object>,false,,,,,*,,DOCUMENTED,The results of the file search tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](file_search_call).results[],body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to a computer use tool. See the [computer use guide](https://developers.openai.com/api/docs/guides/tools-computer-use) for more information.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).type,body,string,true,computer_call,,,computer_call,*,,DOCUMENTED,The type of the computer call. Always `computer_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the computer call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used when responding to the tool call with output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action,body,object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(click),body,object,false,,,,,*,,DOCUMENTED,A click action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(double_click),body,object,false,,,,,*,,DOCUMENTED,A double click action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(drag),body,object,false,,,,,*,,DOCUMENTED,A drag action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(keypress),body,object,false,,,,,*,,DOCUMENTED,A collection of keypresses the model would like to perform.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(move),body,object,false,,,,,*,,DOCUMENTED,A mouse move action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(screenshot),body,object,false,,,,,*,,DOCUMENTED,A screenshot action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(scroll),body,object,false,,,,,*,,DOCUMENTED,A scroll action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(type),body,object,false,,,,,*,,DOCUMENTED,An action to type in text.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).action(wait),body,object,false,,,,,*,,DOCUMENTED,A wait action.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).actions,body,array<object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>>,false,,,,,*,,DOCUMENTED,Flattened batched actions for `computer_use`. Each action includes an `type` discriminator and action-specific fields.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).actions[],body,object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).pending_safety_checks,body,array<object>,true,,,,,*,,DOCUMENTED,The pending safety checks for the computer call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).pending_safety_checks[],body,object,false,,,,,*,,DOCUMENTED,A pending safety check for the computer call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a computer tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the computer tool call output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the computer tool call that produced the output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).type,body,string,true,computer_call_output,,,computer_call_output,*,,DOCUMENTED,The type of the computer tool call output. Always `computer_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).output,body,object,true,,,,,*,,DOCUMENTED,A computer screenshot image used with the computer use tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).output.type,body,string,true,computer_screenshot,,,computer_screenshot,*,,DOCUMENTED,"Specifies the event type. For a computer screenshot, this property is always set to `computer_screenshot`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).output.image_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the screenshot image.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).output.file_id,body,string,false,,,,,*,,DOCUMENTED,The identifier of an uploaded file that contains the screenshot.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).acknowledged_safety_checks,body,array<object>,false,,,,,*,,DOCUMENTED,The safety checks reported by the API that have been acknowledged by the developer.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).acknowledged_safety_checks[],body,object,false,,,,,*,,DOCUMENTED,A pending safety check for the computer call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](computer_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the message input. One of `in_progress`, `completed`, or `incomplete`. Populated when input items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a web search tool call. See the [web search guide](https://developers.openai.com/api/docs/guides/tools-web-search) for more information.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the web search tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).type,body,string,true,,,,web_search_call,*,,DOCUMENTED,The type of the web search tool call. Always `web_search_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).status,body,string,true,,,,in_progress|searching|completed|failed|incomplete,*,,DOCUMENTED,The status of the web search tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).action,body,object<search> | object<open_page> | object<find_in_page>,true,,,,,*,,DOCUMENTED,"An object describing the specific action taken in this web search call. Includes details on how the model used the web (search, open_page, find_in_page).",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).action(search),body,object,true,,,,,*,,DOCUMENTED,"Action type ""search"" - Performs a web search query.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).action(open_page),body,object,true,,,,,*,,DOCUMENTED,"Action type ""open_page"" - Opens a specific URL from search results.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](web_search_call).action(find_in_page),body,object,true,,,,,*,,DOCUMENTED,"Action type ""find_in_page"": Searches for a pattern within a loaded page.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a function. See the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling) for more information.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).type,body,string,true,,,,function_call,*,,DOCUMENTED,The type of the function tool call. Always `function_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the function to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments to pass to the function.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the function tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a function tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).type,body,string,true,function_call_output,,,function_call_output,*,,DOCUMENTED,The type of the function tool call output. Always `function_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,"Text, image, or file output of the function tool call.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).output[],body,array<object<input_text> | object<input_image> | object<input_file>>,false,,,,,*,,DOCUMENTED,"An array of content outputs (text, image, file) for the function tool call.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).name,body,string | null,false,,,,,*,,DOCUMENTED,The name of the tool that produced the output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).namespace,body,string | null,false,,,,,*,,DOCUMENTED,The namespace of the tool that produced the output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](function_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).type,body,string,true,tool_search_call,,,tool_search_call,*,,DOCUMENTED,The item type. Always `tool_search_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).arguments,body,object,true,,,,,*,,DOCUMENTED,The arguments supplied to the tool search call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).type,body,string,true,tool_search_output,,,tool_search_output,*,,DOCUMENTED,The item type. Always `tool_search_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,The loaded tool definitions returned by the tool search output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](tool_search_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this additional tools item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools).type,body,string,true,additional_tools,,,additional_tools,*,,DOCUMENTED,The item type. Always `additional_tools`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools).role,body,string,true,developer,,,developer,*,,DOCUMENTED,The role that provided the additional tools. Only `developer` is supported.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,A list of additional tools made available at this item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](additional_tools).tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](configuration_update),body,object,false,,,,,*,,DOCUMENTED,An update to the conversation's response configuration. The configuration remains in effect for subsequent responses until it is replaced by another configuration update.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](configuration_update).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the configuration update item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](configuration_update).type,body,string,true,configuration_update,,,configuration_update,*,,DOCUMENTED,The item type. Always `configuration_update`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](configuration_update).reasoning,body,object,false,,,,,*,,DOCUMENTED,Updates to reasoning configuration. Only effort is supported.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](configuration_update).reasoning.effort,body,string | null,false,medium,,,none|minimal|low|medium|high|xhigh|max,*,,DOCUMENTED,The reasoning effort to use for subsequent responses until another configuration update replaces it.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning),body,object,false,,,,,*,,DOCUMENTED,A description of the chain of thought used by a reasoning model while generating a response. Be sure to include these items in your `input` to the Responses API for subsequent turns of a conversation if you are manually [managing context](https://developers.openai.com/api/docs/guides/conversation-state).,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).type,body,string,true,,,,reasoning,*,,DOCUMENTED,The type of the object. Always `reasoning`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).id,body,string,true,,,,,*,,DOCUMENTED,The unique identifier of the reasoning content.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).encrypted_content,body,string | null,false,,,,,*,,DOCUMENTED,"The encrypted content of the reasoning item. This is populated by default for reasoning items returned by `POST /v1/responses` and WebSocket `response.create` requests. When streaming, use the completed reasoning item and its `encrypted_content` from the `response.output_item.done` event in subsequent requests. The `encrypted_content` in `response.output_item.added` may be incomplete. This is especially important when `store` is `false` or when using Zero Data Retention.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).summary,body,array<object>,true,,,,,*,,DOCUMENTED,Reasoning summary content.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).summary[],body,object,false,,,,,*,,DOCUMENTED,A summary text from the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).content,body,array<object>,false,,,,,*,,DOCUMENTED,Reasoning text content.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).content[],body,object,false,,,,,*,,DOCUMENTED,Reasoning text from the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](reasoning).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction),body,object,false,,,,,*,,DOCUMENTED,A compaction item generated by the [`v1/responses/compact` API](https://developers.openai.com/api/reference/resources/responses/methods/compact).,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction).id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the compaction item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction).type,body,string,true,compaction,,,compaction,*,,DOCUMENTED,The type of the item. Always `compaction`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction).encrypted_content,body,string,true,,,,,*,,DOCUMENTED,The encrypted content of the compaction summary.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call),body,object,false,,,,,*,,DOCUMENTED,An image generation request made by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).type,body,string,true,,,,image_generation_call,*,,DOCUMENTED,The type of the image generation call. Always `image_generation_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the image generation call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).status,body,string,true,,,,in_progress|completed|generating|failed,*,,DOCUMENTED,The status of the image generation call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).result,body,string | null,true,,,,,*,,DOCUMENTED,The generated image encoded in base64.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).size,body,string,false,,,,,*,,DOCUMENTED,"The image dimensions as a `WIDTHxHEIGHT` string, for example `1536x864`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).quality,body,string | null,false,,,,low|medium|high|xhigh|max|auto,*,,DOCUMENTED,"The quality of the image generated by the image generation tool call. One of `low`, `medium`, `high`, `xhigh`, `max`, or `auto`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).action,body,string | null,false,,,,generate|edit|auto,*,,DOCUMENTED,The action used for image generation.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).background,body,string | null,false,,,,transparent|opaque|auto,*,,DOCUMENTED,The background setting used for generation.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).output_format,body,string | null,false,,,,png|webp|jpeg,*,,DOCUMENTED,The output format used for generation.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](image_generation_call).revised_prompt,body,string | null,false,,,,,*,,DOCUMENTED,The prompt that was used after any model prompt rewriting.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run code.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).type,body,string,true,code_interpreter_call,,,code_interpreter_call,*,,DOCUMENTED,The type of the code interpreter tool call. Always `code_interpreter_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the code interpreter tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).status,body,string,true,,,,in_progress|completed|incomplete|interpreting|failed,*,,DOCUMENTED,"The status of the code interpreter tool call. Valid values are `in_progress`, `completed`, `incomplete`, `interpreting`, and `failed`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).container_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the container used to run the code.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).code,body,string | null,true,,,,,*,,DOCUMENTED,"The code to run, or null if not available.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).outputs,body,array<object<logs> | object<image>>,true,,,,,*,,DOCUMENTED,"The outputs generated by the code interpreter, such as logs or images. Can be null if no outputs are available.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](code_interpreter_call).outputs[],body,object<logs> | object<image>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a command on the local shell.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).type,body,string,true,,,,local_shell_call,*,,DOCUMENTED,The type of the local shell call. Always `local_shell_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action,body,object,true,,,,,*,,DOCUMENTED,Execute a shell command on the server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.type,body,string,true,exec,,,exec,*,,DOCUMENTED,The type of the local shell action. Always `exec`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.command,body,array<string>,true,,,,,*,,DOCUMENTED,The command to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.timeout_ms,body,integer | null,false,,,,,*,,DOCUMENTED,Optional timeout in milliseconds for the command.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.working_directory,body,string | null,false,,,,,*,,DOCUMENTED,Optional working directory to run the command in.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.env,body,"map<string,string>",true,,,,,*,,DOCUMENTED,Environment variables to set for the command.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).action.user,body,string | null,false,,,,,*,,DOCUMENTED,Optional user to run the command as.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the local shell call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a local shell tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call_output).type,body,string,true,,,,local_shell_call_output,*,,DOCUMENTED,The type of the local shell tool call output. Always `local_shell_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call_output).output,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the output of the local shell tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](local_shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool representing a request to execute one or more shell commands.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).type,body,string,true,shell_call,,,shell_call,*,,DOCUMENTED,The type of the item. Always `shell_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).action,body,object,true,,,,,*,,DOCUMENTED,The shell commands and limits that describe how to run the tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).action.commands,body,array<string>,true,,,,,*,,DOCUMENTED,Ordered shell commands for the execution environment to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).action.timeout_ms,body,integer | null,false,,,,,*,,DOCUMENTED,Maximum wall-clock time in milliseconds to allow the shell commands to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).action.max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,Maximum number of UTF-8 characters to capture from combined stdout and stderr output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the shell call. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).environment,body,object<local> | object<container_reference>,false,,,,,*,,DOCUMENTED,The environment to execute the shell commands in.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).environment(local),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call).environment(container_reference),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output items emitted by a shell tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).type,body,string,true,shell_call_output,,,shell_call_output,*,,DOCUMENTED,The type of the item. Always `shell_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).output,body,array<object>,true,,,,,*,,DOCUMENTED,"Captured chunks of stdout and stderr output, along with their associated outcomes.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).output[],body,object,false,,,,,*,,DOCUMENTED,Captured stdout and stderr for a portion of a shell tool call output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the shell call output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](shell_call_output).max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,The maximum number of UTF-8 characters captured for this shell call's combined output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call),body,object,false,,,,,*,,DOCUMENTED,"A tool call representing a request to create, delete, or update files using diff patches.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).type,body,string,true,apply_patch_call,,,apply_patch_call,*,,DOCUMENTED,The type of the item. Always `apply_patch_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).status,body,string,true,,,,in_progress|completed,*,,DOCUMENTED,The status of the apply patch tool call. One of `in_progress` or `completed`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).operation,body,object<create_file> | object<delete_file> | object<update_file>,true,,,,,*,,DOCUMENTED,"The specific create, delete, or update instruction for the apply_patch tool call.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).operation(create_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for creating a new file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).operation(delete_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for deleting an existing file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call).operation(update_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for updating an existing file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output emitted by an apply patch tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).type,body,string,true,apply_patch_call_output,,,apply_patch_call_output,*,,DOCUMENTED,The type of the item. Always `apply_patch_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).status,body,string,true,,,,completed|failed,*,,DOCUMENTED,The status of the apply patch tool call output. One of `completed` or `failed`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](apply_patch_call_output).output,body,string | null,false,,,,,*,,DOCUMENTED,"Optional human-readable log text from the apply patch tool (e.g., patch results or errors).",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools),body,object,false,,,,,*,,DOCUMENTED,A list of tools available on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).type,body,string,true,,,,mcp_list_tools,*,,DOCUMENTED,The type of the item. Always `mcp_list_tools`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the list.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).tools,body,array<object>,true,,,,,*,,DOCUMENTED,The tools available on the server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).tools[],body,object,false,,,,,*,,DOCUMENTED,A tool available on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_list_tools).error,body,string | null,false,,,,,*,,DOCUMENTED,Error message if the server could not list tools.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request),body,object,false,,,,,*,,DOCUMENTED,A request for human approval of a tool invocation.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request).type,body,string,true,,,,mcp_approval_request,*,,DOCUMENTED,The type of the item. Always `mcp_approval_request`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the approval request.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server making the request.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool to run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_request).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of arguments for the tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response),body,object,false,,,,,*,,DOCUMENTED,A response to an MCP approval request.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response).type,body,string,true,,,,mcp_approval_response,*,,DOCUMENTED,The type of the item. Always `mcp_approval_response`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the approval response,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response).approval_request_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the approval request being answered.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response).approve,body,boolean,true,,,,,*,,DOCUMENTED,Whether the request was approved.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_approval_response).reason,body,string | null,false,,,,,*,,DOCUMENTED,Optional reason for the decision.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call),body,object,false,,,,,*,,DOCUMENTED,An invocation of a tool on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).type,body,string,true,,,,mcp_call,*,,DOCUMENTED,The type of the item. Always `mcp_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server running the tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool that was run.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments passed to the tool.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).output,body,string | null,false,,,,,*,,DOCUMENTED,The output from the tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).error,body,object<mcp_protocol_error> | object<mcp_tool_execution_error> | object<http_error>,false,,,,,*,,DOCUMENTED,"The error from the tool call, if any.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).error(mcp_protocol_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).error(mcp_tool_execution_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).error(http_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).status,body,string,false,,,,in_progress|completed|incomplete|calling|failed,*,,DOCUMENTED,"The status of the tool call. One of `in_progress`, `completed`, `incomplete`, `calling`, or `failed`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](mcp_call).approval_request_id,body,string | null,false,,,,,*,,DOCUMENTED,Unique identifier for the MCP tool call approval request. Include this value in a subsequent `mcp_approval_response` input to approve or reject the corresponding tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output),body,object,false,,,,,*,,DOCUMENTED,"The output of a custom tool call from your code, being sent back to the model.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).type,body,string,true,,,,custom_tool_call_output,*,,DOCUMENTED,The type of the custom tool call output. Always `custom_tool_call_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call output in the OpenAI platform.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,"The call ID, used to map this custom tool call output to a custom tool call.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,The output from the custom tool call generated by your code. Can be a string or an list of output content.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call_output).output[],body,array<object<input_text> | object<input_image> | object<input_file>>,false,,,,,*,,DOCUMENTED,"Text, image, or file output of the custom tool call.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).type,body,string,true,,,,custom_tool_call,*,,DOCUMENTED,The type of the custom tool call. Always `custom_tool_call`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call in the OpenAI platform.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used to map this custom tool call to a tool call output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the custom tool being called.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool being called.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).input,body,string,true,,,,,*,,DOCUMENTED,The input for the custom tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](custom_tool_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the custom tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction_trigger),body,object,false,,,,,*,,DOCUMENTED,Compacts the current context. Must be the final input item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction_trigger).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this compaction trigger.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](compaction_trigger).type,body,string,true,compaction_trigger,,,compaction_trigger,*,,DOCUMENTED,The type of the item. Always `compaction_trigger`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](item_reference),body,object,false,,,,,*,,DOCUMENTED,An internal identifier for an item to reference.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](item_reference).type,body,string | null,false,item_reference,,,item_reference,*,,DOCUMENTED,The type of item to reference. Always `item_reference`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](item_reference).id,body,string,true,,,,,*,,DOCUMENTED,The ID of the item to reference.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program).type,body,string,true,program,,,program,*,,DOCUMENTED,The item type. Always `program`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program).call_id,body,string,true,,,,,*,,DOCUMENTED,The stable call ID of the program item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program).code,body,string,true,,,,,*,,DOCUMENTED,The JavaScript source executed by programmatic tool calling.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program).fingerprint,body,string,true,,,,,*,,DOCUMENTED,Opaque program replay fingerprint that must be round-tripped.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program output item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output).type,body,string,true,program_output,,,program_output,*,,DOCUMENTED,The item type. Always `program_output`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The call ID of the program item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output).result,body,string,true,,,,,*,,DOCUMENTED,The result produced by the program item.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](program_output).status,body,string,true,,,,completed|incomplete,*,,DOCUMENTED,The terminal status of the program output.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_text),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A text input to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_text).type,body,string,true,input_text,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,The type of the input item. Always `input_text`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_text).text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text input to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image),body,object,false,,,,,*,,DOCUMENTED,An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).type,body,string,true,input_image,,,input_image,*,,DOCUMENTED,The type of the input item. Always `input_image`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).image_url,body,string | null,false,,,,,*,,DOCUMENTED,The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).detail,body,string,true,,,,low|high|auto|original,*,,DOCUMENTED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_image).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file),body,object,false,,,,,*,,DOCUMENTED,A file input to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).type,body,string,true,input_file,,,input_file,*,,DOCUMENTED,The type of the input item. Always `input_file`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).filename,body,string,false,,,,,*,,DOCUMENTED,The name of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).file_data,body,string,false,,,,,*,,DOCUMENTED,The content of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).file_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations,items[](message).content[](input_file).detail,body,string,false,,,,auto|low|high,*,,DOCUMENTED,"The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to render the file at higher quality. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/conversations/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items,body,array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Initial items to include in the conversation context. You may add up to 20 items at a time.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[],body,object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An item representing part of the context for the response to be generated by the model. Can contain text, images, and audio inputs, as well as previous assistant responses and tool call outputs.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A message input to the model with a role indicating instruction following hierarchy. Instructions given with the `developer` or `system` role take precedence over instructions given with the `user` role. Messages with the `assistant` role are presumed to have been generated by the model in previous interactions.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).role,body,string,true,,,,user|assistant|system|developer,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the message input. One of `user`, `assistant`, `system`, or `developer`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content,body,string | array<object<input_text> | object<input_image> | object<input_file>> | array<object<output_text> | object<refusal>>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or audio input to the model, used to generate a response. Can also contain previous assistant responses.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[],body,array<object<input_text> | object<input_image> | object<input_file>> | object<input_text> | object<input_file> | object<output_text> | object<refusal>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).phase,body,string | null,false,,,,commentary|final_answer,*,,DOCUMENTED,"Labels an `assistant` message as intermediate commentary (`commentary`) or the final answer (`final_answer`). For models like `gpt-5.3-codex` and beyond, when sending follow-up requests, preserve and resend phase on all assistant messages — dropping it can degrade performance. Not used for user messages.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).type,body,string,true,,,,message,*,,DOCUMENTED|LIVE_VERIFIED,The type of the message input. Always `message`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the output message.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a file search tool call. See the [file search guide](https://developers.openai.com/api/docs/guides/tools-file-search) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the file search tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).type,body,string,true,,,,file_search_call,*,,DOCUMENTED,The type of the file search tool call. Always `file_search_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).status,body,string,true,,,,in_progress|searching|completed|incomplete|failed,*,,DOCUMENTED,"The status of the file search tool call. One of `in_progress`, `searching`, `incomplete` or `failed`,",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).queries,body,array<string>,true,,,,,*,,DOCUMENTED,The queries used to search for files.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).queries[],body,string,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).results,body,array<object>,false,,,,,*,,DOCUMENTED,The results of the file search tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](file_search_call).results[],body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to a computer use tool. See the [computer use guide](https://developers.openai.com/api/docs/guides/tools-computer-use) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).type,body,string,true,computer_call,,,computer_call,*,,DOCUMENTED,The type of the computer call. Always `computer_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the computer call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used when responding to the tool call with output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action,body,object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(click),body,object,false,,,,,*,,DOCUMENTED,A click action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(double_click),body,object,false,,,,,*,,DOCUMENTED,A double click action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(drag),body,object,false,,,,,*,,DOCUMENTED,A drag action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(keypress),body,object,false,,,,,*,,DOCUMENTED,A collection of keypresses the model would like to perform.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(move),body,object,false,,,,,*,,DOCUMENTED,A mouse move action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(screenshot),body,object,false,,,,,*,,DOCUMENTED,A screenshot action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(scroll),body,object,false,,,,,*,,DOCUMENTED,A scroll action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(type),body,object,false,,,,,*,,DOCUMENTED,An action to type in text.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).action(wait),body,object,false,,,,,*,,DOCUMENTED,A wait action.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).actions,body,array<object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>>,false,,,,,*,,DOCUMENTED,Flattened batched actions for `computer_use`. Each action includes an `type` discriminator and action-specific fields.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).actions[],body,object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).pending_safety_checks,body,array<object>,true,,,,,*,,DOCUMENTED,The pending safety checks for the computer call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).pending_safety_checks[],body,object,false,,,,,*,,DOCUMENTED,A pending safety check for the computer call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a computer tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the computer tool call output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the computer tool call that produced the output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).type,body,string,true,computer_call_output,,,computer_call_output,*,,DOCUMENTED,The type of the computer tool call output. Always `computer_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).output,body,object,true,,,,,*,,DOCUMENTED,A computer screenshot image used with the computer use tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).output.type,body,string,true,computer_screenshot,,,computer_screenshot,*,,DOCUMENTED,"Specifies the event type. For a computer screenshot, this property is always set to `computer_screenshot`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).output.image_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the screenshot image.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).output.file_id,body,string,false,,,,,*,,DOCUMENTED,The identifier of an uploaded file that contains the screenshot.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).acknowledged_safety_checks,body,array<object>,false,,,,,*,,DOCUMENTED,The safety checks reported by the API that have been acknowledged by the developer.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).acknowledged_safety_checks[],body,object,false,,,,,*,,DOCUMENTED,A pending safety check for the computer call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](computer_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the message input. One of `in_progress`, `completed`, or `incomplete`. Populated when input items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a web search tool call. See the [web search guide](https://developers.openai.com/api/docs/guides/tools-web-search) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the web search tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).type,body,string,true,,,,web_search_call,*,,DOCUMENTED,The type of the web search tool call. Always `web_search_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).status,body,string,true,,,,in_progress|searching|completed|failed|incomplete,*,,DOCUMENTED,The status of the web search tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).action,body,object<search> | object<open_page> | object<find_in_page>,true,,,,,*,,DOCUMENTED,"An object describing the specific action taken in this web search call. Includes details on how the model used the web (search, open_page, find_in_page).",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).action(search),body,object,true,,,,,*,,DOCUMENTED,"Action type ""search"" - Performs a web search query.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).action(open_page),body,object,true,,,,,*,,DOCUMENTED,"Action type ""open_page"" - Opens a specific URL from search results.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](web_search_call).action(find_in_page),body,object,true,,,,,*,,DOCUMENTED,"Action type ""find_in_page"": Searches for a pattern within a loaded page.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a function. See the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).type,body,string,true,,,,function_call,*,,DOCUMENTED,The type of the function tool call. Always `function_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the function to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments to pass to the function.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the function tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a function tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).type,body,string,true,function_call_output,,,function_call_output,*,,DOCUMENTED,The type of the function tool call output. Always `function_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,"Text, image, or file output of the function tool call.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).output[],body,array<object<input_text> | object<input_image> | object<input_file>>,false,,,,,*,,DOCUMENTED,"An array of content outputs (text, image, file) for the function tool call.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).name,body,string | null,false,,,,,*,,DOCUMENTED,The name of the tool that produced the output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).namespace,body,string | null,false,,,,,*,,DOCUMENTED,The namespace of the tool that produced the output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](function_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).type,body,string,true,tool_search_call,,,tool_search_call,*,,DOCUMENTED,The item type. Always `tool_search_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).arguments,body,object,true,,,,,*,,DOCUMENTED,The arguments supplied to the tool search call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).type,body,string,true,tool_search_output,,,tool_search_output,*,,DOCUMENTED,The item type. Always `tool_search_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,The loaded tool definitions returned by the tool search output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](tool_search_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this additional tools item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools).type,body,string,true,additional_tools,,,additional_tools,*,,DOCUMENTED,The item type. Always `additional_tools`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools).role,body,string,true,developer,,,developer,*,,DOCUMENTED,The role that provided the additional tools. Only `developer` is supported.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,A list of additional tools made available at this item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](additional_tools).tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](configuration_update),body,object,false,,,,,*,,DOCUMENTED,An update to the conversation's response configuration. The configuration remains in effect for subsequent responses until it is replaced by another configuration update.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](configuration_update).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the configuration update item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](configuration_update).type,body,string,true,configuration_update,,,configuration_update,*,,DOCUMENTED,The item type. Always `configuration_update`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](configuration_update).reasoning,body,object,false,,,,,*,,DOCUMENTED,Updates to reasoning configuration. Only effort is supported.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](configuration_update).reasoning.effort,body,string | null,false,medium,,,none|minimal|low|medium|high|xhigh|max,*,,DOCUMENTED,The reasoning effort to use for subsequent responses until another configuration update replaces it.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning),body,object,false,,,,,*,,DOCUMENTED,A description of the chain of thought used by a reasoning model while generating a response. Be sure to include these items in your `input` to the Responses API for subsequent turns of a conversation if you are manually [managing context](https://developers.openai.com/api/docs/guides/conversation-state).,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).type,body,string,true,,,,reasoning,*,,DOCUMENTED,The type of the object. Always `reasoning`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).id,body,string,true,,,,,*,,DOCUMENTED,The unique identifier of the reasoning content.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).encrypted_content,body,string | null,false,,,,,*,,DOCUMENTED,"The encrypted content of the reasoning item. This is populated by default for reasoning items returned by `POST /v1/responses` and WebSocket `response.create` requests. When streaming, use the completed reasoning item and its `encrypted_content` from the `response.output_item.done` event in subsequent requests. The `encrypted_content` in `response.output_item.added` may be incomplete. This is especially important when `store` is `false` or when using Zero Data Retention.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).summary,body,array<object>,true,,,,,*,,DOCUMENTED,Reasoning summary content.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).summary[],body,object,false,,,,,*,,DOCUMENTED,A summary text from the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).content,body,array<object>,false,,,,,*,,DOCUMENTED,Reasoning text content.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).content[],body,object,false,,,,,*,,DOCUMENTED,Reasoning text from the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](reasoning).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction),body,object,false,,,,,*,,DOCUMENTED,A compaction item generated by the [`v1/responses/compact` API](https://developers.openai.com/api/reference/resources/responses/methods/compact).,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction).id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the compaction item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction).type,body,string,true,compaction,,,compaction,*,,DOCUMENTED,The type of the item. Always `compaction`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction).encrypted_content,body,string,true,,,,,*,,DOCUMENTED,The encrypted content of the compaction summary.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call),body,object,false,,,,,*,,DOCUMENTED,An image generation request made by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).type,body,string,true,,,,image_generation_call,*,,DOCUMENTED,The type of the image generation call. Always `image_generation_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the image generation call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).status,body,string,true,,,,in_progress|completed|generating|failed,*,,DOCUMENTED,The status of the image generation call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).result,body,string | null,true,,,,,*,,DOCUMENTED,The generated image encoded in base64.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).size,body,string,false,,,,,*,,DOCUMENTED,"The image dimensions as a `WIDTHxHEIGHT` string, for example `1536x864`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).quality,body,string | null,false,,,,low|medium|high|xhigh|max|auto,*,,DOCUMENTED,"The quality of the image generated by the image generation tool call. One of `low`, `medium`, `high`, `xhigh`, `max`, or `auto`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).action,body,string | null,false,,,,generate|edit|auto,*,,DOCUMENTED,The action used for image generation.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).background,body,string | null,false,,,,transparent|opaque|auto,*,,DOCUMENTED,The background setting used for generation.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).output_format,body,string | null,false,,,,png|webp|jpeg,*,,DOCUMENTED,The output format used for generation.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](image_generation_call).revised_prompt,body,string | null,false,,,,,*,,DOCUMENTED,The prompt that was used after any model prompt rewriting.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run code.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).type,body,string,true,code_interpreter_call,,,code_interpreter_call,*,,DOCUMENTED,The type of the code interpreter tool call. Always `code_interpreter_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the code interpreter tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).status,body,string,true,,,,in_progress|completed|incomplete|interpreting|failed,*,,DOCUMENTED,"The status of the code interpreter tool call. Valid values are `in_progress`, `completed`, `incomplete`, `interpreting`, and `failed`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).container_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the container used to run the code.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).code,body,string | null,true,,,,,*,,DOCUMENTED,"The code to run, or null if not available.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).outputs,body,array<object<logs> | object<image>>,true,,,,,*,,DOCUMENTED,"The outputs generated by the code interpreter, such as logs or images. Can be null if no outputs are available.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](code_interpreter_call).outputs[],body,object<logs> | object<image>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a command on the local shell.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).type,body,string,true,,,,local_shell_call,*,,DOCUMENTED,The type of the local shell call. Always `local_shell_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action,body,object,true,,,,,*,,DOCUMENTED,Execute a shell command on the server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.type,body,string,true,exec,,,exec,*,,DOCUMENTED,The type of the local shell action. Always `exec`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.command,body,array<string>,true,,,,,*,,DOCUMENTED,The command to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.timeout_ms,body,integer | null,false,,,,,*,,DOCUMENTED,Optional timeout in milliseconds for the command.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.working_directory,body,string | null,false,,,,,*,,DOCUMENTED,Optional working directory to run the command in.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.env,body,"map<string,string>",true,,,,,*,,DOCUMENTED,Environment variables to set for the command.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).action.user,body,string | null,false,,,,,*,,DOCUMENTED,Optional user to run the command as.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the local shell call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a local shell tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call_output).type,body,string,true,,,,local_shell_call_output,*,,DOCUMENTED,The type of the local shell tool call output. Always `local_shell_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call_output).output,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the output of the local shell tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](local_shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool representing a request to execute one or more shell commands.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).type,body,string,true,shell_call,,,shell_call,*,,DOCUMENTED,The type of the item. Always `shell_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).action,body,object,true,,,,,*,,DOCUMENTED,The shell commands and limits that describe how to run the tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).action.commands,body,array<string>,true,,,,,*,,DOCUMENTED,Ordered shell commands for the execution environment to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).action.timeout_ms,body,integer | null,false,,,,,*,,DOCUMENTED,Maximum wall-clock time in milliseconds to allow the shell commands to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).action.max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,Maximum number of UTF-8 characters to capture from combined stdout and stderr output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the shell call. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).environment,body,object<local> | object<container_reference>,false,,,,,*,,DOCUMENTED,The environment to execute the shell commands in.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).environment(local),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call).environment(container_reference),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output items emitted by a shell tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).type,body,string,true,shell_call_output,,,shell_call_output,*,,DOCUMENTED,The type of the item. Always `shell_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).output,body,array<object>,true,,,,,*,,DOCUMENTED,"Captured chunks of stdout and stderr output, along with their associated outcomes.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).output[],body,object,false,,,,,*,,DOCUMENTED,Captured stdout and stderr for a portion of a shell tool call output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the shell call output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](shell_call_output).max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,The maximum number of UTF-8 characters captured for this shell call's combined output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call),body,object,false,,,,,*,,DOCUMENTED,"A tool call representing a request to create, delete, or update files using diff patches.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).type,body,string,true,apply_patch_call,,,apply_patch_call,*,,DOCUMENTED,The type of the item. Always `apply_patch_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).status,body,string,true,,,,in_progress|completed,*,,DOCUMENTED,The status of the apply patch tool call. One of `in_progress` or `completed`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).operation,body,object<create_file> | object<delete_file> | object<update_file>,true,,,,,*,,DOCUMENTED,"The specific create, delete, or update instruction for the apply_patch tool call.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).operation(create_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for creating a new file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).operation(delete_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for deleting an existing file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call).operation(update_file),body,object,true,,,,,*,,DOCUMENTED,Instruction for updating an existing file via the apply_patch tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output emitted by an apply patch tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).type,body,string,true,apply_patch_call_output,,,apply_patch_call_output,*,,DOCUMENTED,The type of the item. Always `apply_patch_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).status,body,string,true,,,,completed|failed,*,,DOCUMENTED,The status of the apply patch tool call output. One of `completed` or `failed`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](apply_patch_call_output).output,body,string | null,false,,,,,*,,DOCUMENTED,"Optional human-readable log text from the apply patch tool (e.g., patch results or errors).",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools),body,object,false,,,,,*,,DOCUMENTED,A list of tools available on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).type,body,string,true,,,,mcp_list_tools,*,,DOCUMENTED,The type of the item. Always `mcp_list_tools`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the list.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).tools,body,array<object>,true,,,,,*,,DOCUMENTED,The tools available on the server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).tools[],body,object,false,,,,,*,,DOCUMENTED,A tool available on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_list_tools).error,body,string | null,false,,,,,*,,DOCUMENTED,Error message if the server could not list tools.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request),body,object,false,,,,,*,,DOCUMENTED,A request for human approval of a tool invocation.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request).type,body,string,true,,,,mcp_approval_request,*,,DOCUMENTED,The type of the item. Always `mcp_approval_request`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the approval request.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server making the request.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool to run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_request).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of arguments for the tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response),body,object,false,,,,,*,,DOCUMENTED,A response to an MCP approval request.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response).type,body,string,true,,,,mcp_approval_response,*,,DOCUMENTED,The type of the item. Always `mcp_approval_response`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the approval response,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response).approval_request_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the approval request being answered.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response).approve,body,boolean,true,,,,,*,,DOCUMENTED,Whether the request was approved.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_approval_response).reason,body,string | null,false,,,,,*,,DOCUMENTED,Optional reason for the decision.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call),body,object,false,,,,,*,,DOCUMENTED,An invocation of a tool on an MCP server.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).type,body,string,true,,,,mcp_call,*,,DOCUMENTED,The type of the item. Always `mcp_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server running the tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool that was run.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments passed to the tool.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).output,body,string | null,false,,,,,*,,DOCUMENTED,The output from the tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).error,body,object<mcp_protocol_error> | object<mcp_tool_execution_error> | object<http_error>,false,,,,,*,,DOCUMENTED,"The error from the tool call, if any.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).error(mcp_protocol_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).error(mcp_tool_execution_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).error(http_error),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).status,body,string,false,,,,in_progress|completed|incomplete|calling|failed,*,,DOCUMENTED,"The status of the tool call. One of `in_progress`, `completed`, `incomplete`, `calling`, or `failed`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](mcp_call).approval_request_id,body,string | null,false,,,,,*,,DOCUMENTED,Unique identifier for the MCP tool call approval request. Include this value in a subsequent `mcp_approval_response` input to approve or reject the corresponding tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output),body,object,false,,,,,*,,DOCUMENTED,"The output of a custom tool call from your code, being sent back to the model.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).type,body,string,true,,,,custom_tool_call_output,*,,DOCUMENTED,The type of the custom tool call output. Always `custom_tool_call_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call output in the OpenAI platform.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,"The call ID, used to map this custom tool call output to a custom tool call.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,The output from the custom tool call generated by your code. Can be a string or an list of output content.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call_output).output[],body,array<object<input_text> | object<input_image> | object<input_file>>,false,,,,,*,,DOCUMENTED,"Text, image, or file output of the custom tool call.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).type,body,string,true,,,,custom_tool_call,*,,DOCUMENTED,The type of the custom tool call. Always `custom_tool_call`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call in the OpenAI platform.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used to map this custom tool call to a tool call output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).caller(direct),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).caller(program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the custom tool being called.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool being called.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).input,body,string,true,,,,,*,,DOCUMENTED,The input for the custom tool call generated by the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](custom_tool_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the custom tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction_trigger),body,object,false,,,,,*,,DOCUMENTED,Compacts the current context. Must be the final input item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction_trigger).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this compaction trigger.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](compaction_trigger).type,body,string,true,compaction_trigger,,,compaction_trigger,*,,DOCUMENTED,The type of the item. Always `compaction_trigger`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](item_reference),body,object,false,,,,,*,,DOCUMENTED,An internal identifier for an item to reference.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](item_reference).type,body,string | null,false,item_reference,,,item_reference,*,,DOCUMENTED,The type of item to reference. Always `item_reference`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](item_reference).id,body,string,true,,,,,*,,DOCUMENTED,The ID of the item to reference.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program).type,body,string,true,program,,,program,*,,DOCUMENTED,The item type. Always `program`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program).call_id,body,string,true,,,,,*,,DOCUMENTED,The stable call ID of the program item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program).code,body,string,true,,,,,*,,DOCUMENTED,The JavaScript source executed by programmatic tool calling.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program).fingerprint,body,string,true,,,,,*,,DOCUMENTED,Opaque program replay fingerprint that must be round-tripped.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program output item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output).type,body,string,true,program_output,,,program_output,*,,DOCUMENTED,The item type. Always `program_output`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The call ID of the program item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output).result,body,string,true,,,,,*,,DOCUMENTED,The result produced by the program item.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](program_output).status,body,string,true,,,,completed|incomplete,*,,DOCUMENTED,The terminal status of the program output.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_text),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A text input to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_text).type,body,string,true,input_text,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,The type of the input item. Always `input_text`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_text).text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text input to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image),body,object,false,,,,,*,,DOCUMENTED,An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).type,body,string,true,input_image,,,input_image,*,,DOCUMENTED,The type of the input item. Always `input_image`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).image_url,body,string | null,false,,,,,*,,DOCUMENTED,The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).detail,body,string,true,,,,low|high|auto|original,*,,DOCUMENTED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_image).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file),body,object,false,,,,,*,,DOCUMENTED,A file input to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).type,body,string,true,input_file,,,input_file,*,,DOCUMENTED,The type of the input item. Always `input_file`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).filename,body,string,false,,,,,*,,DOCUMENTED,The name of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).file_data,body,string,false,,,,,*,,DOCUMENTED,The content of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).file_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id}/items,items[](message).content[](input_file).detail,body,string,false,,,,auto|low|high,*,,DOCUMENTED,"The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to render the file at higher quality. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,POST /v1/conversations/{conversation_id},metadata,body,"map<string,string>",true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard. Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.",https://developers.openai.com/api/reference/resources/conversations/methods/update
openai,GET /v1/conversations/{conversation_id}/items,limit,query,integer,false,20,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list
openai,GET /v1/conversations/{conversation_id}/items,order,query,string,false,,,,asc|desc,*,,DOCUMENTED|LIVE_VERIFIED,The order to return the input items in. Default is `desc`. - `asc`: Return the input items in ascending order. - `desc`: Return the input items in descending order.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list
openai,GET /v1/conversations/{conversation_id}/items,after,query,string,false,,,,,*,,DOCUMENTED,"An item ID to list items after, used in pagination.",https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list
openai,GET /v1/conversations/{conversation_id}/items,include,query,array<string>,false,,,,,*,,DOCUMENTED,Specify additional output data to include in the model response. Currently supported values are: - `web_search_call.action.sources`: Include the sources of the web search tool call. - `code_interpreter_call.outputs`: Includes the outputs of python code execution in code interpreter tool call items. - `computer_call_output.output.image_url`: Include image urls from the computer call output. - `file_search_call.results`: Include the search results of the file search tool call. - `message.input_image.image_url`: Include image urls from the input message. - `message.output_text.logprobs`: Include logprobs with assistant messages. - `reasoning.encrypted_content`: Includes an encrypted version of reasoning tokens in reasoning item outputs. This enables reasoning items to be used in multi-turn conversations when using the Responses API statelessly (like when the `store` parameter is set to `fal,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list
openai,POST /v1/conversations/{conversation_id}/items,include,query,array<string>,false,,,,,*,,DOCUMENTED,Additional fields to include in the response. See the `include` parameter for [listing Conversation items above](https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list#%28resource%29%20conversations.items%20%3E%20%28method%29%20list%20%3E%20%28params%29%20default%20%3E%20%28param%29%20include%20%3E%20%28schema%29) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/create
openai,GET /v1/conversations/{conversation_id}/items/{item_id},include,query,array<string>,false,,,,,*,,DOCUMENTED,Additional fields to include in the response. See the `include` parameter for [listing Conversation items above](https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/list#%28resource%29%20conversations.items%20%3E%20%28method%29%20list%20%3E%20%28params%29%20default%20%3E%20%28param%29%20include%20%3E%20%28schema%29) for more information.,https://developers.openai.com/api/reference/resources/conversations/subresources/items/methods/retrieve
openai,POST /v1/embeddings,input,body,string | string[] | integer[] | integer[][],true,,1,2048,,text-embedding-3-small|text-embedding-3-large|text-embedding-ada-002,,DOCUMENTED|LIVE_VERIFIED,"Text(s) or token array(s) to embed. Non-empty; <= 8192 tokens per input; arrays 1-2048 items; <= 300,000 tokens summed per request. Tokenizer cl100k_base.",https://developers.openai.com/api/reference/resources/embeddings
openai,POST /v1/embeddings,model,body,string,true,,,,text-embedding-3-small|text-embedding-3-large|text-embedding-ada-002,text-embedding-3-small|text-embedding-3-large|text-embedding-ada-002,,DOCUMENTED|LIVE_VERIFIED,text-embedding-3-small | text-embedding-3-large | text-embedding-ada-002.,https://developers.openai.com/api/reference/resources/embeddings
openai,POST /v1/embeddings,dimensions,body,integer,false,,1,3072,,text-embedding-3-small|text-embedding-3-large,,DOCUMENTED|LIVE_VERIFIED,"Output dimensions (Matryoshka truncation, re-normalized by the API). text-embedding-3-* only; 1..native size (1536 small / 3072 large).",https://developers.openai.com/api/reference/resources/embeddings
openai,POST /v1/embeddings,encoding_format,body,string,false,float,,,float|base64,text-embedding-3-small|text-embedding-3-large|text-embedding-ada-002,,DOCUMENTED|LIVE_VERIFIED,"float (default, JSON numbers) | base64 (little-endian float32 bytes, smaller payload).",https://developers.openai.com/api/reference/resources/embeddings
openai,POST /v1/embeddings,user,body,string,false,,,,,text-embedding-3-small|text-embedding-3-large|text-embedding-ada-002,,DOCUMENTED,End-user identifier for abuse monitoring.,https://developers.openai.com/api/reference/resources/embeddings
openai,GET /v1/evals,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last eval from the previous pagination request.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals,limit,query,integer,false,20,,,,,,DOCUMENTED|DEPRECATED,Number of evals to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED|DEPRECATED,Sort order for evals by timestamp. Use `asc` for ascending order or `desc` for descending order.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals,order_by,query,string,false,created_at,,,created_at|updated_at,,,DOCUMENTED|DEPRECATED,"Evals can be ordered by creation time or last updated time. Use
`created_at` for creation time or `updated_at` for last updated time.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,name,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The name of the evaluation.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,metadata,body,object | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: Metadata],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,data_source_config,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,The configuration for the data source used for the evaluation runs. Dictates the schema of the data used in the evaluation.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,data_source_config.type,body,string,true,stored_completions,,,stored_completions,,,DOCUMENTED|DEPRECATED,The type of data source. Always `stored_completions`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,data_source_config.item_schema,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,The json schema for each row in the data source.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,data_source_config.include_sample_schema,body,boolean,false,,,,,,,DOCUMENTED|DEPRECATED,"Whether the eval should expect you to populate the sample namespace (ie, by generating responses off of your data source)",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,data_source_config.metadata,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,Metadata filters for the logs data source.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,"A list of graders for all eval runs in this group. Graders can reference variables in the data source using double curly braces notation, like `{{item.variable_name}}`. To reference the model's output, use the `sample` namespace (ie, `{{sample.output_text}}`).",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].type,body,string,true,,,,score_model,,,DOCUMENTED|DEPRECATED,"The object type, which is always `score_model`.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation. Must support structured outputs.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].input,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,"A list of chat messages forming the prompt or context. May include variable references to the `item` namespace, ie {{item.name}}.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].input[].role,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The role of the message (e.g. ""system"", ""assistant"", ""user"").",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].input[].content,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The content of the message.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels to classify to each item in the evaluation.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].passing_labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels that indicate a passing result. Must be a subset of labels.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].pass_threshold,body,number,true,,,,,,,DOCUMENTED|DEPRECATED,The threshold for the score.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].sampling_params.reasoning_effort,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].input,body,array<EvalItem>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals,testing_criteria[].range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to update.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id},name,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,Rename the evaluation.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id},metadata,body,object | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: Metadata],https://developers.openai.com/api/reference/resources/evals
openai,DELETE /v1/evals/{eval_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to delete.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs,eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to retrieve runs for.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last run from the previous pagination request.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs,limit,query,integer,false,20,,,,,,DOCUMENTED|DEPRECATED,Number of runs to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED|DEPRECATED,Sort order for runs by timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `asc`.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs,status,query,string,false,,,,queued|in_progress|completed|canceled|failed,,,DOCUMENTED|DEPRECATED,Filter runs by status. One of `queued` | `in_progress` | `failed` | `completed` | `canceled`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to create a run for.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,name,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The name of the run.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,metadata,body,object | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: Metadata],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,Details about the run's data source.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.type,body,string,true,responses,,,responses,,,DOCUMENTED|DEPRECATED,The type of run data source. Always `responses`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source,body,object | object,true,,,,,,,DOCUMENTED|DEPRECATED,Determines what populates the `item` namespace in the data source.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.type,body,string,true,,,,responses,,,DOCUMENTED|DEPRECATED,The type of run data source. Always `responses`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.content,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,The content of the jsonl file.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.content[].item,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.content[].sample,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.id,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The identifier of the file.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages,body,object | object,false,,,,,,,DOCUMENTED|DEPRECATED,"Used when sampling from a model. Dictates the structure of the messages passed into the model. Can either be a reference to a prebuilt trajectory (ie, `item.input_trajectory`), or a template with variable references to the `item` namespace.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages.type,body,string,true,,,,item_reference,,,DOCUMENTED|DEPRECATED,The type of input messages. Always `item_reference`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages.template,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,"A list of chat messages forming the prompt or context. May include variable references to the `item` namespace, ie {{item.name}}.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages.item_reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"A reference to a variable in the `item` namespace. Ie, ""item.input_trajectory""",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.reasoning_effort,body,string | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.temperature,body,number,false,1,,,,,,DOCUMENTED|DEPRECATED,A higher temperature increases randomness in the outputs.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.max_completion_tokens,body,integer,false,,,,,,,DOCUMENTED|DEPRECATED,The maximum number of tokens in the generated output.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.top_p,body,number,false,1,,,,,,DOCUMENTED|DEPRECATED,An alternative to temperature for nucleus sampling; 1.0 includes all tokens.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.seed,body,integer,false,42,,,,,,DOCUMENTED|DEPRECATED,"A seed value to initialize the randomness, during sampling.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format,body,object | object | object,false,,,,,,,DOCUMENTED|DEPRECATED,"An object specifying the format that the model must output.

Setting to `{ ""type"": ""json_schema"", ""json_schema"": {...} }` enables
Structured Outputs which ensures the model will match your supplied JSON
schema. Learn more in the [Structured Outputs
guide](https://developers.openai.com/api/docs/guides/structured-outputs).

Setting to `{ ""type"": ""json_object"" }` enables the older JSON mode, which
ensures the message the model generates is valid JSON. Using `json_schema`
is preferred for models that support it.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.type,body,string,true,,,,json_object,,,DOCUMENTED|DEPRECATED,The type of response format being defined. Always `json_object`.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.json_schema,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,"Structured Outputs configuration options, including a JSON Schema.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.json_schema.description,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"A description of what the response format is for, used by the model to
determine how to respond in the format.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.json_schema.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The name of the response format. Must be a-z, A-Z, 0-9, or contain
underscores and dashes, with a maximum length of 64.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.json_schema.schema,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ResponseFormatJsonSchemaSchema],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.response_format.json_schema.strict,body,boolean | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.tools,body,array<object>,false,,,,,,,DOCUMENTED|DEPRECATED,"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.tools[].type,body,string,true,,,,function,,,DOCUMENTED|DEPRECATED,"The type of the tool. Currently, only `function` is supported.",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.tools[].function,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,[schema: FunctionObject],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.model,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"The name of the model to use for generating completions (e.g. ""o3-mini"").",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source,body,object | object | object,true,,,,,,,DOCUMENTED|DEPRECATED,Determines what populates the `item` namespace in this run's data source.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.metadata,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: Metadata],https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.model,body,string | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.created_after,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.created_before,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.limit,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages.template[].role,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The role of the message (e.g. ""system"", ""assistant"", ""user"").",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.input_messages.template[].content,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The content of the message.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.tools,body,array<Tool>,false,,,,,,,DOCUMENTED|DEPRECATED,"An array of tools the model may call while generating a response. You
can specify which tool to use by setting the `tool_choice` parameter.

The two categories of tools you can provide the model are:

- **Built-in tools**: Tools that are provided by OpenAI that extend the
  model's capabilities, like [web search](https://developers.openai.com/api/docs/guides/tools-web-search)
  or [file search](https://developers.openai.com/api/docs/guides/tools-file-search). Learn more about
  [built-in tools](https://developers.openai.com/api/docs/guides/tools).
- **Function calls (custom tools)**: Functions",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.text,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,"Configuration options for a text response from the model. Can be plain
text or structured JSON data. Learn more:
- [Text inputs and outputs](https://developers.openai.com/api/docs/guides/text)
- [Structured Outputs](https://developers.openai.com/api/docs/guides/structured-outputs)",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.sampling_params.text.format,body,ResponseFormatText | TextResponseFormatJsonSchema | ResponseFormatJsonObject,false,,,,,,,DOCUMENTED|DEPRECATED,"An object specifying the format that the model must output.

Configuring `{ ""type"": ""json_schema"" }` enables Structured Outputs,
which ensures the model will match your supplied JSON schema. Learn more in the
[Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs).

The default format is `{ ""type"": ""text"" }` with no additional options.

**Not recommended for gpt-4o and newer models:**

Setting to `{ ""type"": ""json_object"" }` enables the older JSON mode, which
ensures the message the model generates is valid JSON. Using `json_schema`
is preferred for models t [schema: TextResponseFormatConfiguration]",https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.metadata,body,object | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.instructions_search,body,string | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.reasoning_effort,body,ReasoningEffort | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.users,body,array | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs,data_source.source.tools,body,array | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to retrieve runs for.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id},run_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the run to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs/{run_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation whose run you want to cancel.,https://developers.openai.com/api/reference/resources/evals
openai,POST /v1/evals/{eval_id}/runs/{run_id},run_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the run to cancel.,https://developers.openai.com/api/reference/resources/evals
openai,DELETE /v1/evals/{eval_id}/runs/{run_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to delete the run from.,https://developers.openai.com/api/reference/resources/evals
openai,DELETE /v1/evals/{eval_id}/runs/{run_id},run_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the run to delete.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to retrieve runs for.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,run_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the run to retrieve output items for.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last output item from the previous pagination request.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,limit,query,integer,false,20,,,,,,DOCUMENTED|DEPRECATED,Number of output items to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,status,query,string,false,,,,fail|pass,,,DOCUMENTED|DEPRECATED,"Filter output items by status. Use `failed` to filter by failed output
items or `pass` to filter by passed output items.",https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items,order,query,string,false,asc,,,asc|desc,,,DOCUMENTED|DEPRECATED,Sort order for output items by timestamp. Use `asc` for ascending order or `desc` for descending order. Defaults to `asc`.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items/{output_item_id},eval_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the evaluation to retrieve runs for.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items/{output_item_id},run_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the run to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/evals/{eval_id}/runs/{run_id}/output_items/{output_item_id},output_item_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the output item to retrieve.,https://developers.openai.com/api/reference/resources/evals
openai,GET /v1/files,purpose,query,string,false,,,,,,,DOCUMENTED,Only return files with the given purpose.,https://developers.openai.com/api/reference/resources/files
openai,GET /v1/files,limit,query,integer,false,10000,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 10,000, and the default is 10,000.",https://developers.openai.com/api/reference/resources/files
openai,GET /v1/files,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/files
openai,GET /v1/files,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/files
openai,POST /v1/files,file,body(multipart),string,true,,,,,,,DOCUMENTED,The File object (not file name) to be uploaded.,https://developers.openai.com/api/reference/resources/files
openai,POST /v1/files,purpose,body(multipart),string,true,,,,assistants|batch|fine-tune|vision|user_data|evals,,,DOCUMENTED|LIVE_VERIFIED,"Intended purpose. `evals` is accepted by the API (used by the Evals guide) although the OpenAPI enum only lists assistants|batch|fine-tune|vision|user_data|evals; response-side purposes also include assistants_output, batch_output, fine-tune-results. Files with purpose=batch expire after 30 days by default.",https://developers.openai.com/api/reference/resources/files/methods/create
openai,POST /v1/files,expires_after,body(multipart),object,false,,,,,,,DOCUMENTED,"The expiration policy for a file. By default, files with `purpose=batch` expire after 30 days and all other files are persisted until they are manually deleted. [schema: FileExpirationAfter]",https://developers.openai.com/api/reference/resources/files
openai,POST /v1/files,expires_after.anchor,body(multipart),string,true,,,,created_at,,,DOCUMENTED,Anchor timestamp after which the expiration policy applies. Supported anchors: `created_at`.,https://developers.openai.com/api/reference/resources/files
openai,POST /v1/files,expires_after.seconds,body(multipart),integer,true,,3600,2592000,,,,DOCUMENTED,The number of seconds after the anchor time that the file will expire. Must be between 3600 (1 hour) and 2592000 (30 days).,https://developers.openai.com/api/reference/resources/files
openai,DELETE /v1/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file to use for this request.,https://developers.openai.com/api/reference/resources/files
openai,GET /v1/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file to use for this request.,https://developers.openai.com/api/reference/resources/files
openai,GET /v1/files/{file_id}/content,file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file to use for this request.,https://developers.openai.com/api/reference/resources/files
openai,POST /v1/files,file,body(multipart),binary,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"File bytes (multipart field). Max 512 MB per file; project storage cap 2.5 TB; upload endpoint rate limit 1,000 req/min per user. Batch inputs must be .jsonl <= 200 MB; fine-tune inputs must be .jsonl; vector-store ingestion needs a supported MIME type and <= 5M tokens/file.",https://developers.openai.com/api/reference/resources/files/methods/create
openai,GET /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,fine_tuned_model_checkpoint,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuned model checkpoint to get permissions for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,project_id,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,The ID of the project to get permissions for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last permission ID from the previous pagination request.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,limit,query,integer,false,10,,,,,,DOCUMENTED|DEPRECATED,Number of permissions to retrieve.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,order,query,string,false,descending,,,ascending|descending,,,DOCUMENTED|DEPRECATED,The order in which to retrieve permissions.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,fine_tuned_model_checkpoint,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuned model checkpoint to create a permission for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions,project_ids,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The project identifiers to grant access to.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,DELETE /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions/{permission_id},fine_tuned_model_checkpoint,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuned model checkpoint to delete a permission for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,DELETE /v1/fine_tuning/checkpoints/{fine_tuned_model_checkpoint}/permissions/{permission_id},permission_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuned model checkpoint permission to delete.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,model,body,string | string,true,,,,,,,DOCUMENTED|DEPRECATED,"The name of the model to fine-tune. You can select one of the
[supported models](https://developers.openai.com/api/docs/guides/model-optimization#fine-tuning-methods).",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,training_file,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The ID of an uploaded file that contains training data.

See [upload file](https://developers.openai.com/api/reference/resources/files/methods/create) for how to upload a file.

Your dataset must be formatted as a JSONL file. Additionally, you must upload your file with the purpose `fine-tune`.

The contents of the file should differ depending on if the model uses the [chat](https://developers.openai.com/api/docs/guides/supervised-fine-tuning#formatting-your-data), [completions](https://developers.openai.com/api/docs/guides/supervised-fine-tuning#formatting-your-data) format, or if the fine-tu",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,hyperparameters,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,"The hyperparameters used for the fine-tuning job.
This value is now deprecated in favor of `method`, and should be passed in under the `method` parameter.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,hyperparameters.batch_size,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,"Number of examples in each batch. A larger batch size means that model parameters
are updated less frequently, but with lower variance.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,hyperparameters.learning_rate_multiplier,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,"Scaling factor for the learning rate. A smaller learning rate may be useful to avoid
overfitting.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,hyperparameters.n_epochs,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,"The number of epochs to train the model for. An epoch refers to one full cycle
through the training dataset.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,suffix,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"A string of up to 64 characters that will be added to your fine-tuned model name.

For example, a `suffix` of ""custom-model-name"" would produce a model name like `ft:gpt-4o-mini:openai:custom-model-name:7p4lURel`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,validation_file,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"The ID of an uploaded file that contains validation data.

If you provide this file, the data is used to generate validation
metrics periodically during fine-tuning. These metrics can be viewed in
the fine-tuning results file.
The same data should not be present in both train and validation files.

Your dataset must be formatted as a JSONL file. You must upload your file with the purpose `fine-tune`.

See the [fine-tuning guide](https://developers.openai.com/api/docs/guides/model-optimization) for more details.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations,body,array<object>,false,,,,,,,DOCUMENTED|DEPRECATED,A list of integrations to enable for your fine-tuning job.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].type,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The type of integration to enable. Currently, only ""wandb"" (Weights and Biases) is supported.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].wandb,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,"The settings for your integration with Weights and Biases. This payload specifies the project that
metrics will be sent to. Optionally, you can set an explicit display name for your run, add tags
to your run, and set a default entity (team, username, etc) to be associated with your run.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].wandb.project,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the project that the new run will be created under.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].wandb.name,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"A display name to set for the run. If not set, we will use the Job ID as the name.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].wandb.entity,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,"The entity to use for the run. This allows you to set the team or username of the WandB user that you would
like associated with the run. If not set, the default entity for the registered WandB API key is used.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,integrations[].wandb.tags,body,array<string>,false,,,,,,,DOCUMENTED|DEPRECATED,"A list of tags to be attached to the newly created run. These tags are passed through directly to WandB. Some
default tags are generated by OpenAI: ""openai/finetune"", ""openai/{base-model}"", ""openai/{ftjob-abcdef}"".",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,seed,body,integer,false,,,2147483647,,,,DOCUMENTED|DEPRECATED,"The seed controls the reproducibility of the job. Passing in the same seed and job parameters should produce the same results, but may differ in rare cases.
If a seed is not specified, one will be generated for you.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The method used for fine-tuning. [schema: FineTuneMethod],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.type,body,string,true,,,,supervised|dpo|reinforcement,,,DOCUMENTED|DEPRECATED,"The type of method. Is either `supervised`, `dpo`, or `reinforcement`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.supervised,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,Configuration for the supervised fine-tuning method. [schema: FineTuneSupervisedMethod],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.supervised.hyperparameters,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The hyperparameters used for the fine-tuning job. [schema: FineTuneSupervisedHyperparameters],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.supervised.hyperparameters.batch_size,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,"Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.supervised.hyperparameters.learning_rate_multiplier,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.supervised.hyperparameters.n_epochs,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,Configuration for the DPO fine-tuning method. [schema: FineTuneDPOMethod],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The hyperparameters used for the DPO fine-tuning job. [schema: FineTuneDPOHyperparameters],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters.beta,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,The beta value for the DPO method. A higher beta value will increase the weight of the penalty between the policy and reference model.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters.batch_size,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,"Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters.learning_rate_multiplier,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters.n_epochs,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,Configuration for the reinforcement fine-tuning method. [schema: FineTuneReinforcementMethod],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,The grader used for the fine-tuning job.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.type,body,string,true,multi,,,multi,,,DOCUMENTED|DEPRECATED,"The object type, which is always `multi`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.sampling_params.reasoning_effort,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.input,body,array<EvalItem>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.graders,body,GraderStringCheck | GraderTextSimilarity | GraderPython | GraderScoreModel | GraderLabelModel,true,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.grader.calculate_output,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,A formula to calculate the output based on grader results.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The hyperparameters used for the reinforcement fine-tuning job. [schema: FineTuneReinforcementHyperparameters],https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.batch_size,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,"Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.learning_rate_multiplier,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.n_epochs,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.reasoning_effort,body,string,false,default,,,default|low|medium|high,,,DOCUMENTED|DEPRECATED,Level of reasoning effort.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.compute_multiplier,body,string | number,false,auto,,,,,,DOCUMENTED|DEPRECATED,Multiplier on amount of compute used for exploring search space during training.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.eval_interval,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,The number of training steps between evaluation runs.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.hyperparameters.eval_samples,body,string | integer,false,auto,,,,,,DOCUMENTED|DEPRECATED,Number of evaluation samples to generate per training step.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,metadata,body,object | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: Metadata],https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last job from the previous pagination request.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs,limit,query,integer,false,20,,,,,,DOCUMENTED|DEPRECATED,Number of fine-tuning jobs to retrieve.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs,metadata,query,object,false,,,,,,,DOCUMENTED|DEPRECATED,"Optional metadata filter. To filter, use the syntax `metadata[k]=v`. Alternatively, set `metadata=null` to indicate no metadata.",https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id},fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs/{fine_tuning_job_id}/cancel,fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job to cancel.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/checkpoints,fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job to get checkpoints for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/checkpoints,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last checkpoint ID from the previous pagination request.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/checkpoints,limit,query,integer,false,10,,,,,,DOCUMENTED|DEPRECATED,Number of checkpoints to retrieve.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/events,fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job to get events for.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/events,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Identifier for the last event from the previous pagination request.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,GET /v1/fine_tuning/jobs/{fine_tuning_job_id}/events,limit,query,integer,false,20,,,,,,DOCUMENTED|DEPRECATED,Number of events to retrieve.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs/{fine_tuning_job_id}/pause,fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job to pause.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs/{fine_tuning_job_id}/resume,fine_tuning_job_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,The ID of the fine-tuning job to resume.,https://developers.openai.com/api/reference/resources/fine_tuning
openai,POST /v1/fine_tuning/jobs,method.reinforcement.response_format,body,object,false,,,,,,,DOCUMENTED|UNVERIFIED|DEPRECATED,"Structured-output format applied to samples during RFT training ({type: json_schema, json_schema: {name, strict, schema}}); used by the RFT guide, populates sample.output_json for graders. Present in guide/reference examples; not in the FineTuneReinforcementMethod spec properties.",https://developers.openai.com/api/docs/guides/reinforcement-fine-tuning
openai,POST /v1/fine_tuning/jobs,method.dpo.hyperparameters.beta,body,number | 'auto',false,auto,0,2,,,,DOCUMENTED|DEPRECATED,"DPO only. Float in [0, 2]: higher = more conservative (stick to reference behaviour), lower = follow the preferences more aggressively. Default auto.",https://developers.openai.com/api/docs/guides/direct-preference-optimization
openai,POST /v1/fine_tuning/jobs,model,body,string,true,,,,,gpt-4.1-2025-04-14|gpt-4.1-mini-2025-04-14|gpt-4.1-nano-2025-04-14|gpt-4o-2024-08-06|gpt-4o-mini-2024-07-18|o4-mini-2025-04-16|gpt-4|gpt-3.5-turbo,,DOCUMENTED|DEPRECATED,"Base model or previously fine-tuned model id. Per model pages (2026-09-18) fine-tuning is 'Supported' for: gpt-4.1, gpt-4.1-mini, gpt-4.1-nano, gpt-4o, gpt-4o-mini, o4-mini (RFT), gpt-4, gpt-3.5-turbo (legacy). Guides pin snapshots: SFT/DPO -> gpt-4.1-2025-04-14 / -mini / -nano; vision SFT -> gpt-4o-2024-08-06; RFT -> o4-mini-2025-04-16. No gpt-5.x model supports fine-tuning.",https://developers.openai.com/api/docs/guides/model-optimization#fine-tuning-methods
openai,"training_file (JSONL line, method=supervised)",messages[],file,array<message>,true,,,,,,,DOCUMENTED|DEPRECATED,"Chat-format example: system/developer, user, assistant, tool messages; assistant messages may carry tool_calls and `weight` (0|1) to exclude a turn from the loss. Optional top-level `tools[]`, `parallel_tool_calls`, `functions[]`(legacy). Min 10 lines.",https://developers.openai.com/api/docs/guides/supervised-fine-tuning#formatting-your-data
openai,"training_file (JSONL line, method=supervised)",messages[].weight,file,integer,false,,,,0|1,,,DOCUMENTED|DEPRECATED,"Assistant messages only. 0 = do not train on this message, 1 = train (default).",https://developers.openai.com/api/reference/resources/fine_tuning
openai,"training_file (JSONL line, method=dpo)",input.messages[],file,array<message>,true,,,,,,,DOCUMENTED|DEPRECATED,"Prompt messages (one-turn conversations only). Optional input.tools[], input.parallel_tool_calls.",https://developers.openai.com/api/docs/guides/direct-preference-optimization#data-format
openai,"training_file (JSONL line, method=dpo)",preferred_output[],file,array<assistant message>,true,,,,,,,DOCUMENTED|DEPRECATED,Ideal assistant response (must be the last assistant message).,https://developers.openai.com/api/docs/guides/direct-preference-optimization#data-format
openai,"training_file (JSONL line, method=dpo)",non_preferred_output[],file,array<assistant message>,true,,,,,,,DOCUMENTED|DEPRECATED,Suboptimal assistant response.,https://developers.openai.com/api/docs/guides/direct-preference-optimization#data-format
openai,"training_file (JSONL line, method=reinforcement)",messages[],file,array<message>,true,,,,,,,DOCUMENTED|DEPRECATED,Prompt messages; every other top-level key of the line is exposed to the grader as `item.<key>` (e.g. reference_answer). No assistant answer is required.,https://developers.openai.com/api/docs/guides/reinforcement-fine-tuning#prepare-your-dataset
openai,"training_file (JSONL line, vision SFT)",messages[].content[].image_url,file,object,false,,,,,,,DOCUMENTED|DEPRECATED,"Image inputs as HTTP URL or data URL; JPEG/PNG/WEBP, RGB/RGBA, <= 10 MB each, <= 10 images per example, <= 50,000 image examples per file; `detail: low` -> 85 tokens/image. No images in assistant messages.",https://developers.openai.com/api/docs/guides/vision-fine-tuning#image-data-requirements
openai,POST /v1/fine_tuning/alpha/graders/run,grader,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,The grader used for the fine-tuning job.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.type,body,string,true,multi,,,multi,,,DOCUMENTED|DEPRECATED,"The object type, which is always `multi`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.sampling_params.reasoning_effort,body,string | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.input,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.input[].role,body,string,true,,,,user|assistant|system|developer,,,DOCUMENTED|DEPRECATED,"The role of the message input. One of `user`, `assistant`, `system`, or
`developer`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.input[].content,body,EvalItemContentItem | EvalItemContentArray,true,,,,,,,DOCUMENTED|DEPRECATED,"Inputs to the model - can contain template strings. Supports text, output text, input images, and input audio, either as a single item or an array of items. [schema: EvalItemContent]",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.input[].type,body,string,false,,,,message,,,DOCUMENTED|DEPRECATED,The type of the message input. Always `message`.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders,body,object | object | object | object | object,true,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.type,body,string,true,,,,label_model,,,DOCUMENTED|DEPRECATED,"The object type, which is always `label_model`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.sampling_params.reasoning_effort,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.input,body,array<EvalItem>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels to assign to each item in the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.graders.passing_labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels that indicate a passing result. Must be a subset of labels.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,grader.calculate_output,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,A formula to calculate the output based on grader results.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,item,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,"The dataset item provided to the grader. This will be used to populate
the `item` namespace. See [the guide](https://developers.openai.com/api/docs/guides/graders) for more details.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/run,model_sample,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,"The model sample to be evaluated. This value will be used to populate
the `sample` namespace. See [the guide](https://developers.openai.com/api/docs/guides/graders) for more details.
The `output_json` variable will be populated if the model sample is a
valid JSON string.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader,body,object,true,,,,,,,DOCUMENTED|DEPRECATED,The grader used for the fine-tuning job.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.type,body,string,true,multi,,,multi,,,DOCUMENTED|DEPRECATED,"The object type, which is always `multi`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.sampling_params.reasoning_effort,body,string | null,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.input,body,array<object>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.input[].role,body,string,true,,,,user|assistant|system|developer,,,DOCUMENTED|DEPRECATED,"The role of the message input. One of `user`, `assistant`, `system`, or
`developer`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.input[].content,body,EvalItemContentItem | EvalItemContentArray,true,,,,,,,DOCUMENTED|DEPRECATED,"Inputs to the model - can contain template strings. Supports text, output text, input images, and input audio, either as a single item or an array of items. [schema: EvalItemContent]",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.input[].type,body,string,false,,,,message,,,DOCUMENTED|DEPRECATED,The type of the message input. Always `message`.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders,body,object | object | object | object | object,true,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.type,body,string,true,,,,label_model,,,DOCUMENTED|DEPRECATED,"The object type, which is always `label_model`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.name,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The name of the grader.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.input,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The input text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.reference,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The reference text. This may include template strings.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.operation,body,string,true,,,,eq|ne|like|ilike,,,DOCUMENTED|DEPRECATED,"The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.evaluation_metric,body,string,true,,,,cosine|fuzzy_match|bleu|gleu|meteor|rouge_1|rouge_2|rouge_3|rouge_4|rouge_5|rouge_l,,,DOCUMENTED|DEPRECATED,"The evaluation metric to use. One of `cosine`, `fuzzy_match`, `bleu`, 
`gleu`, `meteor`, `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, 
or `rouge_l`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.source,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The source code of the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.image_tag,body,string,false,,,,,,,DOCUMENTED|DEPRECATED,The image tag to use for the python script.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.model,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,The model to use for the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,The sampling parameters for the model.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params.seed,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params.top_p,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params.temperature,body,number | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params.max_completions_tokens,body,integer | null,false,,,,,,,DOCUMENTED|DEPRECATED,,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.sampling_params.reasoning_effort,body,object,false,,,,,,,DOCUMENTED|DEPRECATED,[schema: ReasoningEffort],https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.input,body,array<EvalItem>,true,,,,,,,DOCUMENTED|DEPRECATED,"The input messages evaluated by the grader. Supports text, output text, input image, and input audio content blocks, and may include template strings.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.range,body,array<number>,false,,,,,,,DOCUMENTED|DEPRECATED,"The range of the score. Defaults to `[0, 1]`.",https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels to assign to each item in the evaluation.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.graders.passing_labels,body,array<string>,true,,,,,,,DOCUMENTED|DEPRECATED,The labels that indicate a passing result. Must be a subset of labels.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/fine_tuning/alpha/graders/validate,grader.calculate_output,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,A formula to calculate the output based on grader results.,https://developers.openai.com/api/reference/resources/graders
openai,POST /v1/images/generations,prompt,body,string,true,,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED|LIVE_VERIFIED,"Text description. Max 32000 chars for GPT image models, 1000 for dall-e-2, 4000 for dall-e-3.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,model,body,string,false,dall-e-2 (documented; effectively required),,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED|LIVE_VERIFIED,Image model id. Docs: defaults to dall-e-2 unless a GPT-image-only parameter is present; DALL-E removed 2026-05-12 so always set it.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,n,body,integer|null,false,1,1,10,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED|LIVE_VERIFIED,Number of images. 1-10; dall-e-3 only n=1.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,quality,body,string|null,false,auto,,,auto|low|medium|high|xhigh|max|standard|hd,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED|LIVE_VERIFIED,"auto (default) | low | medium | high (GPT image) | xhigh, max (gpt-image-2.5 only) | standard, hd (dall-e-3) | standard (dall-e-2).",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,size,body,string|null,false,auto (spec) / 1024x1024 (edits spec),,,auto|1024x1024|1536x1024|1024x1536|256x256|512x512|1792x1024|1024x1792|WIDTHxHEIGHT (gpt-image-2/2.5),gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED|LIVE_VERIFIED,"WIDTHxHEIGHT. gpt-image-2 / 2.5 accept arbitrary sizes: both edges multiples of 16, aspect ratio between 1:3 and 3:1, max edge 3840, total pixels 655,360-8,294,400; above 2560x1440 is experimental. Other GPT image models: 1024x1024, 1536x1024, 1024x1536, auto. dall-e-2: 256x256, 512x512, 1024x1024. dall-e-3: 1024x1024, 1792x1024, 1024x1792.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,background,body,string|null,false,auto,,,transparent|opaque|auto,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,transparent | opaque | auto (default). GPT image models only. gpt-image-2.5 fully supports transparent; gpt-image-2 transparent is PREVIEW. With transparent use output_format png or webp.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,output_format,body,string|null,false,png,,,png|jpeg|webp,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED|LIVE_VERIFIED,png (default) | jpeg | webp. GPT image models only (DALL-E used response_format).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,output_compression,body,integer|null,false,100,0,100,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,Compression 0-100 for jpeg/webp output. GPT image models only.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,moderation,body,string|null,false,auto,,,low|auto,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,Content-moderation strictness: auto (default) or low (less restrictive). GPT image models only.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,stream,body,boolean|null,false,false,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,Stream partial images as SSE events (image_generation.partial_image / .completed). GPT image models only.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,partial_images,body,integer|null,false,0,0,3,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,Number of partial images to emit while streaming (0-3). 0 = final image only in one event. Each partial image costs +100 image output tokens.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,response_format,body,string|null,false,url,,,url|b64_json,dall-e-2|dall-e-3,,DOCUMENTED|RETIRED,url | b64_json. dall-e-2/3 only (URLs valid 60 min). GPT image models always return b64_json.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,style,body,string|null,false,vivid,,,vivid|natural,dall-e-3,,DOCUMENTED|RETIRED,vivid (default) | natural. dall-e-3 only.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/generations,user,body,string,false,,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2|dall-e-3,,DOCUMENTED,Stable end-user identifier for abuse monitoring (see safety identifiers).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,image,body,file | file[] (multipart),true,,,16,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,"Image(s) to edit, multipart field `image` or repeated `image[]`. GPT image models: up to 16 images, each png/webp/jpg < 50 MB. dall-e-2: one square PNG < 4 MB.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,images,body,"array<{file_id?, image_url?}> (JSON)",false,,,16,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,"JSON-body alternative to multipart `image`: up to 16 references, each with exactly one of file_id (Files API) or image_url (URL or base64 data URL).",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,prompt,body,string,true,,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,"Desired edit. Max 32000 chars (GPT image), 1000 (dall-e-2).",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,mask,body,"file (multipart) | {file_id?, image_url?} (JSON)",false,,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,PNG whose fully transparent pixels (alpha 0) mark the area to edit; applied to the first image; must have an alpha channel and the same size/format as the image. Masking with GPT Image is prompt-guided (not pixel-exact).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,model,body,string|null,false,gpt-image-1.5,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,Edit model. Spec default gpt-image-1.5.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,input_fidelity,body,string|null,false,low,,,high|low,gpt-image-1|gpt-image-1.5|chatgpt-image-latest|gpt-image-2.5-sunburst|gpt-image-2.5-flare,,DOCUMENTED,"high | low (default). How strongly input details (esp. faces) are preserved. Supported for gpt-image-1, gpt-image-1.5 and later; NOT gpt-image-1-mini; gpt-image-2 always uses high (parameter rejected/ignored).",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,background,body,string|null,false,auto,,,transparent|opaque|auto,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,transparent | opaque | auto. GPT image models only (gpt-image-2 transparent = preview).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,quality,body,string|null,false,auto,,,auto|low|medium|high|xhigh|max|standard,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,"auto | low | medium | high; xhigh, max for gpt-image-2.5; standard for dall-e-2.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,size,body,string|null,false,1024x1024,,,auto|1024x1024|1536x1024|1024x1536|256x256|512x512|WIDTHxHEIGHT (gpt-image-2/2.5),gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,"WIDTHxHEIGHT. gpt-image-2 / 2.5 accept arbitrary sizes: both edges multiples of 16, aspect ratio between 1:3 and 3:1, max edge 3840, total pixels 655,360-8,294,400; above 2560x1440 is experimental. Other GPT image models: 1024x1024, 1536x1024, 1024x1536, auto. dall-e-2: 256x256, 512x512, 1024x1024. dall-e-3: 1024x1024, 1792x1024, 1024x1792. Edits spec default 1024x1024.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,n,body,integer|null,false,1,1,10,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,1-10 edited images.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,moderation,body,string|null,false,auto,,,low|auto,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,low | auto. GPT image models.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,output_format,body,string|null,false,png,,,png|jpeg|webp,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,png | jpeg | webp (GPT image models).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,output_compression,body,integer|null,false,100,0,100,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,0-100 for jpeg/webp.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,response_format,body,string|null,false,url,,,url|b64_json,dall-e-2,,DOCUMENTED|RETIRED,"url | b64_json, dall-e-2 only.",https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,stream,body,boolean|null,false,false,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,Stream image_edit.partial_image / image_edit.completed events.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,partial_images,body,integer|null,false,0,0,3,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest,,DOCUMENTED,0-3 partial images while streaming (+100 output tokens each).,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/edits,user,body,string,false,,,,,gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-1.5|gpt-image-1|gpt-image-1-mini|chatgpt-image-latest|dall-e-2,,DOCUMENTED,End-user identifier.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,image,body,file (multipart),true,,,,,dall-e-2,,DOCUMENTED|RETIRED,Square PNG < 4 MB.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,model,body,string|null,false,dall-e-2,,,dall-e-2,dall-e-2,,DOCUMENTED|RETIRED,Only dall-e-2.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,n,body,integer|null,false,1,1,10,,dall-e-2,,DOCUMENTED|RETIRED,1-10.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,response_format,body,string|null,false,url,,,url|b64_json,dall-e-2,,DOCUMENTED|RETIRED,url | b64_json.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,size,body,string|null,false,1024x1024,,,256x256|512x512|1024x1024,dall-e-2,,DOCUMENTED|RETIRED,256x256 | 512x512 | 1024x1024.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/images/variations,user,body,string,false,,,,,dall-e-2,,DOCUMENTED|RETIRED,End-user identifier.,https://developers.openai.com/api/reference/resources/images
openai,POST /v1/live/sessions,session,body,object,true,,,,,gpt-live-1,,DOCUMENTED|LIVE_VERIFIED,Startup configuration for a Live media session. Follow the [Live prompting guide](https://developers.openai.com/api/docs/guides/live-prompting) when writing frontend instructions and the backend prompt under delegation.responses.instructions.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.model,body,string | enum,false,,,,gpt-live-1,gpt-live-1,,DOCUMENTED|LIVE_VERIFIED,The Live model. Required in the session configuration for every transport; do not pass it as a URL query parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.audio,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup audio output configuration. WebRTC and SIP negotiate the media format; audio.format is only accepted for primary WebSockets. Voice cannot change after startup.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.audio.output,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Settings for speech generated by the Live model. Choose the voice before starting the session.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.audio.output.voice,body,object | LiveCustomVoiceParam,false,,,,,gpt-live-1,,DOCUMENTED,"The voice used for Live speech, as a built-in voice name or a custom voice object containing its ID. Defaults to `marin` and cannot change after startup.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.audio.output.voice.id,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.type,body,string,false,client,,,client|responses,gpt-live-1,,DOCUMENTED,The delegation owner. Always `client` for tasks handled by your application.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses,body,object,false,,,,,gpt-live-1,,DOCUMENTED,"Model, prompt, and tool settings for tasks delegated by the Live session to a Responses backend.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.model,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The model used for server-owned Responses delegations.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,gpt-live-1,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,gpt-live-1,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools[].name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools[].description,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,gpt-live-1,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tool_choice.name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.store,body,boolean,false,,,,,gpt-live-1,,DOCUMENTED,Whether to store the session for later forking and recording download. Defaults to false for new sessions.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input,body,array<LiveInitialItem>,false,,,,,gpt-live-1,,DOCUMENTED,"Ordered text-only history supplied before startup. Supports developer, user, and assistant messages with one text part each; at most 128 messages and 8,192 rendered tokens in total.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].id,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].type,body,string,false,message,,,message,gpt-live-1,,DOCUMENTED,The history item type. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].status,body,LiveInitialMessageStatus | null,false,,,,incomplete|completed,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].role,body,string,false,developer,,,developer|user|assistant,gpt-live-1,,DOCUMENTED,The author of this history message. Always `developer`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].content,body,array<LiveInitialInputTextContentPartParam>,false,,,,,gpt-live-1,,DOCUMENTED,The message content. Supply exactly one text part for the initial Live conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].content[].type,body,string,false,input_text,,,input_text|text|output_text,gpt-live-1,,DOCUMENTED,The text content type. Always `input_text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.input[].content[].text,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The message text to include in the Live session’s initial conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.client,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup-only capabilities for an untrusted frontend attached to a unified WebRTC session. Trusted sideband connections are unaffected.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.client.data_channel,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Control which Live events an untrusted WebRTC frontend can send and receive over its data channel. These restrictions do not apply to trusted sideband connections.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.client.data_channel.allowed_client_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Client event types that the frontend data channel may send. Use 'all' to allow every client event; an empty array allows none. Omission preserves the existing allow-all behavior.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,session.client.data_channel.allowed_server_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Server events that may be sent to the frontend data channel. Use 'all' to allow every server event; an empty array allows none. Omission preserves the existing allow-all behavior. Responses events use an object with type 'response.event' and a response_event selector.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,transport,body,object,true,,,,,gpt-live-1,,DOCUMENTED|LIVE_VERIFIED,WebRTC transport carrying the offer SDP in a creation request or answer SDP in its response.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,transport.type,body,string,false,,,,webrtc,gpt-live-1,,DOCUMENTED|LIVE_VERIFIED,The transport used for the Live session. Always `webrtc`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions,transport.sdp,body,string,false,,,,,gpt-live-1,,DOCUMENTED|LIVE_VERIFIED,Session Description Protocol message for the WebRTC connection.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/create""}"
openai,POST /v1/live/sessions/{session_id}/fork,session,body,object,false,,,,,gpt-live-1,,DOCUMENTED,"Optional overrides for a stored Live session. Omitted settings are inherited. The model, voice, frontend instructions, and prior conversation come from the stored session. WebRTC negotiates its audio format; audio.format is only supported on WebSocket forks.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.store,body,boolean,false,,,,,gpt-live-1,,DOCUMENTED,Whether to store the forked session. Omission inherits the stored session's setting.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Update the Responses backend for an existing Live session without changing delegation ownership.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.type,body,string,false,responses,,,responses,gpt-live-1,,DOCUMENTED,The delegation owner. Always `responses` for tasks handled by the Responses API.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Updates to the Responses backend of an existing Live session. Omitted settings retain their current values.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.model,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The Responses backend model to use for subsequent delegated requests. Omit to keep the current backend model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,gpt-live-1,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,gpt-live-1,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools[].name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools[].description,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,gpt-live-1,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tool_choice.name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.client,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup-only capabilities for an untrusted frontend attached to a unified WebRTC session. Trusted sideband connections are unaffected.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.client.data_channel,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Control which Live events an untrusted WebRTC frontend can send and receive over its data channel. These restrictions do not apply to trusted sideband connections.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.client.data_channel.allowed_client_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Client event types that the frontend data channel may send. Use 'all' to allow every client event; an empty array allows none. Omission preserves the existing allow-all behavior.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session.client.data_channel.allowed_server_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Server events that may be sent to the frontend data channel. Use 'all' to allow every server event; an empty array allows none. Omission preserves the existing allow-all behavior. Responses events use an object with type 'response.event' and a response_event selector.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,transport,body,object,true,,,,,gpt-live-1,,DOCUMENTED,WebRTC transport carrying the offer SDP in a creation request or answer SDP in its response.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,transport.type,body,string,false,,,,webrtc,gpt-live-1,,DOCUMENTED,The transport used for the Live session. Always `webrtc`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,transport.sdp,body,string,false,,,,,gpt-live-1,,DOCUMENTED,Session Description Protocol message for the WebRTC connection.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/fork,session_id,path,string,true,,,,,gpt-live-1,,DOCUMENTED,Path parameter: stored source session to fork.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/fork""}"
openai,POST /v1/live/sessions/{session_id}/accept,session,body,object,true,,,,,gpt-live-1,,DOCUMENTED,Startup configuration for accepting an incoming Live SIP call. Follow the [Live prompting guide](https://developers.openai.com/api/docs/guides/live-prompting) when writing frontend and backend instructions.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.model,body,string | enum,false,,,,gpt-live-1,gpt-live-1,,DOCUMENTED,The Live model. Required in the session configuration for every transport; do not pass it as a URL query parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.audio,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup audio output configuration. WebRTC and SIP negotiate the media format; audio.format is only accepted for primary WebSockets. Voice cannot change after startup.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.audio.output,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Settings for speech generated by the Live model. Choose the voice before starting the session.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.audio.output.voice,body,object | LiveCustomVoiceParam,false,,,,,gpt-live-1,,DOCUMENTED,"The voice used for Live speech, as a built-in voice name or a custom voice object containing its ID. Defaults to `marin` and cannot change after startup.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.audio.output.voice.id,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.type,body,string,false,client,,,client|responses,gpt-live-1,,DOCUMENTED,The delegation owner. Always `client` for tasks handled by your application.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses,body,object,false,,,,,gpt-live-1,,DOCUMENTED,"Model, prompt, and tool settings for tasks delegated by the Live session to a Responses backend.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.model,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The model used for server-owned Responses delegations.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,gpt-live-1,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,gpt-live-1,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools[].name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools[].description,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,gpt-live-1,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tool_choice.name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.store,body,boolean,false,,,,,gpt-live-1,,DOCUMENTED,Whether to store the session for later forking and recording download. Defaults to false for new sessions.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input,body,array<LiveInitialItem>,false,,,,,gpt-live-1,,DOCUMENTED,"Ordered text-only history supplied before startup. Supports developer, user, and assistant messages with one text part each; at most 128 messages and 8,192 rendered tokens in total.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].id,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].type,body,string,false,message,,,message,gpt-live-1,,DOCUMENTED,The history item type. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].status,body,LiveInitialMessageStatus | null,false,,,,incomplete|completed,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].role,body,string,false,developer,,,developer|user|assistant,gpt-live-1,,DOCUMENTED,The author of this history message. Always `developer`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].content,body,array<LiveInitialInputTextContentPartParam>,false,,,,,gpt-live-1,,DOCUMENTED,The message content. Supply exactly one text part for the initial Live conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].content[].type,body,string,false,input_text,,,input_text|text|output_text,gpt-live-1,,DOCUMENTED,The text content type. Always `input_text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.input[].content[].text,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The message text to include in the Live session’s initial conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session.type,body,string,false,live,,,live,gpt-live-1,,DOCUMENTED,The session type. Always `live`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/accept,session_id,path,string,true,,,,,gpt-live-1,,DOCUMENTED,Path parameter: `data.session_id` from the `live.transport.incoming` webhook.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/accept""}"
openai,POST /v1/live/sessions/{session_id}/refer,target_uri,body,string,true,,,,,,,DOCUMENTED,"URI that should appear in the SIP Refer-To header. Supports values like
`tel:+14155550123` or `sip:agent@example.com`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/refer""}"
openai,POST /v1/live/sessions/{session_id}/refer,session_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/refer""}"
openai,POST /v1/live/sessions/{session_id}/reject,status_code,body,integer,true,,300,699,,,,DOCUMENTED,SIP rejection status sent to the caller. This field is required.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/reject""}"
openai,POST /v1/live/sessions/{session_id}/reject,session_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/reject""}"
openai,POST /v1/live/sessions/{session_id}/hangup,session_id,path,string,true,,,,,,,DOCUMENTED,Path parameter. No body.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/hangup""}"
openai,GET /v1/live/sessions/{session_id}/content,session_id,path,string,true,,,,,,,DOCUMENTED,"Path parameter. Returns audio/wav (stereo: left=input, right=output) of a finalized stored session (store=true).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/subresources/sessions/methods/content""}"
openai,WS /v1/live/sessions session.start,event_id,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,type,body,string,true,session.start,,,session.start,gpt-live-1,,DOCUMENTED,The Live client event type. Always `session.start`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session,body,object,true,,,,,gpt-live-1,,DOCUMENTED,"Initial configuration for a Live session, including its model, conversation instructions, audio, and delegated task handling.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.model,body,string | enum,false,,,,gpt-live-1,gpt-live-1,,DOCUMENTED,The Live model. Required in the session configuration for every transport; do not pass it as a URL query parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input,body,array<LiveInitialItem>,false,,,,,gpt-live-1,,DOCUMENTED,"Ordered text-only history supplied before startup. Supports developer, user, and assistant messages with one text part each; at most 128 messages and 8,192 rendered tokens in total.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].id,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].type,body,string,false,message,,,message,gpt-live-1,,DOCUMENTED,The history item type. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].status,body,LiveInitialMessageStatus | null,false,,,,incomplete|completed,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].role,body,string,false,developer,,,developer|user|assistant,gpt-live-1,,DOCUMENTED,The author of this history message. Always `developer`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].content,body,array<LiveInitialInputTextContentPartParam>,false,,,,,gpt-live-1,,DOCUMENTED,The message content. Supply exactly one text part for the initial Live conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].content[].type,body,string,false,input_text,,,input_text|text|output_text,gpt-live-1,,DOCUMENTED,The text content type. Always `input_text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.input[].content[].text,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The message text to include in the Live session’s initial conversation history.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup audio configuration. Only primary WebSockets accept audio.format; WebRTC and SIP negotiate their media format. Voice and format are immutable after startup.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.format,body,LiveSessionAudioFormatPCMParam | LiveSessionAudioFormatPCMUParam | LiveSessionAudioFormatPCMAParam,false,,,,,gpt-live-1,,DOCUMENTED,Audio encoding and sample rate for audio sent and received over a Live WebSocket connection. WebRTC and SIP negotiate their media format separately.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.format.type,body,string,false,audio/pcm,,,audio/pcm|audio/pcmu|audio/pcma,gpt-live-1,,DOCUMENTED,The audio encoding. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.format.rate,body,integer,false,,16000,24000,16000|24000,gpt-live-1,,DOCUMENTED,Audio sample rate in hertz. Live WebSocket PCM audio supports 16000 or 24000 Hz.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.output,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Settings for speech generated by the Live model. Choose the voice before starting the session.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.output.voice,body,object | LiveCustomVoiceParam,false,,,,,gpt-live-1,,DOCUMENTED,"The voice used for Live speech, as a built-in voice name or a custom voice object containing its ID. Defaults to `marin` and cannot change after startup.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.audio.output.voice.id,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.type,body,string,false,client,,,client|responses,gpt-live-1,,DOCUMENTED,The delegation owner. Always `client` for tasks handled by your application.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses,body,object,false,,,,,gpt-live-1,,DOCUMENTED,"Model, prompt, and tool settings for tasks delegated by the Live session to a Responses backend.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.model,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The model used for server-owned Responses delegations.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.instructions,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,gpt-live-1,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,gpt-live-1,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools[].name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools[].description,body,string | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,gpt-live-1,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tool_choice.name,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,gpt-live-1,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.store,body,boolean,false,,,,,gpt-live-1,,DOCUMENTED,Whether to store the session for later forking and recording download. Defaults to false for new sessions.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.client,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Startup-only capabilities for an untrusted frontend attached to a unified WebRTC session. Trusted sideband connections are unaffected.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.client.data_channel,body,object,false,,,,,gpt-live-1,,DOCUMENTED,Control which Live events an untrusted WebRTC frontend can send and receive over its data channel. These restrictions do not apply to trusted sideband connections.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.client.data_channel.allowed_client_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Client event types that the frontend data channel may send. Use 'all' to allow every client event; an empty array allows none. Omission preserves the existing allow-all behavior.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions session.start,session.client.data_channel.allowed_server_events,body,enum | array,false,,,,all,gpt-live-1,,DOCUMENTED,Server events that may be sent to the frontend data channel. Use 'all' to allow every server event; an empty array allows none. Omission preserves the existing allow-all behavior. Responses events use an object with type 'response.event' and a response_event selector.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,type,body,string,true,session.start,,,session.start,,,DOCUMENTED,The Live client event type. Always `session.start`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session,body,object,true,,,,,,,DOCUMENTED,Overrides for a stored session after connecting to the fork WebSocket. An empty object inherits the stored configuration; do not supply a new model. audio.format applies only to the new WebSocket connection. client overrides are only supported for WebRTC forks.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.audio,body,object,false,,,,,,,DOCUMENTED,Audio format for the new WebSocket connection to a forked Live session. The stored voice is preserved.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.audio.format,body,LiveSessionAudioFormatPCMParam | LiveSessionAudioFormatPCMUParam | LiveSessionAudioFormatPCMAParam,false,,,,,,,DOCUMENTED,Audio encoding and sample rate for audio sent and received over a Live WebSocket connection. WebRTC and SIP negotiate their media format separately.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.audio.format.type,body,string,false,audio/pcm,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio encoding. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.audio.format.rate,body,integer,false,,16000,24000,16000|24000,,,DOCUMENTED,Audio sample rate in hertz. Live WebSocket PCM audio supports 16000 or 24000 Hz.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation,body,object,false,,,,,,,DOCUMENTED,Update the Responses backend for an existing Live session without changing delegation ownership.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.type,body,string,false,responses,,,responses,,,DOCUMENTED,The delegation owner. Always `responses` for tasks handled by the Responses API.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses,body,object,false,,,,,,,DOCUMENTED,Updates to the Responses backend of an existing Live session. Omitted settings retain their current values.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.model,body,string,false,,,,,,,DOCUMENTED,The Responses backend model to use for subsequent delegated requests. Omit to keep the current backend model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.instructions,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools[].description,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tool_choice.name,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.store,body,boolean,false,,,,,,,DOCUMENTED,Whether to store the forked session. Omission inherits the stored session's setting.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.client,body,object,false,,,,,,,DOCUMENTED,Startup-only capabilities for an untrusted frontend attached to a unified WebRTC session. Trusted sideband connections are unaffected.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.client.data_channel,body,object,false,,,,,,,DOCUMENTED,Control which Live events an untrusted WebRTC frontend can send and receive over its data channel. These restrictions do not apply to trusted sideband connections.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.client.data_channel.allowed_client_events,body,enum | array,false,,,,all,,,DOCUMENTED,Client event types that the frontend data channel may send. Use 'all' to allow every client event; an empty array allows none. Omission preserves the existing allow-all behavior.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions/{session_id}/fork session.start,session.client.data_channel.allowed_server_events,body,enum | array,false,,,,all,,,DOCUMENTED,Server events that may be sent to the frontend data channel. Use 'all' to allow every server event; an empty array allows none. Omission preserves the existing allow-all behavior. Responses events use an object with type 'response.event' and a response_event selector.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/fork-websocket#session.start""}"
openai,WS /v1/live/sessions session.update,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,type,body,string,true,session.update,,,session.update,,,DOCUMENTED,The Live client event type. Always `session.update`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session,body,object,true,,,,,,,DOCUMENTED,Changes to an active Live session. Only delegation backend settings can be updated after startup.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.type,body,string,false,client,,,client|responses,,,DOCUMENTED,The delegation owner. Always `client` for tasks handled by your application.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses,body,object,false,,,,,,,DOCUMENTED,Updates to the Responses backend of an existing Live session. Omitted settings retain their current values.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.model,body,string,false,,,,,,,DOCUMENTED,The Responses backend model to use for subsequent delegated requests. Omit to keep the current backend model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.instructions,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.max_output_tokens,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.service_tier,body,LiveResponsesServiceTier | null,false,,,,auto|default|fast_tier_temp_pilot|flex|priority|ultrafast,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.reasoning,body,LiveDelegationReasoningInputParam | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.reasoning.effort,body,LiveReasoningEffort | null,false,,,,none|minimal|low|medium|high|xhigh,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.reasoning.summary,body,LiveReasoningSummary | null,false,,,,concise|detailed|auto,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.text,body,LiveDelegationTextInputParam | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.text.verbosity,body,LiveTextVerbosity | null,false,,,,low|medium|high,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools,body,array<LiveFunctionToolInputParam | LiveWebSearchToolInputParam>,false,,,,,,,DOCUMENTED,Tools available to the Responses backend while it handles tasks delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools[].type,body,string,false,function,,,function|web_search,,,DOCUMENTED,The tool type. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name the delegated Responses model uses when calling this function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools[].description,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools[].parameters,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tools[].strict,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tool_choice,body,LiveToolChoiceEnum | LiveFunctionToolChoiceParam | LiveMCPToolChoiceParam,false,,,,auto|none|required,,,DOCUMENTED,Controls which tool the Responses backend uses when handling a task delegated by the Live model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tool_choice.type,body,string,false,function,,,function|mcp,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tool_choice.name,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.update,session.delegation.responses.parallel_tool_calls,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.update""}"
openai,WS /v1/live/sessions session.input_audio.append,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.append""}"
openai,WS /v1/live/sessions session.input_audio.append,type,body,string,true,session.input_audio.append,,,session.input_audio.append,,,DOCUMENTED,The Live client event type. Always `session.input_audio.append`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.append""}"
openai,WS /v1/live/sessions session.input_audio.append,audio,body,string,true,,,,,,,DOCUMENTED,"Base64-encoded raw audio in the startup-selected format, without a WAV or other container header. Primary WebSocket only; media transports use their audio track. Audio appends have no acknowledgment. Reflected sideband server events reuse this event type and audio key, with no timestamps or event_id; their audio is always mono PCM16LE at 24 kHz.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.append""}"
openai,WS /v1/live/sessions session.input_audio.mute,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.mute""}"
openai,WS /v1/live/sessions session.input_audio.mute,type,body,string,true,session.input_audio.mute,,,session.input_audio.mute,,,DOCUMENTED,The Live client event type. Always `session.input_audio.mute`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.mute""}"
openai,WS /v1/live/sessions session.input_audio.unmute,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.unmute""}"
openai,WS /v1/live/sessions session.input_audio.unmute,type,body,string,true,session.input_audio.unmute,,,session.input_audio.unmute,,,DOCUMENTED,The Live client event type. Always `session.input_audio.unmute`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.input_audio.unmute""}"
openai,WS /v1/live/sessions session.instructions.append,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.instructions.append""}"
openai,WS /v1/live/sessions session.instructions.append,type,body,string,true,session.instructions.append,,,session.instructions.append,,,DOCUMENTED,The Live client event type. Always `session.instructions.append`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.instructions.append""}"
openai,WS /v1/live/sessions session.instructions.append,delegation_id,body,string | null,true,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.instructions.append""}"
openai,WS /v1/live/sessions session.instructions.append,content,body,string,true,,,,,,,DOCUMENTED,"Instruction text to append, limited to 500 tokens. This is a plain string, not an array of content parts.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.instructions.append""}"
openai,WS /v1/live/sessions session.thinking.append,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.thinking.append""}"
openai,WS /v1/live/sessions session.thinking.append,type,body,string,true,session.thinking.append,,,session.thinking.append,,,DOCUMENTED,The Live client event type. Always `session.thinking.append`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.thinking.append""}"
openai,WS /v1/live/sessions session.thinking.append,delegation_id,body,string | null,true,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.thinking.append""}"
openai,WS /v1/live/sessions session.thinking.append,content,body,string,true,,,,,,,DOCUMENTED,"Silent reasoning or progress context, limited to 500 tokens. It does not directly request speech, but can influence later speech and is not a secrecy boundary.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.thinking.append""}"
openai,WS /v1/live/sessions session.commentary.append,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.commentary.append""}"
openai,WS /v1/live/sessions session.commentary.append,type,body,string,true,session.commentary.append,,,session.commentary.append,,,DOCUMENTED,The Live client event type. Always `session.commentary.append`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.commentary.append""}"
openai,WS /v1/live/sessions session.commentary.append,delegation_id,body,string | null,true,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.commentary.append""}"
openai,WS /v1/live/sessions session.commentary.append,content,body,string,true,,,,,,,DOCUMENTED,"Speakable context for the Live model, limited to 500 tokens. Use this for a result the model should communicate; use session.thinking.append for silent context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.commentary.append""}"
openai,WS /v1/live/sessions response.item.create,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,type,body,string,true,response.item.create,,,response.item.create,,,DOCUMENTED,The Live client event type. Always `response.item.create`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item,body,EasyInputMessage | Item | CompactionTriggerItemParam | ItemReferenceParam | ProgramItemParam | ProgramOutputItemParam,true,,,,,,,DOCUMENTED,"An input item to append to the Responses backend conversation, such as a user message or a function tool result.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.role,body,string,false,,,,user|assistant|system|developer,,,DOCUMENTED,"The role of the message input. One of `user`, `assistant`, `system`, or
`developer`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content,body,string | InputMessageContentList,false,,,,,,,DOCUMENTED,"Text, image, or audio input to the model, used to generate a response.
Can also contain previous assistant responses.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.phase,body,MessagePhase | null,false,,,,commentary|final_answer,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.type,body,string,false,,,,message|file_search_call|computer_call|computer_call_output|web_search_call|function_call|function_call_output|tool_search_call|tool_search_output|additional_tools|configuration_update|reasoning|compaction|image_generation_call|code_interpreter_call|local_shell_call|local_shell_call_output|shell_call|shell_call_output|apply_patch_call|apply_patch_call_output|mcp_list_tools|mcp_approval_request|mcp_approval_response|mcp_call|custom_tool_call_output|custom_tool_call|compaction_trigger|item_reference|program|program_output,,,DOCUMENTED,The type of the message input. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.status,body,string,false,,,,in_progress|completed|incomplete|searching|failed|generating|interpreting|calling,,,DOCUMENTED,"The status of item. One of `in_progress`, `completed`, or
`incomplete`. Populated when items are returned via API.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].type,body,string,false,input_text,,,input_text|input_image|input_file|output_text|refusal|reasoning_text,,,DOCUMENTED,The type of the input item. Always `input_text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].text,body,string,false,,,,,,,DOCUMENTED,The text input to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].prompt_cache_breakpoint,body,object,false,,,,,,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].prompt_cache_breakpoint.mode,body,string,false,explicit,,,explicit,,,DOCUMENTED,The breakpoint mode. Always `explicit`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].image_url,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].file_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].detail,body,string,false,,,,low|high|auto|original,,,DOCUMENTED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].filename,body,string,false,,,,,,,DOCUMENTED,The name of the file to be sent to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].file_data,body,string,false,,,,,,,DOCUMENTED,The content of the file to be sent to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].file_url,body,string,false,,,,,,,DOCUMENTED,The URL of the file to be sent to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.id,body,string,false,,,,,,,DOCUMENTED,The unique ID of the output message.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations,body,array<Annotation>,false,,,,,,,DOCUMENTED,The annotations of the text output.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].type,body,string,false,file_citation,,,file_citation|url_citation|container_file_citation|file_path,,,DOCUMENTED,The type of the file citation. Always `file_citation`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].file_id,body,string,false,,,,,,,DOCUMENTED,The ID of the file.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].index,body,integer,false,,,,,,,DOCUMENTED,The index of the file in the list of files.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].filename,body,string,false,,,,,,,DOCUMENTED,The filename of the file cited.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].url,body,string,false,,,,,,,DOCUMENTED,The URL of the web resource.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].start_index,body,integer,false,,,,,,,DOCUMENTED,The index of the first character of the URL citation in the message.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].end_index,body,integer,false,,,,,,,DOCUMENTED,The index of the last character of the URL citation in the message.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].title,body,string,false,,,,,,,DOCUMENTED,The title of the web resource.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].annotations[].container_id,body,string,false,,,,,,,DOCUMENTED,The ID of the container file.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs,body,array<LogProb>,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].token,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].logprob,body,number,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].bytes,body,array<integer>,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].top_logprobs,body,array<TopLogProb>,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].top_logprobs[].token,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].top_logprobs[].logprob,body,number,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].logprobs[].top_logprobs[].bytes,body,array<integer>,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.content[].refusal,body,string,false,,,,,,,DOCUMENTED,The refusal explanation from the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.queries,body,array<string>,false,,,,,,,DOCUMENTED,The queries used to search for files.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.results,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.call_id,body,string,false,,,,,,,DOCUMENTED,An identifier used when responding to the tool call with output.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action,body,ClickParam | DoubleClickAction | DragParam | KeyPressAction | MoveParam | ScreenshotParam | ScrollParam | TypeParam | WaitParam,false,,,,,,,DOCUMENTED,"An object describing the specific action taken in this web search call.
Includes details on how the model used the web (search, open_page, find_in_page).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.type,body,string,false,click,,,click|double_click|drag|keypress|move|screenshot|scroll|type|wait|search|open_page|find_in_page|exec,,,DOCUMENTED,"Specifies the event type. For a click action, this property is always `click`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.button,body,string,false,,,,left|right|wheel|back|forward,,,DOCUMENTED,"Indicates which mouse button was pressed during the click. One of `left`, `right`, `wheel`, `back`, or `forward`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.x,body,integer,false,,,,,,,DOCUMENTED,The x-coordinate where the click occurred.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.y,body,integer,false,,,,,,,DOCUMENTED,The y-coordinate where the click occurred.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.keys,body,array | null,false,,,,,,,DOCUMENTED,"The combination of keys the model is requesting to be pressed. This is an array of strings, each representing a key.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.path,body,array<CoordParam>,false,,,,,,,DOCUMENTED,"An array of coordinates representing the path of the drag action. Coordinates will appear as an array of objects, eg
```
[
  { x: 100, y: 200 },
  { x: 200, y: 300 }
]
```","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.path[].x,body,integer,false,,,,,,,DOCUMENTED,The x-coordinate.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.path[].y,body,integer,false,,,,,,,DOCUMENTED,The y-coordinate.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.scroll_x,body,integer,false,,,,,,,DOCUMENTED,The horizontal scroll distance.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.scroll_y,body,integer,false,,,,,,,DOCUMENTED,The vertical scroll distance.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.text,body,string,false,,,,,,,DOCUMENTED,The text to type.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions,body,array<ComputerAction>,false,,,,,,,DOCUMENTED,"Flattened batched actions for `computer_use`. Each action includes an
`type` discriminator and action-specific fields.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].type,body,string,false,click,,,click|double_click|drag|keypress|move|screenshot|scroll|type|wait,,,DOCUMENTED,"Specifies the event type. For a click action, this property is always `click`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].button,body,string,false,,,,left|right|wheel|back|forward,,,DOCUMENTED,"Indicates which mouse button was pressed during the click. One of `left`, `right`, `wheel`, `back`, or `forward`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].x,body,integer,false,,,,,,,DOCUMENTED,The x-coordinate where the click occurred.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].y,body,integer,false,,,,,,,DOCUMENTED,The y-coordinate where the click occurred.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].keys,body,array | null,false,,,,,,,DOCUMENTED,"The combination of keys the model is requesting to be pressed. This is an array of strings, each representing a key.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].path,body,array<CoordParam>,false,,,,,,,DOCUMENTED,"An array of coordinates representing the path of the drag action. Coordinates will appear as an array of objects, eg
```
[
  { x: 100, y: 200 },
  { x: 200, y: 300 }
]
```","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].path[].x,body,integer,false,,,,,,,DOCUMENTED,The x-coordinate.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].path[].y,body,integer,false,,,,,,,DOCUMENTED,The y-coordinate.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].scroll_x,body,integer,false,,,,,,,DOCUMENTED,The horizontal scroll distance.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].scroll_y,body,integer,false,,,,,,,DOCUMENTED,The vertical scroll distance.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.actions[].text,body,string,false,,,,,,,DOCUMENTED,The text to type.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.pending_safety_checks,body,array<ComputerCallSafetyCheckParam>,false,,,,,,,DOCUMENTED,The pending safety checks for the computer call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.pending_safety_checks[].id,body,string,false,,,,,,,DOCUMENTED,The ID of the pending safety check.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.pending_safety_checks[].code,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.pending_safety_checks[].message,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output,body,object,false,,,,,,,DOCUMENTED,A computer screenshot image used with the computer use tool.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output.type,body,string,false,computer_screenshot,,,computer_screenshot,,,DOCUMENTED,"Specifies the event type. For a computer screenshot, this property is 
always set to `computer_screenshot`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output.image_url,body,string,false,,,,,,,DOCUMENTED,The URL of the screenshot image.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output.file_id,body,string,false,,,,,,,DOCUMENTED,The identifier of an uploaded file that contains the screenshot.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.acknowledged_safety_checks,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.query,body,string,false,,,,,,,DOCUMENTED,The search query.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.queries,body,array<string>,false,,,,,,,DOCUMENTED,The search queries.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.sources,body,array<object>,false,,,,,,,DOCUMENTED,The sources used in the search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.sources[].type,body,string,false,,,,url,,,DOCUMENTED,The type of source. Always `url`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.sources[].url,body,string,false,,,,,,,DOCUMENTED,The URL of the source.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.url,body,string | null,false,,,,,,,DOCUMENTED,The URL opened by the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.pattern,body,string,false,,,,,,,DOCUMENTED,The pattern or text to search for within the page.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.caller,body,ToolCallCaller | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.caller.type,body,string,false,direct,,,direct|program,,,DOCUMENTED,The caller type. Always `direct`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.caller.caller_id,body,string,false,,,,,,,DOCUMENTED,The call ID of the program item that produced this tool call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.namespace,body,string,false,,,,,,,DOCUMENTED,The namespace of the function to run.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.name,body,string,false,,,,,,,DOCUMENTED,The name of the function to run.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.arguments,body,string,false,,,,,,,DOCUMENTED,A JSON string of the arguments to pass to the function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.async,body,boolean,false,,,,,,,DOCUMENTED,Whether the function tool call runs asynchronously.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.execution,body,string,false,,,,server|client,,,DOCUMENTED,Whether tool search was executed by the server or by the client.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools,body,array<Tool>,false,,,,,,,DOCUMENTED,The loaded tool definitions returned by the tool search output.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].type,body,string,false,function,,,function|file_search|computer|computer_use_preview|web_search|web_search_2025_08_26|mcp|code_interpreter|programmatic_tool_calling|image_generation|local_shell|shell|custom|namespace|tool_search|web_search_preview|web_search_preview_2025_03_11|apply_patch,,,DOCUMENTED,The type of the function tool. Always `function`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].async,body,boolean,false,,,,,,,DOCUMENTED,Whether the tool response can be returned asynchronously versus immediately returned on next response creation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].description,body,string | null,false,,,,,,,DOCUMENTED,"Optional description of the custom tool, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].parameters,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].output_schema,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].strict,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this function is deferred and loaded via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].vector_store_ids,body,array<string>,false,,,,,,,DOCUMENTED,The IDs of the vector stores to search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].max_num_results,body,integer,false,,,,,,,DOCUMENTED,The maximum number of results to return. This number should be between 1 and 50 inclusive.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options,body,object,false,,,,,,,DOCUMENTED,Ranking options for search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options.ranker,body,string,false,,,,auto|default-2024-11-15,,,DOCUMENTED,The ranker to use for the file search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options.score_threshold,body,number,false,,,,,,,DOCUMENTED,"The score threshold for the file search, a number between 0 and 1. Numbers closer to 1 will attempt to return only the most relevant results, but may return fewer results.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options.hybrid_search,body,object,false,,,,,,,DOCUMENTED,Weights that control how reciprocal rank fusion balances semantic embedding matches versus sparse keyword matches when hybrid search is enabled.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options.hybrid_search.embedding_weight,body,number,false,,,,,,,DOCUMENTED,The weight of the embedding in the reciprocal ranking fusion.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].ranking_options.hybrid_search.text_weight,body,number,false,,,,,,,DOCUMENTED,The weight of the text in the reciprocal ranking fusion.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters,body,Filters | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.type,body,string,false,eq,,,eq|ne|gt|gte|lt|lte|in|nin|and|or,,,DOCUMENTED,"Specifies the comparison operator: `eq`, `ne`, `gt`, `gte`, `lt`, `lte`, `in`, `nin`.
- `eq`: equals
- `ne`: not equal
- `gt`: greater than
- `gte`: greater than or equal
- `lt`: less than
- `lte`: less than or equal
- `in`: in
- `nin`: not in","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.key,body,string,false,,,,,,,DOCUMENTED,The key to compare against the value.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.value,body,string | number | boolean | array,false,,,,,,,DOCUMENTED,"The value to compare against the attribute key; supports string, number, or boolean types.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.filters,body,array<ComparisonFilter | object>,false,,,,,,,DOCUMENTED,Array of filters to combine. Items can be `ComparisonFilter` or `CompoundFilter`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.filters[].type,body,string,false,eq,,,eq|ne|gt|gte|lt|lte|in|nin,,,DOCUMENTED,"Specifies the comparison operator: `eq`, `ne`, `gt`, `gte`, `lt`, `lte`, `in`, `nin`.
- `eq`: equals
- `ne`: not equal
- `gt`: greater than
- `gte`: greater than or equal
- `lt`: less than
- `lte`: less than or equal
- `in`: in
- `nin`: not in","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.filters[].key,body,string,false,,,,,,,DOCUMENTED,The key to compare against the value.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.filters[].value,body,string | number | boolean | array,false,,,,,,,DOCUMENTED,"The value to compare against the attribute key; supports string, number, or boolean types.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment,body,string,false,,,,windows|mac|linux|ubuntu|browser,,,DOCUMENTED,The type of computer environment to control.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].display_width,body,integer,false,,,,,,,DOCUMENTED,The width of the computer display.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].display_height,body,integer,false,,,,,,,DOCUMENTED,The height of the computer display.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].external_web_access,body,boolean,false,true,,,,,,DOCUMENTED,"Allow live internet access for web search. Defaults to true when omitted. When false, the web search tool runs in offline/cache-only mode and will not fetch new external content.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].filters.allowed_domains,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location.type,body,string,false,approximate,,,approximate,,,DOCUMENTED,The type of location approximation. Always `approximate`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location.country,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location.region,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location.city,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].user_location.timezone,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].search_context_size,body,string,false,medium,,,low|medium|high,,,DOCUMENTED,"High level guidance for the amount of context window space to use for the search. One of `low`, `medium`, or `high`. `medium` is the default.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].server_label,body,string,false,,,,,,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].server_url,body,string,false,,,,,,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tunnel_id,body,string,false,,,,,,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].authorization,body,string,false,,,,,,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].server_description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].headers,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].allowed_tools,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].allowed_tools.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].allowed_tools.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.always,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.always.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.always.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.never,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.never.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].require_approval.never.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container,body,string | AutoCodeInterpreterToolParam,false,,,,,,,DOCUMENTED,"The code interpreter container. Can be a container ID or an object that
specifies uploaded file IDs to make available to your code, along with an
optional `memory_limit` setting.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.type,body,string,false,auto,,,auto,,,DOCUMENTED,Always `auto`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.file_ids,body,array<string>,false,,,,,,,DOCUMENTED,An optional list of uploaded files to make available to your code.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.memory_limit,body,ContainerMemoryLimit | null,false,,,,1g|4g|16g|64g,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy,body,ContainerNetworkPolicyDisabledParam | ContainerNetworkPolicyAllowlistParam,false,,,,,,,DOCUMENTED,Network access policy for the container.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.type,body,string,false,disabled,,,disabled|allowlist,,,DOCUMENTED,Disable outbound network access. Always `disabled`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.allowed_domains,body,array<string>,false,,,,,,,DOCUMENTED,A list of allowed domains when type is `allowlist`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.domain_secrets,body,array<ContainerNetworkPolicyDomainSecretParam>,false,,,,,,,DOCUMENTED,Optional domain-scoped secrets for allowlisted domains.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.domain_secrets[].domain,body,string,false,,,,,,,DOCUMENTED,The domain associated with the secret.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.domain_secrets[].name,body,string,false,,,,,,,DOCUMENTED,The name of the secret to inject for the domain.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].container.network_policy.domain_secrets[].value,body,string,false,,,,,,,DOCUMENTED,The secret value to inject for the domain.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].model,body,string | enum,false,,,,gpt-image-1|gpt-image-1-mini|gpt-image-1.5|gpt-image-2|gpt-image-2-2026-04-21|gpt-image-2.5-sunburst|gpt-image-2.5-sunburst-2026-09-08|gpt-image-2.5-flare|gpt-image-2.5-flare-2026-09-08,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].quality,body,string,false,auto,,,low|medium|high|xhigh|max|auto,,,DOCUMENTED,"The quality of the generated image. The GPT image models support `low`,
`medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,
including their `2026-09-08` snapshots, also support `xhigh` and `max`.
Default: `auto`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].size,body,string | enum,false,auto,,,1024x1024|1024x1536|1536x1024|auto,,,DOCUMENTED,"The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].output_format,body,string,false,png,,,png|webp|jpeg,,,DOCUMENTED,"The output format of the generated image. One of `png`, `webp`, or
`jpeg`. Default: `png`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].output_compression,body,integer,false,100,0,100,,,,DOCUMENTED,Compression level for the output image. Default: 100.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].moderation,body,string,false,auto,,,auto|low,,,DOCUMENTED,Moderation level for the generated image. Default: `auto`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].background,body,string,false,auto,,,transparent|opaque|auto,,,DOCUMENTED,"Set the background of the generated image. One of `transparent`, `opaque`,
or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including
their `2026-09-08` snapshots, support `opaque` and `transparent`
backgrounds. Transparent backgrounds are available for supported GPT Image
models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in
preview. When using `transparent`, set the output format to `png` or `webp`.
Default: `auto`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].input_fidelity,body,InputFidelity | null,false,,,,high|low,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].input_image_mask,body,object,false,,,,,,,DOCUMENTED,"Optional mask for inpainting. Contains `image_url`
(string, optional) and `file_id` (string, optional).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].input_image_mask.image_url,body,string,false,,,,,,,DOCUMENTED,Base64-encoded mask image.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].input_image_mask.file_id,body,string,false,,,,,,,DOCUMENTED,File ID for the mask image.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].partial_images,body,integer,false,0,0,3,,,,DOCUMENTED,"Number of partial images to generate in streaming mode, from 0 (default value) to 3.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].action,body,string,false,,,,generate|edit|auto,,,DOCUMENTED,Whether to generate a new image or edit an existing image. Default: `auto`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.type,body,string,false,container_auto,,,container_auto|local|container_reference,,,DOCUMENTED,Automatically creates a container for this request,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.file_ids,body,array<string>,false,,,,,,,DOCUMENTED,An optional list of uploaded files to make available to your code.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.memory_limit,body,ContainerMemoryLimit | null,false,,,,1g|4g|16g|64g,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy,body,ContainerNetworkPolicyDisabledParam | ContainerNetworkPolicyAllowlistParam,false,,,,,,,DOCUMENTED,Network access policy for the container.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.type,body,string,false,disabled,,,disabled|allowlist,,,DOCUMENTED,Disable outbound network access. Always `disabled`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.allowed_domains,body,array<string>,false,,,,,,,DOCUMENTED,A list of allowed domains when type is `allowlist`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.domain_secrets,body,array<ContainerNetworkPolicyDomainSecretParam>,false,,,,,,,DOCUMENTED,Optional domain-scoped secrets for allowlisted domains.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.domain_secrets[].domain,body,string,false,,,,,,,DOCUMENTED,The domain associated with the secret.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.domain_secrets[].name,body,string,false,,,,,,,DOCUMENTED,The name of the secret to inject for the domain.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.network_policy.domain_secrets[].value,body,string,false,,,,,,,DOCUMENTED,The secret value to inject for the domain.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills,body,array<SkillReferenceParam | InlineSkillParam>,false,,,,,,,DOCUMENTED,An optional list of skills referenced by id or inline data.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].type,body,string,false,skill_reference,,,skill_reference|inline,,,DOCUMENTED,References a skill created with the /v1/skills endpoint.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].skill_id,body,string,false,,,,,,,DOCUMENTED,The ID of the referenced skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].version,body,string,false,,,,,,,DOCUMENTED,Optional skill version. Use a positive integer or 'latest'. Omit for default.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].name,body,string,false,,,,,,,DOCUMENTED,The name of the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].description,body,string,false,,,,,,,DOCUMENTED,The description of the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].source,body,object,false,,,,,,,DOCUMENTED,Inline skill payload,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].source.type,body,string,false,base64,,,base64,,,DOCUMENTED,The type of the inline skill source. Must be `base64`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].source.media_type,body,string,false,application/zip,,,application/zip,,,DOCUMENTED,The media type of the inline skill payload. Must be `application/zip`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].source.data,body,string,false,,,,,,,DOCUMENTED,Base64-encoded skill zip bundle.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.skills[].path,body,string,false,,,,,,,DOCUMENTED,The path to the directory containing the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].environment.container_id,body,string,false,,,,,,,DOCUMENTED,The ID of the referenced container.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].format,body,CustomTextFormatParam | CustomGrammarFormatParam,false,,,,,,,DOCUMENTED,The input format for the custom tool. Default is unconstrained text.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].format.type,body,string,false,text,,,text|grammar,,,DOCUMENTED,Unconstrained text format. Always `text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].format.syntax,body,string,false,,,,lark|regex,,,DOCUMENTED,The syntax of the grammar definition. One of `lark` or `regex`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].format.definition,body,string,false,,,,,,,DOCUMENTED,The grammar definition.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools,body,array<FunctionToolParam | CustomToolParam>,false,,,,,,,DOCUMENTED,The function/custom tools available inside this namespace.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].name,body,string,false,,,,,,,DOCUMENTED,"The name of the custom tool, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].description,body,string | null,false,,,,,,,DOCUMENTED,"Optional description of the custom tool, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].parameters,body,EmptyModelParam | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].strict,body,boolean | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].type,body,string,false,function,,,function|custom,,,DOCUMENTED,The type of the custom tool. Always `custom`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].async,body,boolean,false,,,,,,,DOCUMENTED,Whether the tool response can be returned asynchronously versus immediately returned on next response creation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].output_schema,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this function should be deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].format,body,CustomTextFormatParam | CustomGrammarFormatParam,false,,,,,,,DOCUMENTED,The input format for the custom tool. Default is unconstrained text.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].format.type,body,string,false,text,,,text|grammar,,,DOCUMENTED,Unconstrained text format. Always `text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].format.syntax,body,string,false,,,,lark|regex,,,DOCUMENTED,The syntax of the grammar definition. One of `lark` or `regex`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].tools[].format.definition,body,string,false,,,,,,,DOCUMENTED,The grammar definition.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].execution,body,string,false,,,,server|client,,,DOCUMENTED,Whether tool search is executed by the server or by the client.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].search_content_types,body,array<SearchContentType>,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.reasoning,body,object,false,,,,,,,DOCUMENTED,Updates to reasoning configuration. Only effort is supported.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.reasoning.effort,body,enum | null,false,,,,none|minimal|low|medium|high|xhigh|max,,,DOCUMENTED,"The reasoning effort to use for subsequent responses until another
configuration update replaces it.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.encrypted_content,body,string | null,false,,,,,,,DOCUMENTED,The encrypted content of the compaction summary.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.summary,body,array<SummaryTextContent>,false,,,,,,,DOCUMENTED,Reasoning summary content.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.summary[].type,body,string,false,summary_text,,,summary_text,,,DOCUMENTED,The type of the object. Always `summary_text`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.summary[].text,body,string,false,,,,,,,DOCUMENTED,A summary of the reasoning output from the model so far.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.result,body,string | null,false,,,,,,,DOCUMENTED,The result produced by the program item.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.size,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.quality,body,enum | null,false,,,,low|medium|high|xhigh|max|auto,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.background,body,ImageBackground | null,false,,,,transparent|opaque|auto,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output_format,body,ImageOutputFormat | null,false,,,,png|webp|jpeg,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.revised_prompt,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.container_id,body,string,false,,,,,,,DOCUMENTED,The ID of the container used to run the code.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.code,body,string | null,false,,,,,,,DOCUMENTED,The JavaScript source executed by programmatic tool calling.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.outputs,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.command,body,array<string>,false,,,,,,,DOCUMENTED,The command to run.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.timeout_ms,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.working_directory,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.env,body,object,false,,,,,,,DOCUMENTED,Environment variables to set for the command.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.user,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.commands,body,array<string>,false,,,,,,,DOCUMENTED,Ordered shell commands for the execution environment to run.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.action.max_output_length,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.type,body,string,false,local,,,local|container_reference,,,DOCUMENTED,Use a local computer environment.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.skills,body,array<LocalSkillParam>,false,,,,,,,DOCUMENTED,An optional list of skills.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.skills[].name,body,string,false,,,,,,,DOCUMENTED,The name of the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.skills[].description,body,string,false,,,,,,,DOCUMENTED,The description of the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.skills[].path,body,string,false,,,,,,,DOCUMENTED,The path to the directory containing the skill.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.environment.container_id,body,string,false,,,,,,,DOCUMENTED,The ID of the referenced container.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output[].stdout,body,string,false,,,,,,,DOCUMENTED,Captured stdout output for the shell call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output[].stderr,body,string,false,,,,,,,DOCUMENTED,Captured stderr output for the shell call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output[].outcome,body,FunctionShellCallOutputTimeoutOutcomeParam | FunctionShellCallOutputExitOutcomeParam,false,,,,,,,DOCUMENTED,The exit or timeout outcome associated with this shell call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output[].outcome.type,body,string,false,timeout,,,timeout|exit,,,DOCUMENTED,The outcome type. Always `timeout`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.output[].outcome.exit_code,body,integer,false,,,,,,,DOCUMENTED,The exit code returned by the shell process.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.max_output_length,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.operation,body,ApplyPatchCreateFileOperationParam | ApplyPatchDeleteFileOperationParam | ApplyPatchUpdateFileOperationParam,false,,,,,,,DOCUMENTED,"One of the create_file, delete_file, or update_file operations supplied to the apply_patch tool.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.operation.type,body,string,false,create_file,,,create_file|delete_file|update_file,,,DOCUMENTED,The operation type. Always `create_file`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.operation.path,body,string,false,,,,,,,DOCUMENTED,Path of the file to create relative to the workspace root.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.operation.diff,body,string,false,,,,,,,DOCUMENTED,Unified diff content to apply when creating the file.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].input_schema,body,object,false,,,,,,,DOCUMENTED,The JSON schema describing the tool's input.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.tools[].annotations,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.error,body,string | null,false,,,,,,,DOCUMENTED,"The error from the tool call, if any.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.approval_request_id,body,string,false,,,,,,,DOCUMENTED,The ID of the approval request being answered.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.approve,body,boolean,false,,,,,,,DOCUMENTED,Whether the request was approved.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.reason,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.error.type,body,string,false,mcp_protocol_error,,,mcp_protocol_error|mcp_tool_execution_error|http_error,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.error.code,body,integer,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.error.message,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.error.content,body,any,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.input,body,string,false,,,,,,,DOCUMENTED,The input for the custom tool call generated by the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.item.create,item.fingerprint,body,string,false,,,,,,,DOCUMENTED,Opaque program replay fingerprint that must be round-tripped.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.item.create""}"
openai,WS /v1/live/sessions response.create,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.create""}"
openai,WS /v1/live/sessions response.create,type,body,string,true,response.create,,,response.create,,,DOCUMENTED,The Live client event type. Always `response.create`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#response.create""}"
openai,WS /v1/live/sessions session.close,event_id,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.close""}"
openai,WS /v1/live/sessions session.close,type,body,string,true,session.close,,,session.close,,,DOCUMENTED,The Live client event type. Always `session.close`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/live/primary-websocket#session.close""}"
openai,POST /v1/moderations,input,body,"string | string[] | array<{type:'text', text} | {type:'image_url', image_url:{url}}>",true,,,,,omni-moderation-latest|omni-moderation-2024-09-26,,DOCUMENTED|LIVE_VERIFIED,"Content to classify: one string, many strings, or a multimodal array. Images: URL or base64 data URL, <= 20 MB; only omni-moderation models accept images.",https://developers.openai.com/api/reference/resources/moderations
openai,POST /v1/moderations,input[].type,body,string,false,,,,text|image_url,omni-moderation-latest|omni-moderation-2024-09-26,,DOCUMENTED|LIVE_VERIFIED,text | image_url (multimodal array items).,https://developers.openai.com/api/reference/resources/moderations
openai,POST /v1/moderations,input[].text,body,string,false,,,,,omni-moderation-latest|omni-moderation-2024-09-26,,DOCUMENTED|LIVE_VERIFIED,Text to classify (type=text).,https://developers.openai.com/api/reference/resources/moderations
openai,POST /v1/moderations,input[].image_url.url,body,string (uri | data URL),false,,,,,omni-moderation-latest|omni-moderation-2024-09-26,,DOCUMENTED|LIVE_VERIFIED,"Image URL or data:image/...;base64,... (type=image_url).",https://developers.openai.com/api/reference/resources/moderations
openai,POST /v1/moderations,model,body,string,false,omni-moderation-latest,,,omni-moderation-latest|omni-moderation-2024-09-26|text-moderation-latest|text-moderation-stable|text-moderation-007,omni-moderation-latest|omni-moderation-2024-09-26,,DOCUMENTED|LIVE_VERIFIED,omni-moderation-latest (default) | omni-moderation-2024-09-26. text-moderation-latest/stable/007 were removed 2025-10-27.,https://developers.openai.com/api/reference/resources/moderations
openai,POST /v1/realtime/client_secrets,expires_after,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,"Configuration for the client secret expiration. Expiration refers to the time after which
a client secret will no longer be valid for creating sessions. The session itself may
continue after that time once started. A secret can be used to create multiple sessions
until it expires.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,expires_after.anchor,body,string,false,created_at,,,created_at,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,"The anchor point for the client secret expiration, meaning that `seconds` will be added to the `created_at` time of the client secret to produce an expiration timestamp. Only `created_at` is currently supported.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,expires_after.seconds,body,integer,false,600,10,7200,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,The number of seconds from the anchor point to the expiration. Select a value between `10` and `7200` (2 hours). This default to 600 seconds (10 minutes) if not specified.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session,body,RealtimeSessionCreateRequestGA | RealtimeTranscriptionSessionCreateRequestGA,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,"Session configuration to use for the client secret. Choose either a realtime
session or a transcription session.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.type,body,string,false,,,,realtime|transcription,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,The type of session to create. Always `realtime` for the Realtime API.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.output_modalities,body,array<string>,false,audio,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The set of modalities the model can respond with. It defaults to `[""audio""]`, indicating
that the model will respond with audio plus a transcript. `[""text""]` can be used to make
the model respond with text only. It is not possible to request both `text` and `audio` at the same time.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.model,body,string | enum,false,,,,gpt-realtime|gpt-realtime-1.5|gpt-realtime-2|gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2025-08-28|gpt-4o-realtime-preview|gpt-4o-realtime-preview-2024-10-01|gpt-4o-realtime-preview-2024-12-17|gpt-4o-realtime-preview-2025-06-03|gpt-4o-mini-realtime-preview|gpt-4o-mini-realtime-preview-2024-12-17|gpt-realtime-mini|gpt-realtime-mini-2025-10-06|gpt-realtime-mini-2025-12-15|gpt-audio-1.5|gpt-audio-mini|gpt-audio-mini-2025-10-06|gpt-audio-mini-2025-12-15,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,The Realtime model used for this session.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.instructions,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.

Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,Configuration for input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.format,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The format of the input audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.format.rate,body,integer,false,,,,24000,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,"Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.model,body,string | enum,false,,,,whisper-1|gpt-transcribe|gpt-live-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe|gpt-4o-transcribe-diarize|gpt-realtime-whisper,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED|LIVE_VERIFIED,"The model to use for transcription. Current options are `whisper-1`, `gpt-transcribe`, `gpt-live-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, `gpt-4o-transcribe-diarize`, and `gpt-realtime-whisper`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.language,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The language of the input audio. Supplying the input language in
[ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format
will improve accuracy and latency.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.languages,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.keywords,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.prompt,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"An optional text to guide the model's style or continue a previous audio
segment.
For `whisper-1`, the [prompt is a list of keywords](https://developers.openai.com/api/docs/guides/speech-to-text#prompting).
For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example ""expect words related to technology"".
Prompt is not supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.transcription.delay,body,string,false,,,,minimal|low|medium|high|xhigh,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Controls how long the model waits before emitting transcription text.
Higher values can improve transcription accuracy at the cost of latency.
Only supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.noise_reduction,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Configuration for input audio noise reduction. This can be set to `null` to turn off.
Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model.
Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.type,body,string,false,server_vad,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Type of turn detection, `server_vad` to turn on simple Server VAD.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.threshold,body,number,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.prefix_padding_ms,body,integer,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.silence_duration_ms,body,integer,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.create_response,body,boolean,false,true,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.interrupt_response,body,boolean,false,true,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Whether or not to automatically interrupt (cancel) any ongoing response with output to the default
conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.idle_timeout_ms,body,integer | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.input.turn_detection.eagerness,body,string,false,auto,,,low|medium|high|auto,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.format,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The format of the output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.format.rate,body,integer,false,,,,24000,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.voice,body,VoiceIdsShared | object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.voice.id,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.audio.output.speed,body,number,false,1,0.25,1.5,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The speed of the model's spoken response as a multiple of the original speed.
1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress.

This parameter is a post-processing adjustment to the audio after it is generated, it's
also possible to prompt the model to speak faster or slower.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.include,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Additional fields to include in server outputs.

`item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tracing,body,enum | object,false,,,,auto,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once
tracing is enabled for a session, the configuration cannot be modified.

`auto` will create a trace for the session with default values for the
workflow name, group id, and metadata.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tracing.workflow_name,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The name of the workflow to attach to this trace. This is used to
name the trace in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tracing.group_id,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The group id to attach to this trace to enable filtering and
grouping in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tracing.metadata,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The arbitrary metadata to attach to this trace to enable
filtering in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools,body,array<RealtimeFunctionTool | MCPTool>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Tools available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].type,body,string,false,,,,function|mcp,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].name,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].description,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].parameters,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].server_label,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].server_url,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].tunnel_id,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].authorization,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].server_description,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].headers,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].allowed_tools,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].allowed_tools.tool_names,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].allowed_tools.read_only,body,boolean,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].allowed_callers,body,array | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.always,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.always.tool_names,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.always.read_only,body,boolean,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.never,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.never.tool_names,body,array<string>,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].require_approval.never.read_only,body,boolean,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tools[].defer_loading,body,boolean,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Whether this MCP tool is deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tool_choice,body,ToolChoiceOptions | ToolChoiceFunction | ToolChoiceMCP,false,auto,,,none|auto|required,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"How the model chooses tools. Provide one of the string modes or force a specific
function/MCP tool.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tool_choice.type,body,string,false,,,,function|mcp,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"For function calling, the type is always `function`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tool_choice.name,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.tool_choice.server_label,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The label of the MCP server to use.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.parallel_tool_calls,body,boolean,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Whether the model may call multiple tools in parallel. Only supported by
reasoning Realtime models such as `gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.reasoning,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.reasoning.effort,body,string,false,low,,,minimal|low|medium|high|xhigh,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Constrains effort on reasoning for reasoning-capable Realtime models such as
`gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.max_output_tokens,body,integer | enum,false,,,,inf,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.truncation,body,enum | object,false,,,,auto|disabled,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs.

Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost.

Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate.

Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.truncation.type,body,string,false,,,,retention_ratio,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Use retention ratio truncation.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.truncation.retention_ratio,body,number,false,,0,1,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.truncation.token_limits,body,object,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.truncation.token_limits.post_instructions,body,integer,false,,0,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,"Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.prompt,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.prompt.id,body,string,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.prompt.version,body,string | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,POST /v1/realtime/client_secrets,session.prompt.variables,body,object | null,false,,,,,gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2|gpt-realtime-1.5|gpt-realtime|gpt-realtime-mini|gpt-4o-realtime-preview|gpt-4o-mini-realtime-preview,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create""}"
openai,WS /v1/realtime session.update,event_id,body,string,false,,,,,,,DOCUMENTED,"Optional client-generated ID used to identify this event. This is an arbitrary string that a client may assign. It will be passed back if there is an error with the event, but the corresponding `session.updated` event will not include it.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,type,body,string,true,,,,session.update,,,DOCUMENTED,"The event type, must be `session.update`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session,body,RealtimeSessionCreateRequestGA | RealtimeTranscriptionSessionCreateRequestGA,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Update the Realtime session. Choose either a realtime
session or a transcription session.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.type,body,string,false,,,,realtime|transcription,,,DOCUMENTED|LIVE_VERIFIED,The type of session to create. Always `realtime` for the Realtime API.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.output_modalities,body,array<string>,false,audio,,,,,,DOCUMENTED|LIVE_VERIFIED,"The set of modalities the model can respond with. It defaults to `[""audio""]`, indicating
that the model will respond with audio plus a transcript. `[""text""]` can be used to make
the model respond with text only. It is not possible to request both `text` and `audio` at the same time.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.model,body,string | enum,false,,,,gpt-realtime|gpt-realtime-1.5|gpt-realtime-2|gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2025-08-28|gpt-4o-realtime-preview|gpt-4o-realtime-preview-2024-10-01|gpt-4o-realtime-preview-2024-12-17|gpt-4o-realtime-preview-2025-06-03|gpt-4o-mini-realtime-preview|gpt-4o-mini-realtime-preview-2024-12-17|gpt-realtime-mini|gpt-realtime-mini-2025-10-06|gpt-realtime-mini-2025-12-15|gpt-audio-1.5|gpt-audio-mini|gpt-audio-mini-2025-10-06|gpt-audio-mini-2025-12-15,,,DOCUMENTED,The Realtime model used for this session.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.instructions,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.

Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio,body,object,false,,,,,,,DOCUMENTED,Configuration for input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.format,body,object,false,,,,,,,DOCUMENTED,The format of the input audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.model,body,string | enum,false,,,,whisper-1|gpt-transcribe|gpt-live-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe|gpt-4o-transcribe-diarize|gpt-realtime-whisper,,,DOCUMENTED,"The model to use for transcription. Current options are `whisper-1`, `gpt-transcribe`, `gpt-live-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, `gpt-4o-transcribe-diarize`, and `gpt-realtime-whisper`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.language,body,string,false,,,,,,,DOCUMENTED,"The language of the input audio. Supplying the input language in
[ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format
will improve accuracy and latency.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.languages,body,array<string>,false,,,,,,,DOCUMENTED,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.keywords,body,array<string>,false,,,,,,,DOCUMENTED,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.prompt,body,string,false,,,,,,,DOCUMENTED,"An optional text to guide the model's style or continue a previous audio
segment.
For `whisper-1`, the [prompt is a list of keywords](https://developers.openai.com/api/docs/guides/speech-to-text#prompting).
For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example ""expect words related to technology"".
Prompt is not supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.transcription.delay,body,string,false,,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Controls how long the model waits before emitting transcription text.
Higher values can improve transcription accuracy at the cost of latency.
Only supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.noise_reduction,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio noise reduction. This can be set to `null` to turn off.
Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model.
Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.type,body,string,false,server_vad,,,,,,DOCUMENTED,"Type of turn detection, `server_vad` to turn on simple Server VAD.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.threshold,body,number,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.prefix_padding_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.silence_duration_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.create_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.interrupt_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically interrupt (cancel) any ongoing response with output to the default
conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.idle_timeout_ms,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.input.turn_detection.eagerness,body,string,false,auto,,,low|medium|high|auto,,,DOCUMENTED,"Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.format,body,object,false,,,,,,,DOCUMENTED,The format of the output audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.voice,body,VoiceIdsShared | object,false,,,,,,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.voice.id,body,string,false,,,,,,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.audio.output.speed,body,number,false,1,0.25,1.5,,,,DOCUMENTED,"The speed of the model's spoken response as a multiple of the original speed.
1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress.

This parameter is a post-processing adjustment to the audio after it is generated, it's
also possible to prompt the model to speak faster or slower.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.include,body,array<string>,false,,,,,,,DOCUMENTED,"Additional fields to include in server outputs.

`item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tracing,body,enum | object,false,,,,auto,,,DOCUMENTED,"Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once
tracing is enabled for a session, the configuration cannot be modified.

`auto` will create a trace for the session with default values for the
workflow name, group id, and metadata.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tracing.workflow_name,body,string,false,,,,,,,DOCUMENTED,"The name of the workflow to attach to this trace. This is used to
name the trace in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tracing.group_id,body,string,false,,,,,,,DOCUMENTED,"The group id to attach to this trace to enable filtering and
grouping in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tracing.metadata,body,object,false,,,,,,,DOCUMENTED,"The arbitrary metadata to attach to this trace to enable
filtering in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools,body,array<RealtimeFunctionTool | MCPTool>,false,,,,,,,DOCUMENTED,Tools available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].type,body,string,false,,,,function|mcp,,,DOCUMENTED,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].description,body,string,false,,,,,,,DOCUMENTED,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].parameters,body,object,false,,,,,,,DOCUMENTED,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].server_label,body,string,false,,,,,,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].server_url,body,string,false,,,,,,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].tunnel_id,body,string,false,,,,,,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].authorization,body,string,false,,,,,,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].server_description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].headers,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].allowed_tools,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].allowed_tools.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].allowed_tools.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.always,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.always.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.always.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.never,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.never.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].require_approval.never.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this MCP tool is deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tool_choice,body,ToolChoiceOptions | ToolChoiceFunction | ToolChoiceMCP,false,auto,,,none|auto|required,,,DOCUMENTED,"How the model chooses tools. Provide one of the string modes or force a specific
function/MCP tool.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tool_choice.type,body,string,false,,,,function|mcp,,,DOCUMENTED,"For function calling, the type is always `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tool_choice.name,body,string,false,,,,,,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.parallel_tool_calls,body,boolean,false,,,,,,,DOCUMENTED,"Whether the model may call multiple tools in parallel. Only supported by
reasoning Realtime models such as `gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.reasoning,body,object,false,,,,,,,DOCUMENTED,Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.reasoning.effort,body,string,false,low,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Constrains effort on reasoning for reasoning-capable Realtime models such as
`gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.max_output_tokens,body,integer | enum,false,,,,inf,,,DOCUMENTED,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.truncation,body,enum | object,false,,,,auto|disabled,,,DOCUMENTED,"When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs.

Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost.

Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate.

Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.truncation.type,body,string,false,,,,retention_ratio,,,DOCUMENTED,Use retention ratio truncation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.truncation.retention_ratio,body,number,false,,0,1,,,,DOCUMENTED,Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.truncation.token_limits,body,object,false,,,,,,,DOCUMENTED,"Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.truncation.token_limits.post_instructions,body,integer,false,,0,,,,,DOCUMENTED,"Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.prompt,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.prompt.id,body,string,false,,,,,,,DOCUMENTED,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.prompt.version,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime session.update,session.prompt.variables,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#session.update""}"
openai,WS /v1/realtime response.create,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,type,body,string,true,,,,response.create,,,DOCUMENTED,"The event type, must be `response.create`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Create a new Realtime response with these parameters,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.output_modalities,body,array<string>,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"The set of modalities the model used to respond, currently the only possible values are
`[\""audio\""]`, `[\""text\""]`. Audio output always include a text transcript. Setting the
output to mode `text` will disable audio output from the model.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.instructions,body,string,false,,,,,,,DOCUMENTED,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.
Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio,body,object,false,,,,,,,DOCUMENTED,Configuration for audio input and output.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output.format,body,object,false,,,,,,,DOCUMENTED,The format of the output audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output.voice,body,VoiceIdsShared | object,false,,,,,,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.audio.output.voice.id,body,string,false,,,,,,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools,body,array<RealtimeFunctionTool | MCPTool>,false,,,,,,,DOCUMENTED,Tools available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].type,body,string,false,,,,function|mcp,,,DOCUMENTED,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].description,body,string,false,,,,,,,DOCUMENTED,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].parameters,body,object,false,,,,,,,DOCUMENTED,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].server_label,body,string,false,,,,,,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].server_url,body,string,false,,,,,,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].tunnel_id,body,string,false,,,,,,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].authorization,body,string,false,,,,,,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].server_description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].headers,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].allowed_tools,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].allowed_tools.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].allowed_tools.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.always,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.always.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.always.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.never,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.never.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].require_approval.never.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this MCP tool is deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tool_choice,body,ToolChoiceOptions | ToolChoiceFunction | ToolChoiceMCP,false,auto,,,none|auto|required,,,DOCUMENTED,"How the model chooses tools. Provide one of the string modes or force a specific
function/MCP tool.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tool_choice.type,body,string,false,,,,function|mcp,,,DOCUMENTED,"For function calling, the type is always `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tool_choice.name,body,string,false,,,,,,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.parallel_tool_calls,body,boolean,false,,,,,,,DOCUMENTED,"Whether the model may call multiple tools in parallel. Only supported by
reasoning Realtime models such as `gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.reasoning,body,object,false,,,,,,,DOCUMENTED,Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.reasoning.effort,body,string,false,low,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Constrains effort on reasoning for reasoning-capable Realtime models such as
`gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.max_output_tokens,body,integer | enum,false,,,,inf,,,DOCUMENTED|LIVE_VERIFIED,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.conversation,body,string | enum,false,,,,auto|none,,,DOCUMENTED,"Controls which conversation the response is added to. Currently supports
`auto` and `none`, with `auto` as the default value. The `auto` value
means that the contents of the response will be added to the default
conversation. Set this to `none` to create an out-of-band response which
will not add items to default conversation.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.metadata,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.prompt,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.prompt.id,body,string,false,,,,,,,DOCUMENTED,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.prompt.version,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.prompt.variables,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input,body,array<RealtimeConversationItem>,false,,,,,,,DOCUMENTED,"Input items to include in the prompt for the model. Using this field
creates a new context for this Response instead of using the default
conversation. An empty array `[]` will clear the context for this Response.
Note that this can include references to items that previously appeared in the session
using their id.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].id,body,string,false,,,,,,,DOCUMENTED,The unique ID of the item. This may be provided by the client or generated by the server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].object,body,string,false,,,,realtime.item,,,DOCUMENTED,Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].type,body,string,false,,,,message|function_call|function_call_output|mcp_approval_response|mcp_list_tools|mcp_call|mcp_approval_request,,,DOCUMENTED,The type of the item. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].status,body,string,false,,,,completed|incomplete|in_progress,,,DOCUMENTED,The status of the item. Has no effect on the conversation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].role,body,string,false,,,,system|user|assistant,,,DOCUMENTED,The role of the message sender. Always `system`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content,body,array<object>,false,,,,,,,DOCUMENTED,The content of the message.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].type,body,string,false,,,,input_text|input_audio|input_image|output_text|output_audio,,,DOCUMENTED,The content type. Always `input_text` for system messages.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].text,body,string,false,,,,,,,DOCUMENTED,The text content.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].audio,body,string,false,,,,,,,DOCUMENTED,"Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].image_url,body,string,false,,,,,,,DOCUMENTED,"Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].detail,body,string,false,auto,,,auto|low|high,,,DOCUMENTED,The detail level of the image (for `input_image`). `auto` will default to `high`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].content[].transcript,body,string,false,,,,,,,DOCUMENTED,"Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].call_id,body,string,false,,,,,,,DOCUMENTED,The ID of the function call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function being called.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].arguments,body,string,false,,,,,,,DOCUMENTED,"The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{""arg1"": ""value1"", ""arg2"": 42}`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].output,body,string,false,,,,,,,DOCUMENTED,"The output of the function call, this is free text and can contain any information or simply be empty.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].approval_request_id,body,string,false,,,,,,,DOCUMENTED,The ID of the approval request being answered.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].approve,body,boolean,false,,,,,,,DOCUMENTED,Whether the request was approved.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].reason,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].tools,body,array<MCPListToolsTool>,false,,,,,,,DOCUMENTED,The tools available on the server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the tool.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].tools[].description,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].tools[].input_schema,body,object,false,,,,,,,DOCUMENTED,The JSON schema describing the tool's input.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].tools[].annotations,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].error,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].error.type,body,string,false,,,,protocol_error|tool_execution_error|http_error,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].error.code,body,integer,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime response.create,response.input[].error.message,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.create""}"
openai,WS /v1/realtime conversation.item.create,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,type,body,string,true,,,,conversation.item.create,,,DOCUMENTED,"The event type, must be `conversation.item.create`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,previous_item_id,body,string,false,,,,,,,DOCUMENTED,"The ID of the preceding item after which the new item will be inserted. If not set, the new item will be appended to the end of the conversation.

If set to `root`, the new item will be added to the beginning of the conversation.

If set to an existing ID, it allows an item to be inserted mid-conversation. If the ID cannot be found, an error will be returned and the item will not be added.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item,body,RealtimeConversationItemMessageSystem | RealtimeConversationItemMessageUser | RealtimeConversationItemMessageAssistant | RealtimeConversationItemFunctionCall | RealtimeConversationItemFunctionCallOutput | RealtimeMCPApprovalResponse | RealtimeMCPListTools | RealtimeMCPToolCall | RealtimeMCPApprovalRequest,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,A single item within a Realtime conversation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.id,body,string,false,,,,,,,DOCUMENTED,The unique ID of the item. This may be provided by the client or generated by the server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.object,body,string,false,,,,realtime.item,,,DOCUMENTED,Identifier for the API object being returned - always `realtime.item`. Optional when creating a new item.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.type,body,string,false,,,,message|function_call|function_call_output|mcp_approval_response|mcp_list_tools|mcp_call|mcp_approval_request,,,DOCUMENTED|LIVE_VERIFIED,The type of the item. Always `message`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.status,body,string,false,,,,completed|incomplete|in_progress,,,DOCUMENTED,The status of the item. Has no effect on the conversation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.role,body,string,false,,,,system|user|assistant,,,DOCUMENTED|LIVE_VERIFIED,The role of the message sender. Always `system`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content,body,array<object>,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,The content of the message.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].type,body,string,false,,,,input_text|input_audio|input_image|output_text|output_audio,,,DOCUMENTED|LIVE_VERIFIED,The content type. Always `input_text` for system messages.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].text,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,The text content.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].audio,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Base64-encoded audio bytes (for `input_audio`), these will be parsed as the format specified in the session input audio type configuration. This defaults to PCM 16-bit 24kHz mono if not specified.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].image_url,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Base64-encoded image bytes (for `input_image`) as a data URI. For example `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...`. Supported formats are PNG and JPEG.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].detail,body,string,false,auto,,,auto|low|high,,,DOCUMENTED|LIVE_VERIFIED,The detail level of the image (for `input_image`). `auto` will default to `high`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.content[].transcript,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Transcript of the audio (for `input_audio`). This is not sent to the model, but will be attached to the message item for reference.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.call_id,body,string,false,,,,,,,DOCUMENTED,The ID of the function call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.name,body,string,false,,,,,,,DOCUMENTED,The name of the function being called.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.arguments,body,string,false,,,,,,,DOCUMENTED,"The arguments of the function call. This is a JSON-encoded string representing the arguments passed to the function, for example `{""arg1"": ""value1"", ""arg2"": 42}`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.output,body,string,false,,,,,,,DOCUMENTED,"The output of the function call, this is free text and can contain any information or simply be empty.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.approval_request_id,body,string,false,,,,,,,DOCUMENTED,The ID of the approval request being answered.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.approve,body,boolean,false,,,,,,,DOCUMENTED,Whether the request was approved.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.reason,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.tools,body,array<MCPListToolsTool>,false,,,,,,,DOCUMENTED,The tools available on the server.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the tool.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.tools[].description,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.tools[].input_schema,body,object,false,,,,,,,DOCUMENTED,The JSON schema describing the tool's input.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.tools[].annotations,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.error,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.error.type,body,string,false,,,,protocol_error|tool_execution_error|http_error,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.error.code,body,integer,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime conversation.item.create,item.error.message,body,string,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.create""}"
openai,WS /v1/realtime input_audio_buffer.append,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.append""}"
openai,WS /v1/realtime input_audio_buffer.append,type,body,string,true,,,,input_audio_buffer.append,,,DOCUMENTED,"The event type, must be `input_audio_buffer.append`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.append""}"
openai,WS /v1/realtime input_audio_buffer.append,audio,body,string,true,,,,,,,DOCUMENTED,"Base64-encoded audio bytes. This must be in the format specified by the 
`input_audio_format` field in the session configuration.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.append""}"
openai,WS /v1/realtime input_audio_buffer.commit,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.commit""}"
openai,WS /v1/realtime input_audio_buffer.commit,type,body,string,true,,,,input_audio_buffer.commit,,,DOCUMENTED,"The event type, must be `input_audio_buffer.commit`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.commit""}"
openai,WS /v1/realtime input_audio_buffer.clear,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.clear""}"
openai,WS /v1/realtime input_audio_buffer.clear,type,body,string,true,,,,input_audio_buffer.clear,,,DOCUMENTED,"The event type, must be `input_audio_buffer.clear`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#input_audio_buffer.clear""}"
openai,WS /v1/realtime conversation.item.retrieve,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.retrieve""}"
openai,WS /v1/realtime conversation.item.retrieve,type,body,string,true,,,,conversation.item.retrieve,,,DOCUMENTED,"The event type, must be `conversation.item.retrieve`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.retrieve""}"
openai,WS /v1/realtime conversation.item.retrieve,item_id,body,string,true,,,,,,,DOCUMENTED,The ID of the item to retrieve.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.retrieve""}"
openai,WS /v1/realtime conversation.item.truncate,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.truncate""}"
openai,WS /v1/realtime conversation.item.truncate,type,body,string,true,,,,conversation.item.truncate,,,DOCUMENTED,"The event type, must be `conversation.item.truncate`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.truncate""}"
openai,WS /v1/realtime conversation.item.truncate,item_id,body,string,true,,,,,,,DOCUMENTED,"The ID of the assistant message item to truncate. Only assistant message 
items can be truncated.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.truncate""}"
openai,WS /v1/realtime conversation.item.truncate,content_index,body,integer,true,,,,,,,DOCUMENTED,The index of the content part to truncate. Set this to `0`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.truncate""}"
openai,WS /v1/realtime conversation.item.truncate,audio_end_ms,body,integer,true,,,,,,,DOCUMENTED,"Inclusive duration up to which audio is truncated, in milliseconds. If 
the audio_end_ms is greater than the actual audio duration, the server 
will respond with an error.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.truncate""}"
openai,WS /v1/realtime conversation.item.delete,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.delete""}"
openai,WS /v1/realtime conversation.item.delete,type,body,string,true,,,,conversation.item.delete,,,DOCUMENTED,"The event type, must be `conversation.item.delete`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.delete""}"
openai,WS /v1/realtime conversation.item.delete,item_id,body,string,true,,,,,,,DOCUMENTED,The ID of the item to delete.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#conversation.item.delete""}"
openai,WS /v1/realtime response.cancel,event_id,body,string,false,,,,,,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.cancel""}"
openai,WS /v1/realtime response.cancel,type,body,string,true,,,,response.cancel,,,DOCUMENTED,"The event type, must be `response.cancel`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.cancel""}"
openai,WS /v1/realtime response.cancel,response_id,body,string,false,,,,,,,DOCUMENTED,"A specific response ID to cancel - if not provided, will cancel an 
in-progress response in the default conversation.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#response.cancel""}"
openai,WS /v1/realtime output_audio_buffer.clear,event_id,body,string,false,,,,,,,DOCUMENTED,The unique ID of the client event used for error handling.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#output_audio_buffer.clear""}"
openai,WS /v1/realtime output_audio_buffer.clear,type,body,string,true,,,,output_audio_buffer.clear,,,DOCUMENTED,"The event type, must be `output_audio_buffer.clear`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/client-events#output_audio_buffer.clear""}"
openai,POST /v1/realtime/calls,sdp,body,string,true,,,,,,,DOCUMENTED,WebRTC Session Description Protocol (SDP) offer generated by the caller.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session,body,any,false,,,,,,,DOCUMENTED,"Optional session configuration to apply before the realtime session is
created. Use the same parameters you would send in a [`create client secret`](https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets/methods/create)
request.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.type,body,string,false,,,,realtime,,,DOCUMENTED,The type of session to create. Always `realtime` for the Realtime API.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.output_modalities,body,array<string>,false,audio,,,,,,DOCUMENTED,"The set of modalities the model can respond with. It defaults to `[""audio""]`, indicating
that the model will respond with audio plus a transcript. `[""text""]` can be used to make
the model respond with text only. It is not possible to request both `text` and `audio` at the same time.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.model,body,string | enum,false,,,,gpt-realtime|gpt-realtime-1.5|gpt-realtime-2|gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2025-08-28|gpt-4o-realtime-preview|gpt-4o-realtime-preview-2024-10-01|gpt-4o-realtime-preview-2024-12-17|gpt-4o-realtime-preview-2025-06-03|gpt-4o-mini-realtime-preview|gpt-4o-mini-realtime-preview-2024-12-17|gpt-realtime-mini|gpt-realtime-mini-2025-10-06|gpt-realtime-mini-2025-12-15|gpt-audio-1.5|gpt-audio-mini|gpt-audio-mini-2025-10-06|gpt-audio-mini-2025-12-15,,,DOCUMENTED,The Realtime model used for this session.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.instructions,body,string,false,,,,,,,DOCUMENTED,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.

Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio,body,object,false,,,,,,,DOCUMENTED,Configuration for input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.format,body,object,false,,,,,,,DOCUMENTED,The format of the input audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.model,body,string | enum,false,,,,whisper-1|gpt-transcribe|gpt-live-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe|gpt-4o-transcribe-diarize|gpt-realtime-whisper,,,DOCUMENTED,"The model to use for transcription. Current options are `whisper-1`, `gpt-transcribe`, `gpt-live-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, `gpt-4o-transcribe-diarize`, and `gpt-realtime-whisper`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.language,body,string,false,,,,,,,DOCUMENTED,"The language of the input audio. Supplying the input language in
[ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format
will improve accuracy and latency.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.languages,body,array<string>,false,,,,,,,DOCUMENTED,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.keywords,body,array<string>,false,,,,,,,DOCUMENTED,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.prompt,body,string,false,,,,,,,DOCUMENTED,"An optional text to guide the model's style or continue a previous audio
segment.
For `whisper-1`, the [prompt is a list of keywords](https://developers.openai.com/api/docs/guides/speech-to-text#prompting).
For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example ""expect words related to technology"".
Prompt is not supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.transcription.delay,body,string,false,,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Controls how long the model waits before emitting transcription text.
Higher values can improve transcription accuracy at the cost of latency.
Only supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.noise_reduction,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio noise reduction. This can be set to `null` to turn off.
Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model.
Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.type,body,string,false,server_vad,,,,,,DOCUMENTED,"Type of turn detection, `server_vad` to turn on simple Server VAD.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.threshold,body,number,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.prefix_padding_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.silence_duration_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.create_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.interrupt_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically interrupt (cancel) any ongoing response with output to the default
conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.idle_timeout_ms,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.input.turn_detection.eagerness,body,string,false,auto,,,low|medium|high|auto,,,DOCUMENTED,"Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.format,body,object,false,,,,,,,DOCUMENTED,The format of the output audio.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.voice,body,VoiceIdsShared | object,false,,,,,,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.voice.id,body,string,false,,,,,,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.audio.output.speed,body,number,false,1,0.25,1.5,,,,DOCUMENTED,"The speed of the model's spoken response as a multiple of the original speed.
1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress.

This parameter is a post-processing adjustment to the audio after it is generated, it's
also possible to prompt the model to speak faster or slower.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.include,body,array<string>,false,,,,,,,DOCUMENTED,"Additional fields to include in server outputs.

`item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tracing,body,enum | object,false,,,,auto,,,DOCUMENTED,"Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once
tracing is enabled for a session, the configuration cannot be modified.

`auto` will create a trace for the session with default values for the
workflow name, group id, and metadata.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tracing.workflow_name,body,string,false,,,,,,,DOCUMENTED,"The name of the workflow to attach to this trace. This is used to
name the trace in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tracing.group_id,body,string,false,,,,,,,DOCUMENTED,"The group id to attach to this trace to enable filtering and
grouping in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tracing.metadata,body,object,false,,,,,,,DOCUMENTED,"The arbitrary metadata to attach to this trace to enable
filtering in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools,body,array<RealtimeFunctionTool | MCPTool>,false,,,,,,,DOCUMENTED,Tools available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].type,body,string,false,,,,function|mcp,,,DOCUMENTED,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].description,body,string,false,,,,,,,DOCUMENTED,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].parameters,body,object,false,,,,,,,DOCUMENTED,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].server_label,body,string,false,,,,,,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].server_url,body,string,false,,,,,,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].tunnel_id,body,string,false,,,,,,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].authorization,body,string,false,,,,,,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].server_description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].headers,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].allowed_tools,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].allowed_tools.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].allowed_tools.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.always,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.always.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.always.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.never,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.never.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].require_approval.never.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this MCP tool is deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tool_choice,body,ToolChoiceOptions | ToolChoiceFunction | ToolChoiceMCP,false,auto,,,none|auto|required,,,DOCUMENTED,"How the model chooses tools. Provide one of the string modes or force a specific
function/MCP tool.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tool_choice.type,body,string,false,,,,function|mcp,,,DOCUMENTED,"For function calling, the type is always `function`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tool_choice.name,body,string,false,,,,,,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.parallel_tool_calls,body,boolean,false,,,,,,,DOCUMENTED,"Whether the model may call multiple tools in parallel. Only supported by
reasoning Realtime models such as `gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.reasoning,body,object,false,,,,,,,DOCUMENTED,Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.reasoning.effort,body,string,false,low,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Constrains effort on reasoning for reasoning-capable Realtime models such as
`gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.max_output_tokens,body,integer | enum,false,,,,inf,,,DOCUMENTED,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.truncation,body,enum | object,false,,,,auto|disabled,,,DOCUMENTED,"When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs.

Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost.

Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate.

Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.truncation.type,body,string,false,,,,retention_ratio,,,DOCUMENTED,Use retention ratio truncation.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.truncation.retention_ratio,body,number,false,,0,1,,,,DOCUMENTED,Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.truncation.token_limits,body,object,false,,,,,,,DOCUMENTED,"Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.truncation.token_limits.post_instructions,body,integer,false,,0,,,,,DOCUMENTED,"Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.prompt,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.prompt.id,body,string,false,,,,,,,DOCUMENTED,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.prompt.version,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session.prompt.variables,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls,session (JSON string),body,RealtimeSessionCreateRequestGA | RealtimeTranscriptionSessionCreateRequestGA,false,,,,,,,DOCUMENTED,"multipart/form-data part `session` (type=application/json) carrying the same session object as POST /v1/realtime/client_secrets `session` (see those records). When the request body is raw `application/sdp`, no session config can be attached (use an ephemeral client secret minted with the config instead).","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/create""}"
openai,POST /v1/realtime/calls/{call_id}/accept,type,body,string,true,,,,realtime,,,DOCUMENTED,The type of session to create. Always `realtime` for the Realtime API.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,output_modalities,body,array<string>,false,audio,,,,,,DOCUMENTED,"The set of modalities the model can respond with. It defaults to `[""audio""]`, indicating
that the model will respond with audio plus a transcript. `[""text""]` can be used to make
the model respond with text only. It is not possible to request both `text` and `audio` at the same time.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,model,body,string | enum,false,,,,gpt-realtime|gpt-realtime-1.5|gpt-realtime-2|gpt-realtime-2.1|gpt-realtime-2.1-mini|gpt-realtime-2025-08-28|gpt-4o-realtime-preview|gpt-4o-realtime-preview-2024-10-01|gpt-4o-realtime-preview-2024-12-17|gpt-4o-realtime-preview-2025-06-03|gpt-4o-mini-realtime-preview|gpt-4o-mini-realtime-preview-2024-12-17|gpt-realtime-mini|gpt-realtime-mini-2025-10-06|gpt-realtime-mini-2025-12-15|gpt-audio-1.5|gpt-audio-mini|gpt-audio-mini-2025-10-06|gpt-audio-mini-2025-12-15,,,DOCUMENTED,The Realtime model used for this session.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,instructions,body,string,false,,,,,,,DOCUMENTED,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.

Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio,body,object,false,,,,,,,DOCUMENTED,Configuration for input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.format,body,object,false,,,,,,,DOCUMENTED,The format of the input audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio transcription, defaults to off and can be set to `null` to turn off once on. Input audio transcription is not native to the model, since the model consumes audio directly. Transcription runs asynchronously through [the /audio/transcriptions endpoint](https://developers.openai.com/api/reference/resources/audio/subresources/transcriptions/methods/create) and should be treated as guidance of input audio content rather than precisely what the model heard. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.model,body,string | enum,false,,,,whisper-1|gpt-transcribe|gpt-live-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe|gpt-4o-transcribe-diarize|gpt-realtime-whisper,,,DOCUMENTED,"The model to use for transcription. Current options are `whisper-1`, `gpt-transcribe`, `gpt-live-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, `gpt-4o-transcribe-diarize`, and `gpt-realtime-whisper`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.language,body,string,false,,,,,,,DOCUMENTED,"The language of the input audio. Supplying the input language in
[ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format
will improve accuracy and latency.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.languages,body,array<string>,false,,,,,,,DOCUMENTED,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.keywords,body,array<string>,false,,,,,,,DOCUMENTED,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.prompt,body,string,false,,,,,,,DOCUMENTED,"An optional text to guide the model's style or continue a previous audio
segment.
For `whisper-1`, the [prompt is a list of keywords](https://developers.openai.com/api/docs/guides/speech-to-text#prompting).
For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example ""expect words related to technology"".
Prompt is not supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.transcription.delay,body,string,false,,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Controls how long the model waits before emitting transcription text.
Higher values can improve transcription accuracy at the cost of latency.
Only supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.noise_reduction,body,object,false,,,,,,,DOCUMENTED,"Configuration for input audio noise reduction. This can be set to `null` to turn off.
Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model.
Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.type,body,string,false,server_vad,,,,,,DOCUMENTED,"Type of turn detection, `server_vad` to turn on simple Server VAD.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.threshold,body,number,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.prefix_padding_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.silence_duration_ms,body,integer,false,,,,,,,DOCUMENTED,"Used only for `server_vad` mode. Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.create_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically generate a response when a VAD stop event occurs. If `interrupt_response` is set to `false` this may fail to create a response if the model is already responding.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.interrupt_response,body,boolean,false,true,,,,,,DOCUMENTED,"Whether or not to automatically interrupt (cancel) any ongoing response with output to the default
conversation (i.e. `conversation` of `auto`) when a VAD start event occurs. If `true` then the response will be cancelled, otherwise it will continue until complete.

If both `create_response` and `interrupt_response` are set to `false`, the model will never respond automatically but VAD events will still be emitted.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.idle_timeout_ms,body,integer | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.input.turn_detection.eagerness,body,string,false,auto,,,low|medium|high|auto,,,DOCUMENTED,"Used only for `semantic_vad` mode. The eagerness of the model to respond. `low` will wait longer for the user to continue speaking, `high` will respond more quickly. `auto` is the default and is equivalent to `medium`. `low`, `medium`, and `high` have max timeouts of 8s, 4s, and 2s respectively.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output,body,object,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.format,body,object,false,,,,,,,DOCUMENTED,The format of the output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.format.type,body,string,false,,,,audio/pcm|audio/pcmu|audio/pcma,,,DOCUMENTED,The audio format. Always `audio/pcm`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.format.rate,body,integer,false,,,,24000,,,DOCUMENTED,The sample rate of the audio. Always `24000`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.voice,body,VoiceIdsShared | object,false,,,,,,,DOCUMENTED,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.voice.id,body,string,false,,,,,,,DOCUMENTED,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,audio.output.speed,body,number,false,1,0.25,1.5,,,,DOCUMENTED,"The speed of the model's spoken response as a multiple of the original speed.
1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed. This value can only be changed in between model turns, not while a response is in progress.

This parameter is a post-processing adjustment to the audio after it is generated, it's
also possible to prompt the model to speak faster or slower.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,include,body,array<string>,false,,,,,,,DOCUMENTED,"Additional fields to include in server outputs.

`item.input_audio_transcription.logprobs`: Include logprobs for input audio transcription.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tracing,body,enum | object,false,,,,auto,,,DOCUMENTED,"Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once
tracing is enabled for a session, the configuration cannot be modified.

`auto` will create a trace for the session with default values for the
workflow name, group id, and metadata.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tracing.workflow_name,body,string,false,,,,,,,DOCUMENTED,"The name of the workflow to attach to this trace. This is used to
name the trace in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tracing.group_id,body,string,false,,,,,,,DOCUMENTED,"The group id to attach to this trace to enable filtering and
grouping in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tracing.metadata,body,object,false,,,,,,,DOCUMENTED,"The arbitrary metadata to attach to this trace to enable
filtering in the Traces Dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools,body,array<RealtimeFunctionTool | MCPTool>,false,,,,,,,DOCUMENTED,Tools available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].type,body,string,false,,,,function|mcp,,,DOCUMENTED,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].name,body,string,false,,,,,,,DOCUMENTED,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].description,body,string,false,,,,,,,DOCUMENTED,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].parameters,body,object,false,,,,,,,DOCUMENTED,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].server_label,body,string,false,,,,,,,DOCUMENTED,"A label for this MCP server, used to identify it in tool calls.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].server_url,body,string,false,,,,,,,DOCUMENTED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,,,DOCUMENTED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].tunnel_id,body,string,false,,,,,,,DOCUMENTED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].authorization,body,string,false,,,,,,,DOCUMENTED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].server_description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the MCP server, used to provide more context.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].headers,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].allowed_tools,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].allowed_tools.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].allowed_tools.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].allowed_callers,body,array | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.always,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.always.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.always.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.never,body,object,false,,,,,,,DOCUMENTED,A filter object to specify which tools are allowed.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.never.tool_names,body,array<string>,false,,,,,,,DOCUMENTED,List of allowed tool names.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].require_approval.never.read_only,body,boolean,false,,,,,,,DOCUMENTED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tools[].defer_loading,body,boolean,false,,,,,,,DOCUMENTED,Whether this MCP tool is deferred and discovered via tool search.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tool_choice,body,ToolChoiceOptions | ToolChoiceFunction | ToolChoiceMCP,false,auto,,,none|auto|required,,,DOCUMENTED,"How the model chooses tools. Provide one of the string modes or force a specific
function/MCP tool.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tool_choice.type,body,string,false,,,,function|mcp,,,DOCUMENTED,"For function calling, the type is always `function`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tool_choice.name,body,string,false,,,,,,,DOCUMENTED,The name of the function to call.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,tool_choice.server_label,body,string,false,,,,,,,DOCUMENTED,The label of the MCP server to use.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,parallel_tool_calls,body,boolean,false,,,,,,,DOCUMENTED,"Whether the model may call multiple tools in parallel. Only supported by
reasoning Realtime models such as `gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,reasoning,body,object,false,,,,,,,DOCUMENTED,Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,reasoning.effort,body,string,false,low,,,minimal|low|medium|high|xhigh,,,DOCUMENTED,"Constrains effort on reasoning for reasoning-capable Realtime models such as
`gpt-realtime-2`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,max_output_tokens,body,integer | enum,false,,,,inf,,,DOCUMENTED,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,truncation,body,enum | object,false,,,,auto|disabled,,,DOCUMENTED,"When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs.

Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost.

Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate.

Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,truncation.type,body,string,false,,,,retention_ratio,,,DOCUMENTED,Use retention ratio truncation.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,truncation.retention_ratio,body,number,false,,0,1,,,,DOCUMENTED,Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,truncation.token_limits,body,object,false,,,,,,,DOCUMENTED,"Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,truncation.token_limits.post_instructions,body,integer,false,,0,,,,,DOCUMENTED,"Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,prompt,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,prompt.id,body,string,false,,,,,,,DOCUMENTED,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,prompt.version,body,string | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,prompt.variables,body,object | null,false,,,,,,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/accept,call_id,path,string,true,,,,,,,DOCUMENTED,Path parameter: call ID from the `realtime.call.incoming` webhook.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/accept""}"
openai,POST /v1/realtime/calls/{call_id}/refer,target_uri,body,string,true,,,,,,,DOCUMENTED,"URI that should appear in the SIP Refer-To header. Supports values like
`tel:+14155550123` or `sip:agent@example.com`.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/refer""}"
openai,POST /v1/realtime/calls/{call_id}/refer,call_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/refer""}"
openai,POST /v1/realtime/calls/{call_id}/reject,status_code,body,integer,false,,,,,,,DOCUMENTED,"SIP response code to send back to the caller. Defaults to `603` (Decline)
when omitted.","{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/reject""}"
openai,POST /v1/realtime/calls/{call_id}/reject,call_id,path,string,true,,,,,,,DOCUMENTED,Path parameter.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/reject""}"
openai,POST /v1/realtime/calls/{call_id}/hangup,call_id,path,string,true,,,,,,,DOCUMENTED,Path parameter. No request body.,"{""retrieved_at"": ""2026-09-19T01:31:51Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/calls/methods/hangup""}"
openai,POST /v1/realtime/translations/client_secrets,expires_after,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,"Configuration for the client secret expiration. Expiration refers to the time after which
a client secret will no longer be valid for creating sessions. The session itself may
continue after that time once started. A secret can be used to create multiple sessions
until it expires.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,expires_after.anchor,body,string,false,created_at,,,created_at,gpt-realtime-translate,,DOCUMENTED,"The anchor point for the client secret expiration, meaning that `seconds` will be added to the `created_at` time of the client secret to produce an expiration timestamp. Only `created_at` is currently supported.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,expires_after.seconds,body,integer,false,600,10,7200,,gpt-realtime-translate,,DOCUMENTED,The number of seconds from the anchor point to the expiration. Select a value between `10` and `7200` (2 hours). This default to 600 seconds (10 minutes) if not specified.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session,body,object,true,,,,,gpt-realtime-translate,,DOCUMENTED|LIVE_VERIFIED,"Realtime translation session configuration. Translation sessions stream source
audio in and translated audio plus transcript deltas out continuously.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.model,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED|LIVE_VERIFIED,The Realtime translation model used for this session.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,Configuration for translation input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.input,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.input.transcription,body,object | null,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.input.transcription.model,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,The transcription model to use for source transcript deltas.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.input.noise_reduction,body,object | null,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,gpt-realtime-translate,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.output,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,POST /v1/realtime/translations/client_secrets,session.audio.output.language,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,Target language for translated output audio and transcript deltas.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/subresources/client_secrets""}"
openai,WS /v1/realtime/translations session.update,event_id,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,type,body,string,true,,,,session.update,gpt-realtime-translate,,DOCUMENTED,"The event type, must be `session.update`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session,body,object,true,,,,,gpt-realtime-translate,,DOCUMENTED,Realtime translation session fields that can be updated with `session.update`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,Configuration for translation input and output audio.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.input,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.input.transcription,body,object | null,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.input.transcription.model,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,The transcription model to use for source transcript deltas.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.input.noise_reduction,body,object | null,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.input.noise_reduction.type,body,string,false,,,,near_field|far_field,gpt-realtime-translate,,DOCUMENTED,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.output,body,object,false,,,,,gpt-realtime-translate,,DOCUMENTED,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.update,session.audio.output.language,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,Target language for translated output audio and transcript deltas.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.update""}"
openai,WS /v1/realtime/translations session.input_audio_buffer.append,event_id,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.input_audio_buffer.append""}"
openai,WS /v1/realtime/translations session.input_audio_buffer.append,type,body,string,true,,,,session.input_audio_buffer.append,gpt-realtime-translate,,DOCUMENTED,"The event type, must be `session.input_audio_buffer.append`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.input_audio_buffer.append""}"
openai,WS /v1/realtime/translations session.input_audio_buffer.append,audio,body,string,true,,,,,gpt-realtime-translate,,DOCUMENTED,Base64-encoded 24 kHz PCM16 mono audio bytes.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.input_audio_buffer.append""}"
openai,WS /v1/realtime/translations session.close,event_id,body,string,false,,,,,gpt-realtime-translate,,DOCUMENTED,Optional client-generated ID used to identify this event.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.close""}"
openai,WS /v1/realtime/translations session.close,type,body,string,true,,,,session.close,gpt-realtime-translate,,DOCUMENTED,"The event type, must be `session.close`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/resources/realtime/translation-client-events#session.close""}"
openai,POST /v1/realtime/sessions,client_secret,body,object,true,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Ephemeral key returned by the API.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,client_secret.value,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Ephemeral key usable in client environments to authenticate connections
to the Realtime API. Use this in client-side environments rather than
a standard API token, which should only be used server-side.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,client_secret.expires_at,body,integer,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Timestamp for when the token expires. Currently, all tokens expire
after one minute.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,modalities,body,any,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The set of modalities the model can respond with. To disable audio,
set this to [""text""].","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,instructions,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. ""be extremely succinct"", ""act friendly"", ""here are examples of good responses"") and on audio behavior (e.g. ""talk quickly"", ""inject emotion into your voice"", ""laugh frequently""). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior.
Note that the server sets default instructions which will be used if this field is not set and are visible in the `session.created` event at the start of the session.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,voice,body,VoiceIdsShared | object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,A built-in voice name or a custom voice reference.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,voice.id,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The custom voice ID, e.g. `voice_1234`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,input_audio_format,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,output_audio_format,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,input_audio_transcription,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Configuration for input audio transcription, defaults to off and can be
set to `null` to turn off once on. Input audio transcription is not native
to the model, since the model consumes audio directly. Transcription runs
asynchronously and should be treated as rough guidance
rather than the representation understood by the model.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,input_audio_transcription.model,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,The model to use for transcription.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,speed,body,number,false,1,0.25,1.5,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The speed of the model's spoken response. 1.0 is the default speed. 0.25 is
the minimum speed. 1.5 is the maximum speed. This value can only be changed
in between model turns, not while a response is in progress.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tracing,body,enum | object,false,,,,auto,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Configuration options for tracing. Set to null to disable tracing. Once
tracing is enabled for a session, the configuration cannot be modified.

`auto` will create a trace for the session with default values for the
workflow name, group id, and metadata.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tracing.workflow_name,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The name of the workflow to attach to this trace. This is used to
name the trace in the traces dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tracing.group_id,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The group id to attach to this trace to enable filtering and
grouping in the traces dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tracing.metadata,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The arbitrary metadata to attach to this trace to enable
filtering in the traces dashboard.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,turn_detection,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Configuration for turn detection. Can be set to `null` to turn off. Server
VAD means that the model will detect the start and end of speech based on
audio volume and respond at the end of user speech.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,turn_detection.type,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Type of turn detection, only `server_vad` is currently supported.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,turn_detection.threshold,body,number,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,turn_detection.prefix_padding_ms,body,integer,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,turn_detection.silence_duration_ms,body,integer,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tools,body,array<object>,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Tools (functions) available to the model.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tools[].type,body,string,false,,,,function,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The type of the tool, i.e. `function`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tools[].name,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,The name of the function.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tools[].description,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The description of the function, including guidance on when and how
to call it, and guidance about what to tell the user when calling
(if anything).","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tools[].parameters,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Parameters of the function in JSON Schema.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,tool_choice,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"How the model chooses tools. Options are `auto`, `none`, `required`, or
specify a function.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,temperature,body,number,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Sampling temperature for the model, limited to [0.6, 1.2]. Defaults to 0.8.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,max_response_output_tokens,body,integer | enum,false,,,,inf,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Maximum number of output tokens for a single assistant response,
inclusive of tool calls. Provide an integer between 1 and 4096 to
limit output tokens, or `inf` for the maximum available tokens for a
given model. Defaults to `inf`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,truncation,body,enum | object,false,,,,auto|disabled,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs.

Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost.

Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate.

Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,truncation.type,body,string,false,,,,retention_ratio,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Use retention ratio truncation.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,truncation.retention_ratio,body,number,false,,0,1,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Fraction of post-instruction conversation tokens to retain (`0.0` - `1.0`) when the conversation exceeds the input token limit. Setting this to `0.8` means that messages will be dropped until 80% of the maximum allowed tokens are used. This helps reduce the frequency of truncations and improve cache rates.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,truncation.token_limits,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Optional custom token limits for this truncation strategy. If not provided, the model's default token limits will be used.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,truncation.token_limits.post_instructions,body,integer,false,,0,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Maximum tokens allowed in the conversation after instructions (which including tool definitions). For example, setting this to 5,000 would mean that truncation would occur when the conversation exceeds 5,000 tokens after instructions. This cannot be higher than the model's context window size minus the maximum output tokens.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,prompt,body,object | null,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,prompt.id,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,The unique identifier of the prompt template to use.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,prompt.version,body,string | null,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/sessions,prompt.variables,body,object | null,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,turn_detection,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Configuration for turn detection. Can be set to `null` to turn off. Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,turn_detection.type,body,string,false,,,,server_vad,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Type of turn detection. Only `server_vad` is currently supported for transcription sessions.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,turn_detection.threshold,body,number,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
higher threshold will require louder audio to activate the model, and
thus might perform better in noisy environments.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,turn_detection.prefix_padding_ms,body,integer,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Amount of audio to include before the VAD detected speech (in
milliseconds). Defaults to 300ms.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,turn_detection.silence_duration_ms,body,integer,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Duration of silence to detect speech stop (in milliseconds). Defaults
to 500ms. With shorter values the model will respond more quickly,
but may jump in on short pauses from the user.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_noise_reduction,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Configuration for input audio noise reduction. This can be set to `null` to turn off.
Noise reduction filters audio added to the input audio buffer before it is sent to VAD and the model.
Filtering the audio can improve VAD and turn detection accuracy (reducing false positives) and model performance by improving perception of the input audio.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_noise_reduction.type,body,string,false,,,,near_field|far_field,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Type of noise reduction. `near_field` is for close-talking microphones such as headphones, `far_field` is for far-field microphones such as laptop or conference room microphones.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_format,body,string,false,pcm16,,,pcm16|g711_ulaw|g711_alaw,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
For `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate,
single channel (mono), and little-endian byte order.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription,body,object,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Configuration for input audio transcription. The client can optionally set the language and prompt for transcription, these offer additional guidance to the transcription service.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.model,body,string | enum,false,,,,whisper-1|gpt-transcribe|gpt-live-transcribe|gpt-4o-mini-transcribe|gpt-4o-mini-transcribe-2025-12-15|gpt-4o-transcribe|gpt-4o-transcribe-diarize|gpt-realtime-whisper,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The model to use for transcription. Current options are `whisper-1`, `gpt-transcribe`, `gpt-live-transcribe`, `gpt-4o-mini-transcribe`, `gpt-4o-mini-transcribe-2025-12-15`, `gpt-4o-transcribe`, `gpt-4o-transcribe-diarize`, and `gpt-realtime-whisper`. Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.language,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The language of the input audio. Supplying the input language in
[ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format
will improve accuracy and latency.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.languages,body,array<string>,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.keywords,body,array<string>,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.prompt,body,string,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"An optional text to guide the model's style or continue a previous audio
segment.
For `whisper-1`, the [prompt is a list of keywords](https://developers.openai.com/api/docs/guides/speech-to-text#prompting).
For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the prompt is a free text string, for example ""expect words related to technology"".
Prompt is not supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,input_audio_transcription.delay,body,string,false,,,,minimal|low|medium|high|xhigh,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"Controls how long the model waits before emitting transcription text.
Higher values can improve transcription accuracy at the cost of latency.
Only supported with `gpt-realtime-whisper` in GA Realtime sessions.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,POST /v1/realtime/transcription_sessions,include,body,array<string>,false,,,,,,,DOCUMENTED|LEGACY|BETA|FAILED_VERIFICATION,"The set of items to include in the transcription. Current available items are:
`item.input_audio_transcription.logprobs`","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/reference/realtime-beta/overview""}"
openai,WS /v1/realtime (connect),model,query,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Realtime model to use when opening a fresh WebSocket session with a standard API key (`wss://api.openai.com/v1/realtime?model=…`). Ignored when `call_id` is given.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/voice-websockets?api=realtime""}"
openai,WS /v1/realtime (connect),call_id,query,string,false,,,,,,,DOCUMENTED,Attach a sideband WebSocket to an existing WebRTC/SIP call (`wss://api.openai.com/v1/realtime?call_id=rtc_…`). The model is already configured by the call/accept step.,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/voice-websockets?api=realtime""}"
openai,WS /v1/realtime (connect),Authorization,header,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"`Bearer <standard API key>` (server) or `Bearer <ek_… client secret>` (browser/mobile). In browsers without header support, use WebSocket subprotocols `realtime`, `openai-insecure-api-key.<ek_…>`, optional `openai-organization.<org>`, `openai-project.<proj>`.","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/voice-websockets?api=realtime""}"
openai,WS /v1/realtime (connect),OpenAI-Safety-Identifier,header,string,false,,,,,,,DOCUMENTED,"Stable, privacy-preserving end-user identifier (hashed). Set on the server-side connection request or on the client_secrets request (bound to the ephemeral token).","{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/voice-websockets?api=realtime""}"
openai,WS /v1/realtime (connect),OpenAI-Beta,header,string,false,,,,realtime=v1,,realtime=v1,DOCUMENTED,BETA-ERA header only. Remove for the GA interface (docs: 'Remove the OpenAI-Beta: realtime=v1 header when calling the GA interface').,"{""retrieved_at"": ""2026-09-19T01:31:52Z"", ""url"": ""https://developers.openai.com/api/docs/guides/voice-websockets?api=realtime""}"
openai,POST /v1/responses,metadata,body,"map<string,string>",false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard. Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,top_logprobs,body,integer,false,,0,20,,*,,DOCUMENTED|LIVE_VERIFIED,"An integer between 0 and 20 specifying the maximum number of most likely tokens to return at each token position, each with an associated log probability. In some cases, the number of returned tokens may be fewer than requested.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED|LIVE_VERIFIED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,top_p,body,number | null,false,1,0,1,,*,,DOCUMENTED|LIVE_VERIFIED,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,user,body,string,false,,,,,*,,DOCUMENTED|LEGACY,This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use `prompt_cache_key` instead to maintain caching optimizations. A stable identifier for your end-users. Used to boost cache hit rates by better bucketing similar requests and to help OpenAI detect and prevent abuse. [Learn more](https://developers.openai.com/api/docs/guides/safety-best-practices#implement-safety-identifiers).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,safety_identifier,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A stable identifier used to help detect users of your application that may be violating OpenAI's usage policies. The IDs should be a string that uniquely identifies each user, with a maximum length of 64 characters. We recommend hashing their username or email address, in order to avoid sending us any identifying information. [Learn more](https://developers.openai.com/api/docs/guides/safety-best-practices#implement-safety-identifiers).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_key,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Used by OpenAI to cache responses for similar requests to optimize your cache hit rates. Replaces the `user` field. [Learn more](https://developers.openai.com/api/docs/guides/prompt-caching).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_retention,body,string | null,false,,,,in_memory|24h,gpt-5.5|gpt-5.5-pro|gpt-5.4|gpt-5.2|gpt-5.1-codex-max|gpt-5.1|gpt-5.1-codex|gpt-5.1-codex-mini|gpt-5.1-chat-latest|gpt-5|gpt-5-codex|gpt-4.1|(in_memory: all cache-capable models),,DOCUMENTED|LEGACY|LIVE_VERIFIED,"Deprecated. Use `prompt_cache_options.ttl` instead. The retention policy for the prompt cache. Set to `24h` to enable extended prompt caching, which keeps cached prefixes active for longer, up to a maximum of 24 hours. [Learn more](https://developers.openai.com/api/docs/guides/prompt-caching#prompt-cache-retention). This field expresses a maximum retention policy, while `prompt_cache_options.ttl` expresses a minimum cache lifetime. The two fields are independent and do not interact. For `gpt-5.5`, `gpt-5.5-pro`, and future models, only `24h` is supported. For older models that support both `in_memory` and `24h`, the default depends on your organization's data retention policy: - Organizations without ZDR enabled default to `24h`. - Organizations with ZDR enabled default to `in_memory` when `prompt_cache_retention` is not specified.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_options,body,object,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Options for prompt caching. Supported for `gpt-5.6` and later models. By default, OpenAI automatically chooses one implicit cache breakpoint. You can add explicit breakpoints to content blocks with `prompt_cache_breakpoint`. Each request can write up to four breakpoints. For cache matching, OpenAI considers up to the latest 80 breakpoints in the conversation, without a content-block lookback limit. Set `mode` to `explicit` to disable the implicit breakpoint. The `ttl` defaults to `30m`, which is currently the only supported value. See the [prompt caching guide](https://developers.openai.com/api/docs/guides/prompt-caching) for current details.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_options.ttl,body,string,false,,,,30m,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"The minimum lifetime applied to every implicit and explicit cache breakpoint written by the request. Defaults to `30m`, which is currently the only supported value. The backend may retain cache entries for longer.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_options.mode,body,string,false,,,,implicit|explicit,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Controls whether OpenAI automatically creates an implicit cache breakpoint. Defaults to `implicit`. With `implicit`, OpenAI creates one implicit breakpoint and writes up to the latest three explicit breakpoints in the request. With `explicit`, OpenAI does not create an implicit breakpoint and writes up to the latest four explicit breakpoints. If there are no explicit breakpoints, the request does not use prompt caching.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_options.prewarm,body,boolean,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Prepares the prompt cache without generating output. Defaults to `false`. When set to `true`, overrides the `generate` field to `false`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt_cache_options.comparison_response_id,body,string | null,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,The ID of a response to compare when diagnosing prompt cache reuse. Supplying this field requests prompt cache diagnostics when the feature is enabled.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,previous_response_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the previous response to the model. Use this to create multi-turn conversations. Learn more about [conversation state](https://developers.openai.com/api/docs/guides/conversation-state). Cannot be used in conjunction with `conversation`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,model,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Model ID used to generate the response, like `gpt-6-astra`. OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points. Refer to the [model guide](https://developers.openai.com/api/docs/models) to browse and compare available models.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,background,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,Whether to run the model response in the background. [Learn more](https://developers.openai.com/api/docs/guides/background).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,max_tool_calls,body,integer | null,false,,,,,*,,DOCUMENTED,"The maximum number of total calls to built-in tools that can be processed in a response. This maximum number applies across all built-in tool calls, not per individual tool. Any further attempts to call a tool by the model will be ignored.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text,body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Configuration options for a text response from the model. Can be plain text or structured JSON data. Learn more: - [Text inputs and outputs](https://developers.openai.com/api/docs/guides/text) - [Structured Outputs](https://developers.openai.com/api/docs/guides/structured-outputs),https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format,body,object<text> | object<json_schema> | object<json_object>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An object specifying the format that the model must output. Configuring `{ ""type"": ""json_schema"" }` enables Structured Outputs, which ensures the model will match your supplied JSON schema. Learn more in the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs). The default format is `{ ""type"": ""text"" }` with no additional options. **Not recommended for gpt-4o and newer models:** Setting to `{ ""type"": ""json_object"" }` enables the older JSON mode, which ensures the message the model generates is valid JSON. Using `json_schema` is preferred for models that support it.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(text),body,object,false,,,,,*,,DOCUMENTED,Default response format. Used to generate text responses.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(text).type,body,string,true,,,,text,*,,DOCUMENTED,The type of response format being defined. Always `text`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,JSON Schema response format. Used to generate structured JSON responses. Learn more about [Structured Outputs](https://developers.openai.com/api/docs/guides/structured-outputs).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema).type,body,string,true,,,,json_schema,*,,DOCUMENTED|LIVE_VERIFIED,The type of response format being defined. Always `json_schema`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema).description,body,string,false,,,,,*,,DOCUMENTED,"A description of what the response format is for, used by the model to determine how to respond in the format.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema).name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema).schema,body,"map<string,any>",true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The schema for the response format, described as a JSON Schema object. Learn how to build JSON schemas [here](https://json-schema.org/).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_schema).strict,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the `schema` field. Only a subset of JSON Schema is supported when `strict` is `true`. To learn more, read the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_object),body,object,false,,,,,*,,DOCUMENTED,JSON object response format. An older method of generating JSON responses. Using `json_schema` is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.format(json_object).type,body,string,true,,,,json_object,*,,DOCUMENTED,The type of response format being defined. Always `json_object`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,text.verbosity,body,string | null,false,medium,,,low|medium|high,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4*|gpt-5.2*|gpt-5.1*|gpt-5*,,DOCUMENTED|LIVE_VERIFIED,"Constrains the verbosity of the model's response. Lower values will result in more concise responses, while higher values will result in more verbose responses. Currently supported values are `low`, `medium`, and `high`. The default is `medium`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,false,,,,,*,,DOCUMENTED,"An array of tools the model may call while generating a response. You can specify which tool to use by setting the `tool_choice` parameter. We support the following categories of tools: - **Built-in tools**: Tools that are provided by OpenAI that extend the model's capabilities, like [web search](https://developers.openai.com/api/docs/guides/tools-web-search) or [file search](https://developers.openai.com/api/docs/guides/tools-file-search). Learn more about [built-in tools](https://developers.openai.com/api/docs/guides/tools). - **MCP Tools**: Integrations with third-party systems via custom MCP servers or predefined connectors such as Google Drive and SharePoint. Learn more about [MCP Tools](https://developers.openai.com/api/docs/guides/tools-connectors-mcp). - **Function calls (custom tools)**: Functions that are defined by you, enabling the model to call your own code with strongly ty",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](function),body,object,false,,,,,*,,DOCUMENTED,Defines a function in your own code the model can choose to call. Learn more about [function calling](https://developers.openai.com/api/docs/guides/function-calling).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](file_search),body,object,false,,,,,*,,DOCUMENTED,A tool that searches for relevant content from uploaded files. Learn more about the [file search tool](https://developers.openai.com/api/docs/guides/tools-file-search).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](computer),body,object,false,,,,,*,,DOCUMENTED,A tool that controls a virtual computer. Learn more about the [computer tool](https://developers.openai.com/api/docs/guides/tools-computer-use).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](computer_use_preview),body,object,false,,,,,*,,DOCUMENTED,A tool that controls a virtual computer. Learn more about the [computer tool](https://developers.openai.com/api/docs/guides/tools-computer-use).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](web_search),body,object,false,,,,,*,,DOCUMENTED,Search the Internet for sources related to the prompt. Learn more about the [web search tool](https://developers.openai.com/api/docs/guides/tools-web-search).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](mcp),body,object,false,,,,,*,,DOCUMENTED,Give the model access to additional tools via remote Model Context Protocol (MCP) servers. [Learn more about MCP](https://developers.openai.com/api/docs/guides/tools-connectors-mcp).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](code_interpreter),body,object,false,,,,,*,,DOCUMENTED,A tool that runs Python code to help generate a response to a prompt.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](programmatic_tool_calling),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](image_generation),body,object,false,,,,,*,,DOCUMENTED,A tool that generates images using the GPT image models.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](local_shell),body,object,false,,,,,*,,DOCUMENTED,A tool that allows the model to execute shell commands in a local environment.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](shell),body,object,false,,,,,*,,DOCUMENTED,A tool that allows the model to execute shell commands.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](custom),body,object,false,,,,,*,,DOCUMENTED,A custom tool that processes input using a specified format. Learn more about [custom tools](https://developers.openai.com/api/docs/guides/function-calling#custom-tools),https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](namespace),body,object,false,,,,,*,,DOCUMENTED,Groups function/custom tools under a shared namespace.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](tool_search),body,object,false,,,,,*,,DOCUMENTED,Hosted or BYOT tool search configuration for deferred tools.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](web_search_preview),body,object,false,,,,,*,,DOCUMENTED,This tool searches the web for relevant results to use in a response. Learn more about the [web search tool](https://developers.openai.com/api/docs/guides/tools-web-search).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tools[](apply_patch),body,object,false,,,,,*,,DOCUMENTED,"Allows the assistant to create, delete, or update files using unified diffs.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice,body,string | object<allowed_tools> | ToolChoiceTypes | object<function> | object<mcp> | object<custom> | object<programmatic_tool_calling> | object<apply_patch> | object<shell>,false,,,,,*,,DOCUMENTED,How the model should select which tool (or tools) to use when generating a response. See the `tools` parameter to see how to specify which tools the model can call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(allowed_tools),body,object,false,,,,,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(allowed_tools).type,body,string,true,,,,allowed_tools,*,,DOCUMENTED,Allowed tool configuration type. Always `allowed_tools`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(allowed_tools).mode,body,string,true,,,,auto|required,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set. `auto` allows the model to pick from among the allowed tools and generate a message. `required` requires the model to call one or more of the allowed tools.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(allowed_tools).tools,body,"array<map<string,any>>",true,,,,,*,,DOCUMENTED,"A list of tool definitions that the model should be allowed to call. For the Responses API, the list of tool definitions might look like: ```json [ { ""type"": ""function"", ""name"": ""get_weather"" }, { ""type"": ""mcp"", ""server_label"": ""deepwiki"" }, { ""type"": ""image_generation"" } ] ```",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(allowed_tools).tools[],body,"map<string,any>",false,,,,,*,,DOCUMENTED,A tool definition that the model should be allowed to call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(hosted_tool),body,object,false,,,,,*,,DOCUMENTED,Indicates that the model should use a built-in tool to generate a response. [Learn more about built-in tools](https://developers.openai.com/api/docs/guides/tools).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(hosted_tool).type,body,string,true,,,,file_search|web_search_preview|computer|computer_use_preview|computer_use|web_search_preview_2025_03_11|image_generation|code_interpreter,*,,DOCUMENTED,The type of hosted tool the model should to use. Learn more about [built-in tools](https://developers.openai.com/api/docs/guides/tools). Allowed values are: - `file_search` - `web_search_preview` - `computer` - `computer_use_preview` - `computer_use` - `code_interpreter` - `image_generation`,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(function),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific function.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(function).type,body,string,true,,,,function,*,,DOCUMENTED,"For function calling, the type is always `function`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(function).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(mcp),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific tool on a remote MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(mcp).type,body,string,true,,,,mcp,*,,DOCUMENTED,"For MCP tools, the type is always `mcp`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(mcp).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server to use.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(mcp).name,body,string | null,false,,,,,*,,DOCUMENTED,The name of the tool to call on the server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(custom),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific custom tool.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(custom).type,body,string,true,,,,custom,*,,DOCUMENTED,"For custom tool calling, the type is always `custom`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(custom).name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool to call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(programmatic_tool_calling),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(programmatic_tool_calling).type,body,string,true,programmatic_tool_calling,,,programmatic_tool_calling,*,,DOCUMENTED,The tool to call. Always `programmatic_tool_calling`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(apply_patch),body,object,false,,,,,*,,DOCUMENTED,Forces the model to call the apply_patch tool when executing a tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(apply_patch).type,body,string,true,apply_patch,,,apply_patch,*,,DOCUMENTED,The tool to call. Always `apply_patch`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(shell),body,object,false,,,,,*,,DOCUMENTED,Forces the model to call the shell tool when a tool call is required.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,tool_choice(shell).type,body,string,true,shell,,,shell,*,,DOCUMENTED,The tool to call. Always `shell`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt,body,object | null,false,,,,,*,,DOCUMENTED,Reference to a prompt template and its variables. [Learn more](https://developers.openai.com/api/docs/guides/text?api-mode=responses#version-prompts-in-code).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt.id,body,string,true,,,,,*,,DOCUMENTED,The unique identifier of the prompt template to use.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt.version,body,string | null,false,,,,,*,,DOCUMENTED,Optional version of the prompt template.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,prompt.variables,body,"map<string,string | object<input_text> | object<input_image> | object<input_file>>",false,,,,,*,,DOCUMENTED,"Optional map of values to substitute in for variables in your prompt. The substitution values can either be strings, or other Response input types like images or files.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,service_tier,body,string | null,false,auto,,,auto|default|flex|scale|priority|fast|ultrafast,*,,DOCUMENTED|LIVE_VERIFIED,"Specifies the processing type used for serving the request. - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'. - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model. - If set to '[flex](https://developers.openai.com/api/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier. - To opt-in to [Fast mode](https://developers.openai.com/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. The response will show `service_tier=priority` regardless of if you specify `service_tier=fast` or `priority` in your request. - If set to 'ultrafast', then the request will",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,truncation,body,string | null,false,disabled,,,auto|disabled,*,,DOCUMENTED|LEGACY|LIVE_VERIFIED,"The truncation strategy to use for the model response. - `auto`: If the input to this Response exceeds the model's context window size, the model will truncate the response to fit the context window by dropping items from the beginning of the conversation. - `disabled` (default): If the input size will exceed the context window size for a model, the request will fail with a 400 error.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning,body,object | null,false,,,,,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Configuration options for [reasoning models](https://developers.openai.com/api/docs/guides/reasoning).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning.mode,body,string,false,,,,,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"Controls the reasoning execution mode for the request. When returned on a response, this is the effective execution mode.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning.effort,body,string | null,false,medium,,,none|minimal|low|medium|high|xhigh|max,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"Constrains effort on reasoning for reasoning models. Currently supported values are `none`, `minimal`, `low`, `medium`, `high`, `xhigh`, and `max`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Not all reasoning models support every value. See the [reasoning guide](https://developers.openai.com/api/docs/guides/reasoning) for model-specific support.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning.summary,body,string | null,false,,,,auto|concise|detailed,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"A summary of the reasoning performed by the model. This can be useful for debugging and understanding the model's reasoning process. One of `auto`, `concise`, or `detailed`. `concise` is supported for `computer-use-preview` models and all reasoning models after `gpt-5`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning.context,body,string | null,false,,,,auto|current_turn|all_turns,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"Controls which reasoning items are rendered back to the model on later turns. If omitted or set to `auto`, the model determines the context mode. The `gpt-5.6` model family defaults to `all_turns`; earlier models default to `current_turn`. When returned on a response, this is the effective reasoning context mode used for the response.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,reasoning.generate_summary,body,string | null,false,,,,auto|concise|detailed,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LEGACY,"**Deprecated:** use `summary` instead. A summary of the reasoning performed by the model. This can be useful for debugging and understanding the model's reasoning process. One of `auto`, `concise`, or `detailed`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input,body,string | array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or file inputs to the model, used to generate a response. Learn more: - [Text inputs and outputs](https://developers.openai.com/api/docs/guides/text) - [Image inputs](https://developers.openai.com/api/docs/guides/images-vision) - [File inputs](https://developers.openai.com/api/docs/guides/file-inputs) - [Conversation state](https://developers.openai.com/api/docs/guides/conversation-state) - [Function calling](https://developers.openai.com/api/docs/guides/function-calling)",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[],body,array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>> | object<program_output>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A message input to the model with a role indicating instruction following hierarchy. Instructions given with the `developer` or `system` role take precedence over instructions given with the `user` role. Messages with the `assistant` role are presumed to have been generated by the model in previous interactions.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).role,body,string,true,,,,user|assistant|system|developer,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the message input. One of `user`, `assistant`, `system`, or `developer`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content,body,string | array<object<input_text> | object<input_image> | object<input_file>> | array<object<output_text> | object<refusal>>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or audio input to the model, used to generate a response. Can also contain previous assistant responses.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[],body,array<object<input_text> | object<input_image> | object<input_file>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).phase,body,string | null,false,,,,commentary|final_answer,*,,DOCUMENTED|LIVE_VERIFIED,"Labels an `assistant` message as intermediate commentary (`commentary`) or the final answer (`final_answer`). For models like `gpt-5.3-codex` and beyond, when sending follow-up requests, preserve and resend phase on all assistant messages — dropping it can degrade performance. Not used for user messages.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).type,body,string,true,,,,message,*,,DOCUMENTED|LIVE_VERIFIED,The type of the message input. Always `message`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the output message.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a file search tool call. See the [file search guide](https://developers.openai.com/api/docs/guides/tools-file-search) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the file search tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call).type,body,string,true,,,,file_search_call,*,,DOCUMENTED,The type of the file search tool call. Always `file_search_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call).status,body,string,true,,,,in_progress|searching|completed|incomplete|failed,*,,DOCUMENTED,"The status of the file search tool call. One of `in_progress`, `searching`, `incomplete` or `failed`,",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call).queries,body,array<string>,true,,,,,*,,DOCUMENTED,The queries used to search for files.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](file_search_call).results,body,array<object>,false,,,,,*,,DOCUMENTED,The results of the file search tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to a computer use tool. See the [computer use guide](https://developers.openai.com/api/docs/guides/tools-computer-use) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).type,body,string,true,computer_call,,,computer_call,*,,DOCUMENTED,The type of the computer call. Always `computer_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the computer call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used when responding to the tool call with output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).action,body,object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).actions,body,array<object<click> | object<double_click> | object<drag> | object<keypress> | object<move> | object<screenshot> | object<scroll> | object<type> | object<wait>>,false,,,,,*,,DOCUMENTED,Flattened batched actions for `computer_use`. Each action includes an `type` discriminator and action-specific fields.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).pending_safety_checks,body,array<object>,true,,,,,*,,DOCUMENTED,The pending safety checks for the computer call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a computer tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the computer tool call output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the computer tool call that produced the output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).type,body,string,true,computer_call_output,,,computer_call_output,*,,DOCUMENTED,The type of the computer tool call output. Always `computer_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).output,body,object,true,,,,,*,,DOCUMENTED,A computer screenshot image used with the computer use tool.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).acknowledged_safety_checks,body,array<object>,false,,,,,*,,DOCUMENTED,The safety checks reported by the API that have been acknowledged by the developer.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](computer_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the message input. One of `in_progress`, `completed`, or `incomplete`. Populated when input items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](web_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a web search tool call. See the [web search guide](https://developers.openai.com/api/docs/guides/tools-web-search) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](web_search_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the web search tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](web_search_call).type,body,string,true,,,,web_search_call,*,,DOCUMENTED,The type of the web search tool call. Always `web_search_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](web_search_call).status,body,string,true,,,,in_progress|searching|completed|failed|incomplete,*,,DOCUMENTED,The status of the web search tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](web_search_call).action,body,object<search> | object<open_page> | object<find_in_page>,true,,,,,*,,DOCUMENTED,"An object describing the specific action taken in this web search call. Includes details on how the model used the web (search, open_page, find_in_page).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a function. See the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).type,body,string,true,,,,function_call,*,,DOCUMENTED,The type of the function tool call. Always `function_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the function to run.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the function to run.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments to pass to the function.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the function tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a function tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).type,body,string,true,function_call_output,,,function_call_output,*,,DOCUMENTED,The type of the function tool call output. Always `function_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,"Text, image, or file output of the function tool call.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).name,body,string | null,false,,,,,*,,DOCUMENTED,The name of the tool that produced the output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).namespace,body,string | null,false,,,,,*,,DOCUMENTED,The namespace of the tool that produced the output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).type,body,string,true,tool_search_call,,,tool_search_call,*,,DOCUMENTED,The item type. Always `tool_search_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).arguments,body,object,true,,,,,*,,DOCUMENTED,The arguments supplied to the tool search call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this tool search output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).call_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the tool search call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).type,body,string,true,tool_search_output,,,tool_search_output,*,,DOCUMENTED,The item type. Always `tool_search_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).execution,body,string,false,,,,server|client,*,,DOCUMENTED,Whether tool search was executed by the server or by the client.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,The loaded tool definitions returned by the tool search output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](tool_search_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the tool search output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](additional_tools),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](additional_tools).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this additional tools item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](additional_tools).type,body,string,true,additional_tools,,,additional_tools,*,,DOCUMENTED,The item type. Always `additional_tools`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](additional_tools).role,body,string,true,developer,,,developer,*,,DOCUMENTED,The role that provided the additional tools. Only `developer` is supported.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](additional_tools).tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,true,,,,,*,,DOCUMENTED,A list of additional tools made available at this item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](configuration_update),body,object,false,,,,,gpt-6-astra,,DOCUMENTED,An update to the conversation's response configuration. The configuration remains in effect for subsequent responses until it is replaced by another configuration update.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](configuration_update).id,body,string | null,false,,,,,gpt-6-astra,,DOCUMENTED,The unique ID of the configuration update item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](configuration_update).type,body,string,true,configuration_update,,,configuration_update,gpt-6-astra,,DOCUMENTED,The item type. Always `configuration_update`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](configuration_update).reasoning,body,object,false,,,,,gpt-6-astra,,DOCUMENTED,Updates to reasoning configuration. Only effort is supported.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning),body,object,false,,,,,*,,DOCUMENTED,A description of the chain of thought used by a reasoning model while generating a response. Be sure to include these items in your `input` to the Responses API for subsequent turns of a conversation if you are manually [managing context](https://developers.openai.com/api/docs/guides/conversation-state).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).type,body,string,true,,,,reasoning,*,,DOCUMENTED,The type of the object. Always `reasoning`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).id,body,string,true,,,,,*,,DOCUMENTED,The unique identifier of the reasoning content.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).encrypted_content,body,string | null,false,,,,,*,,DOCUMENTED,"The encrypted content of the reasoning item. This is populated by default for reasoning items returned by `POST /v1/responses` and WebSocket `response.create` requests. When streaming, use the completed reasoning item and its `encrypted_content` from the `response.output_item.done` event in subsequent requests. The `encrypted_content` in `response.output_item.added` may be incomplete. This is especially important when `store` is `false` or when using Zero Data Retention.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).summary,body,array<object>,true,,,,,*,,DOCUMENTED,Reasoning summary content.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).content,body,array<object>,false,,,,,*,,DOCUMENTED,Reasoning text content.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](reasoning).status,body,string,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`. Populated when items are returned via API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction),body,object,false,,,,,* (responses models supporting compaction),,DOCUMENTED,A compaction item generated by the [`v1/responses/compact` API](https://developers.openai.com/api/reference/resources/responses/methods/compact).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction).id,body,string | null,false,,,,,* (responses models supporting compaction),,DOCUMENTED,The ID of the compaction item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction).type,body,string,true,compaction,,,compaction,* (responses models supporting compaction),,DOCUMENTED,The type of the item. Always `compaction`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction).encrypted_content,body,string,true,,,,,* (responses models supporting compaction),,DOCUMENTED,The encrypted content of the compaction summary.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call),body,object,false,,,,,*,,DOCUMENTED,An image generation request made by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).type,body,string,true,,,,image_generation_call,*,,DOCUMENTED,The type of the image generation call. Always `image_generation_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the image generation call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).status,body,string,true,,,,in_progress|completed|generating|failed,*,,DOCUMENTED,The status of the image generation call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).result,body,string | null,true,,,,,*,,DOCUMENTED,The generated image encoded in base64.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).size,body,string,false,,,,,*,,DOCUMENTED,"The image dimensions as a `WIDTHxHEIGHT` string, for example `1536x864`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).quality,body,string | null,false,,,,low|medium|high|xhigh|max|auto,*,,DOCUMENTED,"The quality of the image generated by the image generation tool call. One of `low`, `medium`, `high`, `xhigh`, `max`, or `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).action,body,string | null,false,,,,generate|edit|auto,*,,DOCUMENTED,The action used for image generation.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).background,body,string | null,false,,,,transparent|opaque|auto,*,,DOCUMENTED,The background setting used for generation.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).output_format,body,string | null,false,,,,png|webp|jpeg,*,,DOCUMENTED,The output format used for generation.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](image_generation_call).revised_prompt,body,string | null,false,,,,,*,,DOCUMENTED,The prompt that was used after any model prompt rewriting.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run code.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).type,body,string,true,code_interpreter_call,,,code_interpreter_call,*,,DOCUMENTED,The type of the code interpreter tool call. Always `code_interpreter_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the code interpreter tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).status,body,string,true,,,,in_progress|completed|incomplete|interpreting|failed,*,,DOCUMENTED,"The status of the code interpreter tool call. Valid values are `in_progress`, `completed`, `incomplete`, `interpreting`, and `failed`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).container_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the container used to run the code.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).code,body,string | null,true,,,,,*,,DOCUMENTED,"The code to run, or null if not available.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](code_interpreter_call).outputs,body,array<object<logs> | object<image>>,true,,,,,*,,DOCUMENTED,"The outputs generated by the code interpreter, such as logs or images. Can be null if no outputs are available.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a command on the local shell.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call).type,body,string,true,,,,local_shell_call,*,,DOCUMENTED,The type of the local shell call. Always `local_shell_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call).action,body,object,true,,,,,*,,DOCUMENTED,Execute a shell command on the server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call).status,body,string,true,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the local shell call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a local shell tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call_output).type,body,string,true,,,,local_shell_call_output,*,,DOCUMENTED,The type of the local shell tool call output. Always `local_shell_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the local shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call_output).output,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the output of the local shell tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](local_shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the item. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool representing a request to execute one or more shell commands.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).type,body,string,true,shell_call,,,shell_call,*,,DOCUMENTED,The type of the item. Always `shell_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).action,body,object,true,,,,,*,,DOCUMENTED,The shell commands and limits that describe how to run the tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,"The status of the shell call. One of `in_progress`, `completed`, or `incomplete`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call).environment,body,object<local> | object<container_reference>,false,,,,,*,,DOCUMENTED,The environment to execute the shell commands in.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output items emitted by a shell tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the shell tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).type,body,string,true,shell_call_output,,,shell_call_output,*,,DOCUMENTED,The type of the item. Always `shell_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).output,body,array<object>,true,,,,,*,,DOCUMENTED,"Captured chunks of stdout and stderr output, along with their associated outcomes.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).status,body,string | null,false,,,,in_progress|completed|incomplete,*,,DOCUMENTED,The status of the shell call output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](shell_call_output).max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,The maximum number of UTF-8 characters captured for this shell call's combined output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call),body,object,false,,,,,*,,DOCUMENTED,"A tool call representing a request to create, delete, or update files using diff patches.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).type,body,string,true,apply_patch_call,,,apply_patch_call,*,,DOCUMENTED,The type of the item. Always `apply_patch_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).status,body,string,true,,,,in_progress|completed,*,,DOCUMENTED,The status of the apply patch tool call. One of `in_progress` or `completed`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call).operation,body,object<create_file> | object<delete_file> | object<update_file>,true,,,,,*,,DOCUMENTED,"The specific create, delete, or update instruction for the apply_patch tool call.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output emitted by an apply patch tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).type,body,string,true,apply_patch_call_output,,,apply_patch_call_output,*,,DOCUMENTED,The type of the item. Always `apply_patch_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call output. Populated when this item is returned via API.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the apply patch tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).status,body,string,true,,,,completed|failed,*,,DOCUMENTED,The status of the apply patch tool call output. One of `completed` or `failed`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](apply_patch_call_output).output,body,string | null,false,,,,,*,,DOCUMENTED,"Optional human-readable log text from the apply patch tool (e.g., patch results or errors).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools),body,object,false,,,,,*,,DOCUMENTED,A list of tools available on an MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools).type,body,string,true,,,,mcp_list_tools,*,,DOCUMENTED,The type of the item. Always `mcp_list_tools`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the list.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools).tools,body,array<object>,true,,,,,*,,DOCUMENTED,The tools available on the server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_list_tools).error,body,string | null,false,,,,,*,,DOCUMENTED,Error message if the server could not list tools.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request),body,object,false,,,,,*,,DOCUMENTED,A request for human approval of a tool invocation.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request).type,body,string,true,,,,mcp_approval_request,*,,DOCUMENTED,The type of the item. Always `mcp_approval_request`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the approval request.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server making the request.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool to run.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_request).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of arguments for the tool.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response),body,object,false,,,,,*,,DOCUMENTED,A response to an MCP approval request.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response).type,body,string,true,,,,mcp_approval_response,*,,DOCUMENTED,The type of the item. Always `mcp_approval_response`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the approval response,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response).approval_request_id,body,string,true,,,,,*,,DOCUMENTED,The ID of the approval request being answered.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response).approve,body,boolean,true,,,,,*,,DOCUMENTED,Whether the request was approved.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_approval_response).reason,body,string | null,false,,,,,*,,DOCUMENTED,Optional reason for the decision.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call),body,object,false,,,,,*,,DOCUMENTED,An invocation of a tool on an MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).type,body,string,true,,,,mcp_call,*,,DOCUMENTED,The type of the item. Always `mcp_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server running the tool.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool that was run.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments passed to the tool.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).output,body,string | null,false,,,,,*,,DOCUMENTED,The output from the tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).error,body,object<mcp_protocol_error> | object<mcp_tool_execution_error> | object<http_error>,false,,,,,*,,DOCUMENTED,"The error from the tool call, if any.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).status,body,string,false,,,,in_progress|completed|incomplete|calling|failed,*,,DOCUMENTED,"The status of the tool call. One of `in_progress`, `completed`, `incomplete`, `calling`, or `failed`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](mcp_call).approval_request_id,body,string | null,false,,,,,*,,DOCUMENTED,Unique identifier for the MCP tool call approval request. Include this value in a subsequent `mcp_approval_response` input to approve or reject the corresponding tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output),body,object,false,,,,,*,,DOCUMENTED,"The output of a custom tool call from your code, being sent back to the model.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).type,body,string,true,,,,custom_tool_call_output,*,,DOCUMENTED,The type of the custom tool call output. Always `custom_tool_call_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call output in the OpenAI platform.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).call_id,body,string,true,,,,,*,,DOCUMENTED,"The call ID, used to map this custom tool call output to a custom tool call.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output,body,string | array<object<input_text> | object<input_image> | object<input_file>>,true,,,,,*,,DOCUMENTED,The output from the custom tool call generated by your code. Can be a string or an list of output content.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).type,body,string,true,,,,custom_tool_call,*,,DOCUMENTED,The type of the custom tool call. Always `custom_tool_call`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the custom tool call in the OpenAI platform.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).call_id,body,string,true,,,,,*,,DOCUMENTED,An identifier used to map this custom tool call to a tool call output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).caller,body,object<direct> | object<program>,false,,,,,*,,DOCUMENTED,The execution context that produced this tool call.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).namespace,body,string,false,,,,,*,,DOCUMENTED,The namespace of the custom tool being called.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).name,body,string,true,,,,,*,,DOCUMENTED,The name of the custom tool being called.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).input,body,string,true,,,,,*,,DOCUMENTED,The input for the custom tool call generated by the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call).async,body,boolean,false,,,,,*,,DOCUMENTED,Whether the custom tool call runs asynchronously.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction_trigger),body,object,false,,,,,*,,DOCUMENTED,Compacts the current context. Must be the final input item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction_trigger).id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of this compaction trigger.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](compaction_trigger).type,body,string,true,compaction_trigger,,,compaction_trigger,*,,DOCUMENTED,The type of the item. Always `compaction_trigger`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](item_reference),body,object,false,,,,,*,,DOCUMENTED,An internal identifier for an item to reference.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](item_reference).type,body,string | null,false,item_reference,,,item_reference,*,,DOCUMENTED,The type of item to reference. Always `item_reference`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](item_reference).id,body,string,true,,,,,*,,DOCUMENTED,The ID of the item to reference.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program).type,body,string,true,program,,,program,*,,DOCUMENTED,The item type. Always `program`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program).call_id,body,string,true,,,,,*,,DOCUMENTED,The stable call ID of the program item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program).code,body,string,true,,,,,*,,DOCUMENTED,The JavaScript source executed by programmatic tool calling.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program).fingerprint,body,string,true,,,,,*,,DOCUMENTED,Opaque program replay fingerprint that must be round-tripped.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of this program output item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output).type,body,string,true,program_output,,,program_output,*,,DOCUMENTED,The item type. Always `program_output`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output).call_id,body,string,true,,,,,*,,DOCUMENTED,The call ID of the program item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output).result,body,string,true,,,,,*,,DOCUMENTED,The result produced by the program item.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](program_output).status,body,string,true,,,,completed|incomplete,*,,DOCUMENTED,The terminal status of the program output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,include,body,array<string>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Specify additional output data to include in the model response. Currently supported values are: - `web_search_call.action.sources`: Include the sources of the web search tool call. - `code_interpreter_call.outputs`: Includes the outputs of python code execution in code interpreter tool call items. - `computer_call_output.output.image_url`: Include image urls from the computer call output. - `file_search_call.results`: Include the search results of the file search tool call. - `message.input_image.image_url`: Include image urls from the input message. - `message.output_text.logprobs`: Include logprobs with assistant messages. - `reasoning.encrypted_content`: Includes an encrypted version of reasoning tokens in reasoning item outputs. This enables reasoning items to be used in multi-turn conversations when using the Responses API statelessly (like when the `store` parameter is set to `fal,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,include[],body,string,false,,,,file_search_call.results|web_search_call.results|web_search_call.action.sources|message.input_image.image_url|computer_call_output.output.image_url|code_interpreter_call.outputs|reasoning.encrypted_content|message.output_text.logprobs,*,,DOCUMENTED|LIVE_VERIFIED,Specify additional output data to include in the model response. Currently supported values are: - `web_search_call.results`: Include the search results of the web search tool call. - `web_search_call.action.sources`: Include the sources of the web search tool call. - `code_interpreter_call.outputs`: Includes the outputs of python code execution in code interpreter tool call items. - `computer_call_output.output.image_url`: Include image urls from the computer call output. - `file_search_call.results`: Include the search results of the file search tool call. - `message.input_image.image_url`: Include image urls from the input message. - `message.output_text.logprobs`: Include logprobs with assistant messages. - `reasoning.encrypted_content`: Includes an encrypted version of reasoning tokens in reasoning item outputs. This enables reasoning items to be used in multi-turn conversations whe,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,parallel_tool_calls,body,boolean | null,false,true,,,,*,,DOCUMENTED|LIVE_VERIFIED,Whether to allow the model to run tool calls in parallel.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,store,body,boolean | null,false,true,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to store the generated model response for later retrieval via API. Defaults to true when omitted. If set to true, response data will be stored for at least 30 days, subject to the [data retention exceptions](https://developers.openai.com/api/docs/guides/your-data#v1responses).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,instructions,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A system (or developer) message inserted into the model's context. When using along with `previous_response_id`, the instructions from a previous response will not be carried over to the next response. This makes it simple to swap out system (or developer) messages in new responses.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation,body,object | null,false,,,,,*,,DOCUMENTED,Configuration for running moderation on the input and output of this response.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.model,body,string,true,,,,,*,,DOCUMENTED,"The moderation model to use for moderated completions, e.g. 'omni-moderation-latest'.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.policy,body,object | null,false,,,,,*,,DOCUMENTED,The policy to apply to moderated response input and output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.policy.input,body,object | null,false,,,,,*,,DOCUMENTED,The moderation policy for the response input.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.policy.input.mode,body,string,true,,,,score|block,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.policy.output,body,object | null,false,,,,,*,,DOCUMENTED,The moderation policy for the response output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,moderation.policy.output.mode,body,string,true,,,,score|block,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,stream,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set to true, the model response data will be streamed to the client as it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format). See the [Streaming section below](https://developers.openai.com/api/reference/resources/responses/streaming-events) for more information.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,stream_options,body,object | null,false,,,,,*,,DOCUMENTED,Options for streaming responses. Only set this when you set `stream: true`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,stream_options.include_obfuscation,body,boolean,false,,,,,*,,DOCUMENTED,"When true, stream obfuscation will be enabled. Stream obfuscation adds random characters to an `obfuscation` field on streaming delta events to normalize payload sizes as a mitigation to certain side-channel attacks. These obfuscation fields are included by default, but add a small amount of overhead to the data stream. You can set `include_obfuscation` to false to optimize for bandwidth if you trust the network links between your application and the OpenAI API.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,conversation,body,string | object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The conversation that this response belongs to. Items from this conversation are prepended to `input_items` for this response request. Input items and output items from this response are automatically added to this conversation after this response completes.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,conversation(object),body,object,false,,,,,*,,DOCUMENTED,The conversation that this response belongs to.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,conversation(object).id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the conversation.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,context_management,body,array<object>,false,,,,,"* (responses models supporting compaction, e.g. gpt-5.3-codex, gpt-5.4*, gpt-6-astra)",,DOCUMENTED,Context management configuration for this request.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,context_management[],body,object,false,,,,,"* (responses models supporting compaction, e.g. gpt-5.3-codex, gpt-5.4*, gpt-6-astra)",,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,context_management[].type,body,string,true,,,,,"* (responses models supporting compaction, e.g. gpt-5.3-codex, gpt-5.4*, gpt-6-astra)",,DOCUMENTED,The context management entry type. Currently only 'compaction' is supported.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,context_management[].compact_threshold,body,integer | null,false,,1000,,,"* (responses models supporting compaction, e.g. gpt-5.3-codex, gpt-5.4*, gpt-6-astra)",,DOCUMENTED,Token threshold at which compaction should be triggered for this entry.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,max_output_tokens,body,integer | null,false,,16,,,*,,DOCUMENTED|LIVE_VERIFIED,"An upper bound for the number of tokens that can be generated for a response, including visible output tokens and [reasoning tokens](https://developers.openai.com/api/docs/guides/reasoning).",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_text),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_text).type,body,string,true,input_text,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,The type of the input item. Always `input_text`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_text).text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image),body,object,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).type,body,string,true,input_image,,,input_image,*,,DOCUMENTED|LIVE_VERIFIED,The type of the input item. Always `input_image`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).image_url,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).detail,body,string,true,,,,low|high|auto|original,*,,DOCUMENTED|LIVE_VERIFIED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_image).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file),body,object,false,,,,,*,,DOCUMENTED,A file input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).type,body,string,true,input_file,,,input_file,*,,DOCUMENTED,The type of the input item. Always `input_file`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).filename,body,string,false,,,,,*,,DOCUMENTED,The name of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).file_data,body,string,false,,,,,*,,DOCUMENTED,The content of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).file_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](input_file).detail,body,string,false,,,,auto|low|high,*,,DOCUMENTED,"The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to render the file at higher quality. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).type,body,string,true,output_text,,,output_text,*,,DOCUMENTED,The type of the output text. Always `output_text`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).text,body,string,true,,,,,*,,DOCUMENTED,The text output from the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations,body,array<object<file_citation> | object<url_citation> | object<container_file_citation> | object<file_path>>,true,,,,,*,,DOCUMENTED,The annotations of the text output.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations[],body,object<file_citation> | object<url_citation> | object<container_file_citation> | object<file_path>,false,,,,,*,,DOCUMENTED,An annotation that applies to a span of output text.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations[](file_citation),body,object,false,,,,,*,,DOCUMENTED,A citation to a file.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations[](url_citation),body,object,false,,,,,*,,DOCUMENTED,A citation for a web resource used to generate a model response.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations[](container_file_citation),body,object,false,,,,,*,,DOCUMENTED,A citation for a container file used to generate a model response.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).annotations[](file_path),body,object,false,,,,,*,,DOCUMENTED,A path to a file.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs,body,array<object>,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs[],body,object,false,,,,,*,,DOCUMENTED,The log probability of a token.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs[].token,body,string,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs[].logprob,body,number,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs[].bytes,body,array<integer>,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](output_text).logprobs[].top_logprobs,body,array<object>,true,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](refusal).type,body,string,true,refusal,,,refusal,*,,DOCUMENTED,The type of the refusal. Always `refusal`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](message).content[](refusal).refusal,body,string,true,,,,,*,,DOCUMENTED,The refusal explanation from the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_text),body,object,false,,,,,*,,DOCUMENTED,A text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_text).type,body,string,true,input_text,,,input_text,*,,DOCUMENTED,The type of the input item. Always `input_text`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_text).text,body,string,true,,,,,*,,DOCUMENTED,The text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image),body,object,false,,,,,*,,DOCUMENTED,An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).type,body,string,true,input_image,,,input_image,*,,DOCUMENTED,The type of the input item. Always `input_image`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).image_url,body,string | null,false,,,,,*,,DOCUMENTED,The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).detail,body,string,true,,,,low|high|auto|original,*,,DOCUMENTED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_image).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file),body,object,false,,,,,*,,DOCUMENTED,A file input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).type,body,string,true,input_file,,,input_file,*,,DOCUMENTED,The type of the input item. Always `input_file`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).filename,body,string,false,,,,,*,,DOCUMENTED,The name of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).file_data,body,string,false,,,,,*,,DOCUMENTED,The content of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).file_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](function_call_output).output[](input_file).detail,body,string,false,,,,auto|low|high,*,,DOCUMENTED,"The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to render the file at higher quality. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_text),body,object,false,,,,,*,,DOCUMENTED,A text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_text).type,body,string,true,input_text,,,input_text,*,,DOCUMENTED,The type of the input item. Always `input_text`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_text).text,body,string,true,,,,,*,,DOCUMENTED,The text input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_text).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_text).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image),body,object,false,,,,,*,,DOCUMENTED,An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).type,body,string,true,input_image,,,input_image,*,,DOCUMENTED,The type of the input item. Always `input_image`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).image_url,body,string | null,false,,,,,*,,DOCUMENTED,The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).detail,body,string,true,,,,low|high|auto|original,*,,DOCUMENTED,"The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_image).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file),body,object,false,,,,,*,,DOCUMENTED,A file input to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).type,body,string,true,input_file,,,input_file,*,,DOCUMENTED,The type of the input item. Always `input_file`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).file_id,body,string | null,false,,,,,*,,DOCUMENTED,The ID of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).filename,body,string,false,,,,,*,,DOCUMENTED,The name of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).file_data,body,string,false,,,,,*,,DOCUMENTED,The content of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).prompt_cache_breakpoint,body,object,false,,,,,*,,DOCUMENTED,Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).prompt_cache_breakpoint.mode,body,string,true,explicit,,,explicit,*,,DOCUMENTED,The breakpoint mode. Always `explicit`.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).file_url,body,string,false,,,,,*,,DOCUMENTED,The URL of the file to be sent to the model.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,input[](custom_tool_call_output).output[](input_file).detail,body,string,false,,,,auto|low|high,*,,DOCUMENTED,"The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to render the file at higher quality. Defaults to `auto`.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses,personality,body,string,false,,,,friendly|pragmatic,unknown,,LIVE_DISCOVERED,A model-owned style preset to apply to this request (PersonalityEnum). Not listed in the CreateResponse schema; documented for POST /v1/responses/input_tokens.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,WS /v1/responses (response.create),stream_id,body,string,false,,1,256,,*,,DOCUMENTED,"WebSocket-only lane identifier (1-256 chars, [A-Za-z0-9_.-]). Requests with the same stream_id run FIFO; different lanes run concurrently (max 16 in-flight, 32 distinct named lanes per connection). Not accepted on HTTP POST /v1/responses.",https://developers.openai.com/api/docs/guides/websocket-mode
openai,WS /v1/responses (response.create),generate,body,boolean,false,true,,,,*,,DOCUMENTED,"WebSocket warm-up: `generate: false` prepares request state (tools, instructions) without producing output; returns a response id usable as previous_response_id.",https://developers.openai.com/api/docs/guides/websocket-mode
openai,POST /v1/responses?beta=true,multi_agent,body,object | null,false,,,,,*,,DOCUMENTED|BETA,Configuration for server-hosted multi-agent execution.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses?beta=true,multi_agent.enabled,body,boolean,true,,,,,*,,DOCUMENTED|BETA,Whether to enable server-hosted multi-agent execution for this response.,https://developers.openai.com/api/reference/resources/responses/methods/create
openai,POST /v1/responses?beta=true,multi_agent.max_concurrent_subagents,body,integer,false,,1,,,*,,DOCUMENTED|BETA,"`max_concurrent_subagents` sets the maximum number of subagents that can be active simultaneously across the entire agent tree. It includes all descendants—children, grandchildren, and deeper subagents—but excludes the root agent. The API does not impose a fixed upper bound on this setting. The default is `3`, which is recommended for most workloads. Multi-agent runs also have no fixed limit on tree depth or the total number of subagents created during a run.",https://developers.openai.com/api/reference/resources/responses/methods/create
openai,GET /v1/responses/{response_id},include,query,array<string>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Additional fields to include in the response. See the `include` parameter for Response creation above for more information.,https://developers.openai.com/api/reference/resources/responses/methods/retrieve
openai,GET /v1/responses/{response_id},stream,query,boolean,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set to true, the model response data will be streamed to the client as it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format). See the [Streaming section below](https://developers.openai.com/api/reference/resources/responses/streaming-events) for more information.",https://developers.openai.com/api/reference/resources/responses/methods/retrieve
openai,GET /v1/responses/{response_id},starting_after,query,integer,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The sequence number of the event after which to start streaming.,https://developers.openai.com/api/reference/resources/responses/methods/retrieve
openai,GET /v1/responses/{response_id},include_obfuscation,query,boolean,false,,,,,*,,DOCUMENTED,"When true, stream obfuscation will be enabled. Stream obfuscation adds random characters to an `obfuscation` field on streaming delta events to normalize payload sizes as a mitigation to certain side-channel attacks. These obfuscation fields are included by default, but add a small amount of overhead to the data stream. You can set `include_obfuscation` to false to optimize for bandwidth if you trust the network links between your application and the OpenAI API.",https://developers.openai.com/api/reference/resources/responses/methods/retrieve
openai,GET /v1/responses/{response_id}/input_items,limit,query,integer,false,20,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/responses/subresources/input_items/methods/list
openai,GET /v1/responses/{response_id}/input_items,order,query,string,false,,,,asc|desc,*,,DOCUMENTED,The order to return the input items in. Default is `desc`. - `asc`: Return the input items in ascending order. - `desc`: Return the input items in descending order.,https://developers.openai.com/api/reference/resources/responses/subresources/input_items/methods/list
openai,GET /v1/responses/{response_id}/input_items,after,query,string,false,,,,,*,,DOCUMENTED,"An item ID to list items after, used in pagination.",https://developers.openai.com/api/reference/resources/responses/subresources/input_items/methods/list
openai,GET /v1/responses/{response_id}/input_items,include,query,array<string>,false,,,,,*,,DOCUMENTED,Additional fields to include in the response. See the `include` parameter for Response creation above for more information.,https://developers.openai.com/api/reference/resources/responses/subresources/input_items/methods/list
openai,POST /v1/responses/input_tokens,model,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Model ID used to generate the response, like `gpt-4o` or `o3`. OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points. Refer to the [model guide](https://developers.openai.com/api/docs/models) to browse and compare available models.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input,body,string | array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or file inputs to the model, used to generate a response",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[],body,array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>> | object<program_output>,false,,,,,*,,DOCUMENTED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](message),body,object,false,,,,,*,,DOCUMENTED,A message input to the model with a role indicating instruction following hierarchy. Instructions given with the `developer` or `system` role take precedence over instructions given with the `user` role. Messages with the `assistant` role are presumed to have been generated by the model in previous interactions.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](file_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a file search tool call. See the [file search guide](https://developers.openai.com/api/docs/guides/tools-file-search) for more information.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](computer_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to a computer use tool. See the [computer use guide](https://developers.openai.com/api/docs/guides/tools-computer-use) for more information.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](computer_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a computer tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](web_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a web search tool call. See the [web search guide](https://developers.openai.com/api/docs/guides/tools-web-search) for more information.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](function_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a function. See the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling) for more information.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](function_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a function tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](tool_search_call),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](tool_search_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](additional_tools),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](configuration_update),body,object,false,,,,,gpt-6-astra,,DOCUMENTED,An update to the conversation's response configuration. The configuration remains in effect for subsequent responses until it is replaced by another configuration update.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](reasoning),body,object,false,,,,,*,,DOCUMENTED,A description of the chain of thought used by a reasoning model while generating a response. Be sure to include these items in your `input` to the Responses API for subsequent turns of a conversation if you are manually [managing context](https://developers.openai.com/api/docs/guides/conversation-state).,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](compaction),body,object,false,,,,,* (responses models supporting compaction),,DOCUMENTED,A compaction item generated by the [`v1/responses/compact` API](https://developers.openai.com/api/reference/resources/responses/methods/compact).,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](image_generation_call),body,object,false,,,,,*,,DOCUMENTED,An image generation request made by the model.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](code_interpreter_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run code.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](local_shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a command on the local shell.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](local_shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a local shell tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool representing a request to execute one or more shell commands.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output items emitted by a shell tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](apply_patch_call),body,object,false,,,,,*,,DOCUMENTED,"A tool call representing a request to create, delete, or update files using diff patches.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](apply_patch_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output emitted by an apply patch tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](mcp_list_tools),body,object,false,,,,,*,,DOCUMENTED,A list of tools available on an MCP server.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](mcp_approval_request),body,object,false,,,,,*,,DOCUMENTED,A request for human approval of a tool invocation.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](mcp_approval_response),body,object,false,,,,,*,,DOCUMENTED,A response to an MCP approval request.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](mcp_call),body,object,false,,,,,*,,DOCUMENTED,An invocation of a tool on an MCP server.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](custom_tool_call_output),body,object,false,,,,,*,,DOCUMENTED,"The output of a custom tool call from your code, being sent back to the model.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](custom_tool_call),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](compaction_trigger),body,object,false,,,,,*,,DOCUMENTED,Compacts the current context. Must be the final input item.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](item_reference),body,object,false,,,,,*,,DOCUMENTED,An internal identifier for an item to reference.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,input[](program_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,previous_response_id,body,string | null,false,,,,,*,,DOCUMENTED,The unique ID of the previous response to the model. Use this to create multi-turn conversations. Learn more about [conversation state](https://developers.openai.com/api/docs/guides/conversation-state). Cannot be used in conjunction with `conversation`.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tools,body,array<object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An array of tools the model may call while generating a response. You can specify which tool to use by setting the `tool_choice` parameter.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tools[],body,object<function> | object<file_search> | object<computer> | object<computer_use_preview> | WebSearchTool | object<mcp> | object<code_interpreter> | object<programmatic_tool_calling> | object<image_generation> | object<local_shell> | object<shell> | object<custom> | object<namespace> | object<tool_search> | WebSearchPreviewTool | object<apply_patch>,false,,,,,*,,DOCUMENTED,A tool that can be used to generate a response.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,text,body,object | null,false,,,,,*,,DOCUMENTED,Configuration options for a text response from the model. Can be plain text or structured JSON data. Learn more: - [Text inputs and outputs](https://developers.openai.com/api/docs/guides/text) - [Structured Outputs](https://developers.openai.com/api/docs/guides/structured-outputs),https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,text.format,body,object<text> | object<json_schema> | object<json_object>,false,,,,,*,,DOCUMENTED,"An object specifying the format that the model must output. Configuring `{ ""type"": ""json_schema"" }` enables Structured Outputs, which ensures the model will match your supplied JSON schema. Learn more in the [Structured Outputs guide](https://developers.openai.com/api/docs/guides/structured-outputs). The default format is `{ ""type"": ""text"" }` with no additional options. **Not recommended for gpt-4o and newer models:** Setting to `{ ""type"": ""json_object"" }` enables the older JSON mode, which ensures the message the model generates is valid JSON. Using `json_schema` is preferred for models that support it.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,text.verbosity,body,string | null,false,medium,,,low|medium|high,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4*|gpt-5.2*|gpt-5.1*|gpt-5*,,DOCUMENTED,"Constrains the verbosity of the model's response. Lower values will result in more concise responses, while higher values will result in more verbose responses. Currently supported values are `low`, `medium`, and `high`. The default is `medium`.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning,body,object | null,false,,,,,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,**gpt-5 and o-series models only** Configuration options for [reasoning models](https://developers.openai.com/api/docs/guides/reasoning).,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning.mode,body,string,false,,,,,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"Controls the reasoning execution mode for the request. When returned on a response, this is the effective execution mode.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning.effort,body,string | null,false,medium,,,none|minimal|low|medium|high|xhigh|max,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"Constrains effort on reasoning for reasoning models. Currently supported values are `none`, `minimal`, `low`, `medium`, `high`, `xhigh`, and `max`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Not all reasoning models support every value. See the [reasoning guide](https://developers.openai.com/api/docs/guides/reasoning) for model-specific support.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning.summary,body,string | null,false,,,,auto|concise|detailed,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"A summary of the reasoning performed by the model. This can be useful for debugging and understanding the model's reasoning process. One of `auto`, `concise`, or `detailed`. `concise` is supported for `computer-use-preview` models and all reasoning models after `gpt-5`.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning.context,body,string | null,false,,,,auto|current_turn|all_turns,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED,"Controls which reasoning items are rendered back to the model on later turns. If omitted or set to `auto`, the model determines the context mode. The `gpt-5.6` model family defaults to `all_turns`; earlier models default to `current_turn`. When returned on a response, this is the effective reasoning context mode used for the response.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,reasoning.generate_summary,body,string | null,false,,,,auto|concise|detailed,gpt-6-astra|gpt-5.6-*|gpt-5.5|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.2|gpt-5.1|gpt-5|gpt-5-mini|gpt-5-nano|gpt-5*-codex*|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LEGACY,"**Deprecated:** use `summary` instead. A summary of the reasoning performed by the model. This can be useful for debugging and understanding the model's reasoning process. One of `auto`, `concise`, or `detailed`.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,truncation,body,string,false,,,,auto|disabled,*,,DOCUMENTED|LEGACY,"The truncation strategy to use for the model response. - `auto`: If the input to this Response exceeds the model's context window size, the model will truncate the response to fit the context window by dropping items from the beginning of the conversation. - `disabled` (default): If the input size will exceed the context window size for a model, the request will fail with a 400 error.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,instructions,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A system (or developer) message inserted into the model's context. When used along with `previous_response_id`, the instructions from a previous response will not be carried over to the next response. This makes it simple to swap out system (or developer) messages in new responses.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,personality,body,string,false,,,,,*,,DOCUMENTED,A model-owned style preset to apply to this request. Omit this parameter to use the model's default style. Supported values may expand over time. Values must be at most 64 characters.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,conversation,body,string | object,false,,,,,*,,DOCUMENTED,The conversation that this response belongs to. Items from this conversation are prepended to `input_items` for this response request. Input items and output items from this response are automatically added to this conversation after this response completes.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,conversation(object),body,object,false,,,,,*,,DOCUMENTED,The conversation that this response belongs to.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice,body,string | object<allowed_tools> | ToolChoiceTypes | object<function> | object<mcp> | object<custom> | object<programmatic_tool_calling> | object<apply_patch> | object<shell>,false,,,,,*,,DOCUMENTED,"Controls which tool the model should use, if any.",https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(allowed_tools),body,object,false,,,,,*,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(hosted_tool),body,object,false,,,,,*,,DOCUMENTED,Indicates that the model should use a built-in tool to generate a response. [Learn more about built-in tools](https://developers.openai.com/api/docs/guides/tools).,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(function),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific function.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(mcp),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific tool on a remote MCP server.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(custom),body,object,false,,,,,*,,DOCUMENTED,Use this option to force the model to call a specific custom tool.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(programmatic_tool_calling),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(apply_patch),body,object,false,,,,,*,,DOCUMENTED,Forces the model to call the apply_patch tool when executing a tool call.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,tool_choice(shell),body,object,false,,,,,*,,DOCUMENTED,Forces the model to call the shell tool when a tool call is required.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/input_tokens,parallel_tool_calls,body,boolean | null,false,,,,,*,,DOCUMENTED,Whether to allow the model to run tool calls in parallel.,https://developers.openai.com/api/reference/resources/responses/subresources/input_tokens/methods/count
openai,POST /v1/responses/compact,model,body,string | null,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Model ID used to generate the response, like `gpt-6-astra`. OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points. Refer to the [model guide](https://developers.openai.com/api/docs/models) to browse and compare available models.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input,body,string | array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image, or file inputs to the model, used to generate a response",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[],body,array<object<message> | object<message> | object<file_search_call> | object<computer_call> | object<computer_call_output> | object<web_search_call> | object<function_call> | object<function_call_output> | object<tool_search_call> | object<tool_search_output> | object<additional_tools> | object<configuration_update> | object<reasoning> | object<compaction> | object<image_generation_call> | object<code_interpreter_call> | object<local_shell_call> | object<local_shell_call_output> | object<shell_call> | object<shell_call_output> | object<apply_patch_call> | object<apply_patch_call_output> | object<mcp_list_tools> | object<mcp_approval_request> | object<mcp_approval_response> | object<mcp_call> | object<custom_tool_call_output> | object<custom_tool_call> | object<compaction_trigger> | ItemReferenceParam | object<program> | object<program_output>> | object<program_output>,false,,,,,*,,DOCUMENTED,"A list of one or many input items to the model, containing different content types.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](message),body,object,false,,,,,*,,DOCUMENTED,A message input to the model with a role indicating instruction following hierarchy. Instructions given with the `developer` or `system` role take precedence over instructions given with the `user` role. Messages with the `assistant` role are presumed to have been generated by the model in previous interactions.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](file_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a file search tool call. See the [file search guide](https://developers.openai.com/api/docs/guides/tools-file-search) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](computer_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to a computer use tool. See the [computer use guide](https://developers.openai.com/api/docs/guides/tools-computer-use) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](computer_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a computer tool call.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](web_search_call),body,object,false,,,,,*,,DOCUMENTED,The results of a web search tool call. See the [web search guide](https://developers.openai.com/api/docs/guides/tools-web-search) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](function_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a function. See the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling) for more information.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](function_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a function tool call.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](tool_search_call),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](tool_search_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](additional_tools),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](configuration_update),body,object,false,,,,,gpt-6-astra,,DOCUMENTED,An update to the conversation's response configuration. The configuration remains in effect for subsequent responses until it is replaced by another configuration update.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](reasoning),body,object,false,,,,,*,,DOCUMENTED,A description of the chain of thought used by a reasoning model while generating a response. Be sure to include these items in your `input` to the Responses API for subsequent turns of a conversation if you are manually [managing context](https://developers.openai.com/api/docs/guides/conversation-state).,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](compaction),body,object,false,,,,,* (responses models supporting compaction),,DOCUMENTED,A compaction item generated by the [`v1/responses/compact` API](https://developers.openai.com/api/reference/resources/responses/methods/compact).,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](image_generation_call),body,object,false,,,,,*,,DOCUMENTED,An image generation request made by the model.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](code_interpreter_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run code.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](local_shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool call to run a command on the local shell.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](local_shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The output of a local shell tool call.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](shell_call),body,object,false,,,,,*,,DOCUMENTED,A tool representing a request to execute one or more shell commands.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](shell_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output items emitted by a shell tool call.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](apply_patch_call),body,object,false,,,,,*,,DOCUMENTED,"A tool call representing a request to create, delete, or update files using diff patches.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](apply_patch_call_output),body,object,false,,,,,*,,DOCUMENTED,The streamed output emitted by an apply patch tool call.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](mcp_list_tools),body,object,false,,,,,*,,DOCUMENTED,A list of tools available on an MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](mcp_approval_request),body,object,false,,,,,*,,DOCUMENTED,A request for human approval of a tool invocation.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](mcp_approval_response),body,object,false,,,,,*,,DOCUMENTED,A response to an MCP approval request.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](mcp_call),body,object,false,,,,,*,,DOCUMENTED,An invocation of a tool on an MCP server.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](custom_tool_call_output),body,object,false,,,,,*,,DOCUMENTED,"The output of a custom tool call from your code, being sent back to the model.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](custom_tool_call),body,object,false,,,,,*,,DOCUMENTED,A call to a custom tool created by the model.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](compaction_trigger),body,object,false,,,,,*,,DOCUMENTED,Compacts the current context. Must be the final input item.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](item_reference),body,object,false,,,,,*,,DOCUMENTED,An internal identifier for an item to reference.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](program),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,input[](program_output),body,object,false,,,,,*,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,previous_response_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the previous response to the model. Use this to create multi-turn conversations. Learn more about [conversation state](https://developers.openai.com/api/docs/guides/conversation-state). Cannot be used in conjunction with `conversation`.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,instructions,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A system (or developer) message inserted into the model's context. When used along with `previous_response_id`, the instructions from a previous response will not be carried over to the next response. This makes it simple to swap out system (or developer) messages in new responses.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,prompt_cache_key,body,string | null,false,,,,,*,,DOCUMENTED,A key to use when reading from or writing to the prompt cache.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,prompt_cache_retention,body,string | null,false,,,,in_memory|24h,gpt-5.5|gpt-5.5-pro|gpt-5.4|gpt-5.2|gpt-5.1-codex-max|gpt-5.1|gpt-5.1-codex|gpt-5.1-codex-mini|gpt-5.1-chat-latest|gpt-5|gpt-5-codex|gpt-4.1|(in_memory: all cache-capable models),,DOCUMENTED|LEGACY,How long to retain a prompt cache entry created by this request.,https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,prompt_cache_options,body,object | null,false,,,,,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Options for prompt caching. Supported for `gpt-5.6` and later models. By default, OpenAI automatically chooses one implicit cache breakpoint. You can add explicit breakpoints to content blocks with `prompt_cache_breakpoint`. Each request can write up to four breakpoints. For cache matching, OpenAI considers up to the latest 80 breakpoints in the conversation, without a content-block lookback limit. Set `mode` to `explicit` to disable the implicit breakpoint. The `ttl` defaults to `30m`, which is currently the only supported value. See the [prompt caching guide](https://developers.openai.com/api/docs/guides/prompt-caching) for current details.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,prompt_cache_options.ttl,body,string,false,,,,30m,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"The minimum lifetime applied to every implicit and explicit cache breakpoint written by the request. Defaults to `30m`, which is currently the only supported value. The backend may retain cache entries for longer.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,prompt_cache_options.mode,body,string,false,,,,implicit|explicit,gpt-6-astra|gpt-5.6-sol|gpt-5.6-terra|gpt-5.6-luna,,DOCUMENTED,"Controls whether OpenAI automatically creates an implicit cache breakpoint. Defaults to `implicit`. With `implicit`, OpenAI creates one implicit breakpoint and writes up to the latest three explicit breakpoints in the request. With `explicit`, OpenAI does not create an implicit breakpoint and writes up to the latest four explicit breakpoints. If there are no explicit breakpoints, the request does not use prompt caching.",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,POST /v1/responses/compact,service_tier,body,string | null,false,,,,auto|default|fast|flex|priority,*,,DOCUMENTED,"Specifies the processing type used for serving the request. - If set to 'auto', then the request will be processed with the service tier configured in the Project settings. Unless otherwise configured, the Project will use 'default'. - If set to 'default', then the request will be processed with the standard pricing and performance for the selected model. - If set to '[flex](https://developers.openai.com/api/docs/guides/flex-processing)', then the request will be processed with the Flex Processing service tier. - To opt-in to [Fast mode](https://developers.openai.com/api/docs/guides/fast-mode) at the request level, include the `service_tier=fast` or `service_tier=priority` parameter for Responses or Chat Completions. For models with a dedicated Fast tier, either value resolves to `service_tier=fast`; for other models, either value resolves to `service_tier=priority`. - When not set, the ",https://developers.openai.com/api/reference/resources/responses/methods/compact
openai,GET /v1/safety/alerts/{id},id,path,string,true,,,38,,,,DOCUMENTED|LIVE_VERIFIED,Project safety alert id (maxLength 38; webhook pattern ^alert_[0-9a-f]{32}$; guide examples use 'salert_123').,https://developers.openai.com/api/docs/guides/safety-checks/misalignment-monitoring
openai,GET /v1/safety/cases/{id},id,path,string,true,,,128,,,,LIVE_DISCOVERED|ACCOUNT_RESTRICTED,Safety case id (maxLength 128; webhook example 'C-abc123').,https://raw.githubusercontent.com/openai/openai-openapi/master/openapi.yaml
openai,POST /v1/content_provenance_checks,file,body,file (multipart; set the part's media type),true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Image (image/png, image/jpeg, image/webp) or audio (MP3, Opus as audio/ogg, AAC, FLAC, WAV, PCM; <= 60 s decoded). <= 50 MiB, one file per request. Do not add a separate `type` field.",https://developers.openai.com/api/docs/guides/content-provenance
openai,POST /v1/responses,safety_identifier,body,string|null,false,,,64,,,,DOCUMENTED,"Stable, privacy-preserving end-user id (hash of username/email), max 64 chars. Lets OpenAI block an abusive end user instead of the whole org; blocked ids get an 'identifier blocked' error. Replaces the legacy `user` field together with prompt_cache_key.",https://developers.openai.com/api/docs/guides/safety-best-practices
openai,POST /v1/chat/completions,safety_identifier,body,string|null,false,,,64,,,,DOCUMENTED,Same semantics as on /v1/responses (max 64 chars).,https://developers.openai.com/api/docs/guides/safety-best-practices
openai,POST /v1/realtime/client_secrets,OpenAI-Safety-Identifier,header,string,false,,,,,,,DOCUMENTED,Header carrying the safety identifier for Realtime sessions (also on direct WebSocket/WebRTC connection requests from a trusted backend). Identifiers do not carry over between APIs or sessions.,https://developers.openai.com/api/docs/guides/safety-checks
openai,POST /v1/responses,moderation,body,{model: string},false,,,,,,,DOCUMENTED,Top-level object requesting moderation scores for the input and the generated output in the same response (omni-moderation-latest). Scores arrive after the full output when streaming; may contain an error object if moderation failed.,https://developers.openai.com/api/docs/guides/moderation
openai,POST /v1/responses,tools[type=function].type,body,string,true,function,,,function,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The type of the function tool. Always `function`.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].name,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The name of the function to call.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].async,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].description,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,A description of the function. Used by the model to determine whether or not to call the function.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].parameters,body,object,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,A JSON schema object describing the parameters of the function.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].output_schema,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,A JSON schema object describing the JSON value encoded in string outputs for this function.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].strict,body,boolean,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether strict parameter validation is enforced for this function tool.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].defer_loading,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether this function is deferred and loaded via tool search.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=function].allowed_callers,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=custom].type,body,string,true,custom,,,custom,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The type of the custom tool. Always `custom`.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].name,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"The name of the custom tool, used to identify it in tool calls.",https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].async,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether the tool response can be returned asynchronously versus immediately returned on next response creation.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].description,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"Optional description of the custom tool, used to provide more context.",https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].format,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The input format for the custom tool. Default is unconstrained text.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].format.type,body,string,true,text,,,text,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Unconstrained text format. Always `text`.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].format.syntax,body,string,true,,,,lark|regex,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].format.definition,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The grammar definition.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].defer_loading,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether this tool should be deferred and discovered via tool search.,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=custom].allowed_callers,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/function-calling#custom-tools
openai,POST /v1/responses,tools[type=namespace].type,body,string,true,namespace,,,namespace,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The type of the tool. Always `namespace`.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].name,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"The namespace name used in tool calls (for example, `crm`).",https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].description,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,A description of the namespace shown to the model.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools,body,array,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The function/custom tools available inside this namespace.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].name,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].description,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].parameters,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].strict,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"Whether to enforce strict parameter validation. If omitted, Responses attempts to use strict validation when the schema is compatible, and falls back to non-strict validation otherwise.",https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].type,body,string,true,function,,,function,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].async,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether the tool response can be returned asynchronously versus immediately returned on next response creation.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].output_schema,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,A JSON Schema describing the JSON value encoded in string outputs for this function tool. This does not describe content-array outputs.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].defer_loading,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Whether this function should be deferred and discovered via tool search.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].allowed_callers,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].format,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The input format for the custom tool. Default is unconstrained text.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].format.type,body,string,true,text,,,"{""$comment"": ""depth-limited""}",chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Unconstrained text format. Always `text`.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].format.syntax,body,"{""$comment"": ""depth-limited""}",true,,,,"{""$comment"": ""depth-limited""}",chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=namespace].tools[].format.definition,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-audio-mini|gpt-oss-120b|gpt-oss-20b|o1|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The grammar definition.,https://developers.openai.com/api/docs/guides/function-calling#namespaces
openai,POST /v1/responses,tools[type=tool_search].type,body,string,true,tool_search,,,tool_search,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The type of the tool. Always `tool_search`.,https://developers.openai.com/api/docs/guides/tools-tool-search
openai,POST /v1/responses,tools[type=tool_search].execution,body,string,false,,,,server|client,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-tool-search
openai,POST /v1/responses,tools[type=tool_search].description,body,string,false,,,,,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,Description shown to the model for a client-executed tool search tool.,https://developers.openai.com/api/docs/guides/tools-tool-search
openai,POST /v1/responses,tools[type=tool_search].parameters,body,object,false,,,,,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-tool-search
openai,POST /v1/responses,tools[type=programmatic_tool_calling].type,body,string,true,programmatic_tool_calling,,,programmatic_tool_calling,,,DOCUMENTED,The type of the tool. Always `programmatic_tool_calling`.,https://developers.openai.com/api/docs/guides/tools-programmatic-tool-calling
openai,POST /v1/responses,tools[type=web_search].type,body,string,true,web_search,,,web_search|web_search_2025_08_26,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The type of the web search tool. One of `web_search` or `web_search_2025_08_26`.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].external_web_access,body,boolean,false,true,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Allow live internet access for web search. Defaults to true when omitted. When false, the web search tool runs in offline/cache-only mode and will not fetch new external content.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].filters,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Filters for the search.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].filters.allowed_domains,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Allowed domains for the search. If not provided, all domains are allowed.
Subdomains of the provided domains are allowed as well.

Example: `[""pubmed.ncbi.nlm.nih.gov""]`",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The approximate location of the user.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location.type,body,string,false,approximate,,,approximate,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The type of location approximation. Always `approximate`.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location.country,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"The two-letter [ISO country code](https://en.wikipedia.org/wiki/ISO_3166-1) of the user, e.g. `US`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location.region,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Free text input for the region of the user, e.g. `California`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location.city,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Free text input for the city of the user, e.g. `San Francisco`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].user_location.timezone,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"The [IANA timezone](https://timeapi.io/documentation/iana-timezones) of the user, e.g. `America/Los_Angeles`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].search_context_size,body,string,false,medium,,,low|medium|high,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"High level guidance for the amount of context window space to use for the search. One of `low`, `medium`, or `high`. `medium` is the default.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].type,body,string,true,web_search_preview,,,web_search_preview|web_search_preview_2025_03_11,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,The type of the web search tool. One of `web_search_preview` or `web_search_preview_2025_03_11`.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location.type,body,string,true,approximate,,,approximate,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,The type of location approximation. Always `approximate`.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location.country,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,"The two-letter [ISO country code](https://en.wikipedia.org/wiki/ISO_3166-1) of the user, e.g. `US`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location.region,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,"Free text input for the region of the user, e.g. `California`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location.city,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,"Free text input for the city of the user, e.g. `San Francisco`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].user_location.timezone,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,"The [IANA timezone](https://timeapi.io/documentation/iana-timezones) of the user, e.g. `America/Los_Angeles`.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].search_context_size,body,string,false,,,,low|medium|high,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search_preview].search_content_types,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LEGACY,,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=file_search].type,body,string,true,file_search,,,file_search,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The type of the file search tool. Always `file_search`.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].vector_store_ids,body,array,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The IDs of the vector stores to search.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].max_num_results,body,integer,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The maximum number of results to return. This number should be between 1 and 50 inclusive.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options.ranker,body,string,false,,,,auto|default-2024-11-15,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options.score_threshold,body,number,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"The score threshold for the file search, a number between 0 and 1. Numbers closer to 1 will attempt to return only the most relevant results, but may return fewer results.",https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options.hybrid_search,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options.hybrid_search.embedding_weight,body,number,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The weight of the embedding in the reciprocal ranking fusion.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].ranking_options.hybrid_search.text_weight,body,number,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The weight of the text in the reciprocal ranking fusion.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].filters,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].filters.type,body,string,true,eq,,,eq|ne|gt|gte|lt|lte|in|nin,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"Specifies the comparison operator: `eq`, `ne`, `gt`, `gte`, `lt`, `lte`, `in`, `nin`.
- `eq`: equals
- `ne`: not equal
- `gt`: greater than
- `gte`: greater than or equal
- `lt`: less than
- `lte`: less than or equal
- `in`: in
- `nin`: not in",https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].filters.key,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,The key to compare against the value.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].filters.value,body,oneOf,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,"The value to compare against the attribute key; supports string, number, or boolean types.",https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=file_search].filters.filters,body,array,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o1|o1-mini|o1-pro|o3|o3-mini|o3-pro|o4-mini,,DOCUMENTED|LIVE_VERIFIED,Array of filters to combine. Items can be `ComparisonFilter` or `CompoundFilter`.,https://developers.openai.com/api/docs/guides/tools-file-search
openai,POST /v1/responses,tools[type=code_interpreter].type,body,string,true,,,,code_interpreter,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The type of the code interpreter tool. Always `code_interpreter`.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container,body,oneOf,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"The code interpreter container. Can be a container ID or an object that
specifies uploaded file IDs to make available to your code, along with an
optional `memory_limit` setting.",https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.type,body,string,true,auto,,,auto,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Always `auto`.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.file_ids,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,An optional list of uploaded files to make available to your code.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.memory_limit,body,string,false,,,,1g|4g|16g|64g,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.network_policy,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Network access policy for the container.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.network_policy.type,body,string,true,disabled,,,disabled,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Disable outbound network access. Always `disabled`.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.network_policy.allowed_domains,body,array,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,A list of allowed domains when type is `allowlist`.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].container.network_policy.domain_secrets,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Optional domain-scoped secrets for allowlisted domains.,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=code_interpreter].allowed_callers,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1-mini|o3|o3-deep-research|o3-mini|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/tools-code-interpreter
openai,POST /v1/responses,tools[type=computer].type,body,string,true,computer,,,computer,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED,The type of the computer tool. Always `computer`.,https://developers.openai.com/api/docs/guides/tools-computer-use
openai,POST /v1/responses,tools[type=computer_use_preview].type,body,string,true,computer_use_preview,,,computer_use_preview,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|PREVIEW,The type of the computer use tool. Always `computer_use_preview`.,https://developers.openai.com/api/docs/guides/tools-computer-use
openai,POST /v1/responses,tools[type=computer_use_preview].environment,body,string,true,,,,windows|mac|linux|ubuntu|browser,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|PREVIEW,,https://developers.openai.com/api/docs/guides/tools-computer-use
openai,POST /v1/responses,tools[type=computer_use_preview].display_width,body,integer,true,,,,,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|PREVIEW,The width of the computer display.,https://developers.openai.com/api/docs/guides/tools-computer-use
openai,POST /v1/responses,tools[type=computer_use_preview].display_height,body,integer,true,,,,,gpt-5.4|gpt-5.4-mini|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|PREVIEW,The height of the computer display.,https://developers.openai.com/api/docs/guides/tools-computer-use
openai,POST /v1/responses,tools[type=image_generation].type,body,string,true,,,,image_generation,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,The type of the image generation tool. Always `image_generation`.,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].model,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].quality,body,string,false,auto,,,low|medium|high|xhigh|max|auto,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"The quality of the generated image. The GPT image models support `low`,
`medium`, and `high`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`,
including their `2026-09-08` snapshots, also support `xhigh` and `max`.
Default: `auto`.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].size,body,oneOf,false,auto,,,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"The size of the generated images. For `gpt-image-2`, `gpt-image-2-2026-04-21`, `gpt-image-2.5-sunburst`, `gpt-image-2.5-sunburst-2026-09-08`, `gpt-image-2.5-flare`, and `gpt-image-2.5-flare-2026-09-08`, arbitrary resolutions are supported as `WIDTHxHEIGHT` strings, for example `1536x864`. Width and height must both be divisible by 16 and the requested aspect ratio must be between 1:3 and 3:1. Resolutions above `2560x1440` are experimental, and the maximum supported resolution is `3840x2160`. The requested size must also satisfy the model's current pixel and edge limits. The standard sizes `1024x1024`, `1536x1024`, and `1024x1536` are supported by the GPT image models; `auto` is supported for models that allow automatic sizing. For `dall-e-2`, use one of `256x256`, `512x512`, or `1024x1024`. For `dall-e-3`, use one of `1024x1024`, `1792x1024`, or `1024x1792`.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].output_format,body,string,false,png,,,png|webp|jpeg,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"The output format of the generated image. One of `png`, `webp`, or
`jpeg`. Default: `png`.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].output_compression,body,integer,false,100,0,100,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,Compression level for the output image. Default: 100.,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].moderation,body,string,false,auto,,,auto|low,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,Moderation level for the generated image. Default: `auto`.,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].background,body,string,false,auto,,,transparent|opaque|auto,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"Set the background of the generated image. One of `transparent`, `opaque`,
or `auto`. `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare`, including
their `2026-09-08` snapshots, support `opaque` and `transparent`
backgrounds. Transparent backgrounds are available for supported GPT Image
models. For `gpt-image-2` and `gpt-image-2-2026-04-21`, this support is in
preview. When using `transparent`, set the output format to `png` or `webp`.
Default: `auto`.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].input_fidelity,body,string,false,,,,high|low,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"Control how much effort the model will exert to match the style and features, especially facial features, of input images. This parameter is only supported for `gpt-image-1` and `gpt-image-1.5` and later models, unsupported for `gpt-image-1-mini`. Supports `high` and `low`. Defaults to `low`.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].input_image_mask,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"Optional mask for inpainting. Contains `image_url`
(string, optional) and `file_id` (string, optional).",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].input_image_mask.image_url,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,Base64-encoded mask image.,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].input_image_mask.file_id,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,File ID for the mask image.,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].partial_images,body,integer,false,0,0,3,,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,"Number of partial images to generate in streaming mode, from 0 (default value) to 3.",https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=image_generation].action,body,string,false,,,,generate|edit|auto,chat-latest|gpt-4.1|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-5|gpt-5-chat-latest|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|o3|o3-mini|o3-pro,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-image-generation
openai,POST /v1/responses,tools[type=mcp].type,body,string,true,,,,mcp,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The type of the MCP tool. Always `mcp`.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].server_label,body,string,true,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"A label for this MCP server, used to identify it in tool calls.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].server_url,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"The URL for the MCP server. One of `server_url`, `connector_id`, or
`tunnel_id` must be provided.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].connector_id,body,string,false,,,,connector_dropbox|connector_gmail|connector_googlecalendar|connector_googledrive|connector_microsoftteams|connector_outlookcalendar|connector_outlookemail|connector_sharepoint,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Identifier for service connectors, like those available in ChatGPT. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more
about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).

This field is deprecated for models released after September 1, 2026.
Use `server_url` to connect to a remote MCP server, or `tunnel_id` to
connect through a Secure MCP Tunnel.

Currently supported `connector_id` values are:

- Dropbox: `connector_dropbox`
- Gmail: `connector_gmail`
- Google Calendar: `connector_googlecalendar`
- Google Drive: `connector_googledrive`
- Microsoft Teams: `connector_microsoftteams`
- Outlook Calendar: `connector_outlookcalendar`
- Outlook Email: `connector_outlookemail`
- SharePoint: `connector_sharepoint`",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].tunnel_id,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"The Secure MCP Tunnel ID to use instead of a direct server URL. One of
`server_url`, `connector_id`, or `tunnel_id` must be provided.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].authorization,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"An OAuth access token that can be used with a remote MCP server, either
with a custom MCP server URL or a service connector. Your application
must handle the OAuth authorization flow and provide the token here.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].server_description,body,string,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Optional description of the MCP server, used to provide more context.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].headers,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Optional HTTP headers to send to the MCP server. Use for authentication
or other purposes.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].allowed_tools,body,oneOf,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,List of allowed tool names or a filter object.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].allowed_tools.tool_names,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,List of allowed tool names.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].allowed_tools.read_only,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].allowed_callers,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval,body,oneOf,false,always,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Specify which of the MCP server's tools require approval.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.always,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,A filter object to specify which tools are allowed.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.always.tool_names,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,List of allowed tool names.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.always.read_only,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.never,body,object,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,A filter object to specify which tools are allowed.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.never.tool_names,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,List of allowed tool names.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].require_approval.never.read_only,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,"Indicates whether or not a tool modifies data or is read-only. If an
MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint),
it will match this filter.",https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=mcp].defer_loading,body,boolean,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4.1-nano|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o1|o1-mini|o1-pro|o3|o3-deep-research|o3-mini|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_VERIFIED,Whether this MCP tool is deferred and discovered via tool search.,https://developers.openai.com/api/docs/guides/tools-connectors-mcp
openai,POST /v1/responses,tools[type=shell].type,body,string,true,shell,,,shell,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The type of the shell tool. Always `shell`.,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment,body,oneOf,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.type,body,string,true,container_auto,,,container_auto,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,Automatically creates a container for this request,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.file_ids,body,array,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,An optional list of uploaded files to make available to your code.,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.memory_limit,body,string,false,,,,1g|4g|16g|64g,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.network_policy,body,oneOf,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,Network access policy for the container.,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.network_policy.type,body,"{""$comment"": ""depth-limited""}",true,"{""$comment"": ""depth-limited""}",,,"{""$comment"": ""depth-limited""}",gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,{'$comment': 'depth-limited'},https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.network_policy.allowed_domains,body,"{""$comment"": ""depth-limited""}",true,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,{'$comment': 'depth-limited'},https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.network_policy.domain_secrets,body,"{""$comment"": ""depth-limited""}",false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,{'$comment': 'depth-limited'},https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills,body,array,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,An optional list of skills referenced by id or inline data.,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].type,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].skill_id,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].version,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].name,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].description,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.skills[].source,body,object,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].environment.container_id,body,string,true,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The ID of the referenced container.,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=shell].allowed_callers,body,array,false,,,,,gpt-5.2|gpt-5.2-codex|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/tools-shell
openai,POST /v1/responses,tools[type=local_shell].type,body,string,true,local_shell,,,local_shell,codex-mini-latest,,DOCUMENTED|LEGACY,The type of the local shell tool. Always `local_shell`.,https://developers.openai.com/api/docs/guides/tools-local-shell
openai,POST /v1/responses,tools[type=apply_patch].type,body,string,true,apply_patch,,,apply_patch,gpt-5.1|gpt-5.2|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The type of the tool. Always `apply_patch`.,https://developers.openai.com/api/docs/guides/tools-apply-patch
openai,POST /v1/responses,tools[type=apply_patch].allowed_callers,body,array,false,,,,,gpt-5.1|gpt-5.2|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest,,DOCUMENTED|LIVE_VERIFIED,The tool invocation context(s).,https://developers.openai.com/api/docs/guides/tools-apply-patch
openai,POST /v1/responses,tool_choice,body,oneOf,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"How the model should select which tool (or tools) to use when generating
a response. See the `tools` parameter to see how to specify which tools
the model can call.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tool_choice.type,body,string,true,,,,allowed_tools,,,DOCUMENTED|LIVE_VERIFIED,Allowed tool configuration type. Always `allowed_tools`.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tool_choice.mode,body,string,true,,,,auto|required,,,DOCUMENTED|LIVE_VERIFIED,"Constrains the tools available to the model to a pre-defined set.

`auto` allows the model to pick from among the allowed tools and generate a
message.

`required` requires the model to call one or more of the allowed tools.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tool_choice.tools,body,array,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"A list of tool definitions that the model should be allowed to call.

For the Responses API, the list of tool definitions might look like:
```json
[
  { ""type"": ""function"", ""name"": ""get_weather"" },
  { ""type"": ""mcp"", ""server_label"": ""deepwiki"" },
  { ""type"": ""image_generation"" }
]
```",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tool_choice.name,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The name of the function to call.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tool_choice.server_label,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The label of the MCP server to use.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,parallel_tool_calls,body,boolean,false,true,,,,,,DOCUMENTED|LIVE_VERIFIED,Whether to enable [parallel function calling](https://developers.openai.com/api/docs/guides/function-calling#parallel-function-calling) during tool use.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,include,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Extra output fields; tool-related values: file_search_call.results, web_search_call.results, web_search_call.action.sources, code_interpreter_call.outputs, computer_call_output.output.image_url.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,max_tool_calls,body,integer,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"Maximum number of built-in tool calls (web search, file search, MCP…) in a response; used notably with deep research models.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/responses,tools[type=web_search].return_token_budget,body,string,false,,,,default|unlimited,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_DISCOVERED,"Token budget for returned web search content (GPT-5+ reasoning web search). Documented in the guide, echoed live (default), absent from OpenAPI spec.",https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].filters.blocked_domains,body,array,false,,,,,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_DISCOVERED,Up to 100 blocked domains (guide); absent from OpenAPI spec.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/responses,tools[type=web_search].search_content_types,body,array,false,,,,text|image,chat-latest|gpt-4.1|gpt-4.1-mini|gpt-4o|gpt-4o-mini|gpt-4o-mini-audio-preview|gpt-5|gpt-5-chat-latest|gpt-5-codex|gpt-5-mini|gpt-5-nano|gpt-5-pro|gpt-5.1|gpt-5.1-chat-latest|gpt-5.1-codex|gpt-5.1-codex-max|gpt-5.1-codex-mini|gpt-5.2|gpt-5.2-chat-latest|gpt-5.2-codex|gpt-5.2-pro|gpt-5.3-chat-latest|gpt-5.3-codex|gpt-5.4|gpt-5.4-mini|gpt-5.4-nano|gpt-5.4-pro|gpt-5.5|gpt-5.5-pro|gpt-5.6-cyber|gpt-5.6-luna|gpt-5.6-sol|gpt-5.6-terra|gpt-6-astra|gpt-daybreak-blue-latest|gpt-daybreak-red-latest|gpt-oss-120b|gpt-oss-20b|o3|o3-deep-research|o3-pro|o4-mini|o4-mini-deep-research,,DOCUMENTED|LIVE_DISCOVERED,Image search: include 'image' (and 'text'). Echoed live as ['text']; present in spec only on web_search_preview.,https://developers.openai.com/api/docs/guides/tools-web-search
openai,POST /v1/chat/completions,tools[],body,oneOf,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].type,body,string,true,,,,function,,,DOCUMENTED,"The type of the tool. Currently, only `function` is supported.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].function,body,object,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].function.description,body,string,false,,,,,,,DOCUMENTED,"A description of what the function does, used by the model to choose when and how to call the function.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].function.name,body,string,true,,,,,,,DOCUMENTED,"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].function.parameters,body,object,false,,,,,,,DOCUMENTED,"The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://developers.openai.com/api/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.

Omitting `parameters` defines a function with an empty parameter list.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].function.strict,body,boolean,false,false,,,,,,DOCUMENTED,"Whether to enable strict schema adherence when generating the function call. If set to true, the model will follow the exact schema defined in the `parameters` field. Only a subset of JSON Schema is supported when `strict` is `true`. Learn more about Structured Outputs in the [function calling guide](https://developers.openai.com/api/docs/guides/function-calling).",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom,body,object,true,,,,,,,DOCUMENTED,Properties of the custom tool.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.name,body,string,true,,,,,,,DOCUMENTED,"The name of the custom tool, used to identify it in tool calls.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.description,body,string,false,,,,,,,DOCUMENTED,"Optional description of the custom tool, used to provide more context.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.format,body,oneOf,false,,,,,,,DOCUMENTED,The input format for the custom tool. Default is unconstrained text.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.format.type,body,string,true,,,,text,,,DOCUMENTED,Unconstrained text format. Always `text`.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.format.grammar,body,object,true,,,,,,,DOCUMENTED,Your chosen grammar.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.format.grammar.definition,body,"{""$comment"": ""depth-limited""}",true,,,,,,,DOCUMENTED,{'$comment': 'depth-limited'},https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tools[].custom.format.grammar.syntax,body,"{""$comment"": ""depth-limited""}",true,,,,"{""$comment"": ""depth-limited""}",,,DOCUMENTED,{'$comment': 'depth-limited'},https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice,body,oneOf,false,,,,,,,DOCUMENTED,"Controls which (if any) tool is called by the model.
`none` means the model will not call any tool and instead generates a message.
`auto` means the model can pick between generating a message or calling one or more tools.
`required` means the model must call one or more tools.
Specifying a particular tool via `{""type"": ""function"", ""function"": {""name"": ""my_function""}}` forces the model to call that tool.

`none` is the default when no tools are present. `auto` is the default if tools are present.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.type,body,string,true,,,,allowed_tools,,,DOCUMENTED,Allowed tool configuration type. Always `allowed_tools`.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.allowed_tools,body,object,true,,,,,,,DOCUMENTED,Constrains the tools available to the model to a pre-defined set.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.allowed_tools.mode,body,string,true,,,,auto|required,,,DOCUMENTED,"Constrains the tools available to the model to a pre-defined set.

`auto` allows the model to pick from among the allowed tools and generate a
message.

`required` requires the model to call one or more of the allowed tools.",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.allowed_tools.tools,body,array,true,,,,,,,DOCUMENTED,"A list of tool definitions that the model should be allowed to call.

For the Chat Completions API, the list of tool definitions might look like:
```json
[
  { ""type"": ""function"", ""function"": { ""name"": ""get_weather"" } },
  { ""type"": ""function"", ""function"": { ""name"": ""get_time"" } }
]
```",https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.function,body,object,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.function.name,body,string,true,,,,,,,DOCUMENTED,The name of the function to call.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.custom,body,object,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/chat/completions,tool_choice.custom.name,body,string,true,,,,,,,DOCUMENTED,The name of the custom tool to call.,https://developers.openai.com/api/docs/guides/function-calling
openai,POST /v1/containers,name,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Name of the container to create.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,file_ids,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,IDs of files to copy to the container.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,expires_after,body,object,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Container expiration time in seconds relative to the 'anchor' time.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,expires_after.anchor,body,string,true,,,,last_active_at,,,DOCUMENTED|LIVE_VERIFIED,Time anchor for the expiration time. Currently only 'last_active_at' is supported.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,expires_after.minutes,body,integer,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,An optional list of skills referenced by id or inline data.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].type,body,string,true,skill_reference,,,skill_reference,,,DOCUMENTED|LIVE_VERIFIED,References a skill created with the /v1/skills endpoint.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].skill_id,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The ID of the referenced skill.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].version,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Optional skill version. Use a positive integer or 'latest'. Omit for default.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].name,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The name of the skill.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].description,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The description of the skill.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].source,body,object,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Inline skill payload,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].source.type,body,string,true,base64,,,base64,,,DOCUMENTED|LIVE_VERIFIED,The type of the inline skill source. Must be `base64`.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].source.media_type,body,string,true,application/zip,,,application/zip,,,DOCUMENTED|LIVE_VERIFIED,The media type of the inline skill payload. Must be `application/zip`.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,skills[].source.data,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Base64-encoded skill zip bundle.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,memory_limit,body,string,false,,,,1g|4g|16g|64g,,,DOCUMENTED|LIVE_VERIFIED,"Optional memory limit for the container. Defaults to ""1g"".",https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy,body,oneOf,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Network access policy for the container.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.type,body,string,true,disabled,,,disabled,,,DOCUMENTED|LIVE_VERIFIED,Disable outbound network access. Always `disabled`.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.allowed_domains,body,array,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,A list of allowed domains when type is `allowlist`.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.domain_secrets,body,array,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Optional domain-scoped secrets for allowlisted domains.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.domain_secrets[].domain,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The domain associated with the secret.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.domain_secrets[].name,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The name of the secret to inject for the domain.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers,network_policy.domain_secrets[].value,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,The secret value to inject for the domain.,https://developers.openai.com/api/reference/resources/containers/methods/create
openai,POST /v1/containers/{container_id}/files,file_id,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Name of the file to create.,https://developers.openai.com/api/reference/resources/containers/subresources/files/methods/create
openai,POST /v1/containers/{container_id}/files,file,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,The File object (not file name) to be uploaded.,https://developers.openai.com/api/reference/resources/containers/subresources/files/methods/create
openai,POST /v1/skills,files,body,oneOf,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/reference/resources/skills/methods/create
openai,POST /v1/skills/{skill_id}/versions,files,body,oneOf,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,,https://developers.openai.com/api/reference/resources/skills/subresources/versions/methods/create
openai,POST /v1/skills/{skill_id}/versions,default,body,boolean,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Whether to set this version as the default.,https://developers.openai.com/api/reference/resources/skills/subresources/versions/methods/create
openai,POST /v1/skills/{skill_id},default_version,body,string,true,,,,,,,DOCUMENTED|FAILED_VERIFICATION,The skill version number to set as default.,https://developers.openai.com/api/reference/resources/skills/methods/update
openai,GET /v1/containers,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers,name,query,string,false,,,,,,,DOCUMENTED,Filter results by container name.,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id},container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,DELETE /v1/containers/{container_id},container_id,path,string,true,,,,,,,DOCUMENTED,The ID of the container to delete.,https://developers.openai.com/api/reference/resources/containers
openai,POST /v1/containers/{container_id}/files,container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files,container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files/{file_id},container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,DELETE /v1/containers/{container_id}/files/{file_id},container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,DELETE /v1/containers/{container_id}/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files/{file_id}/content,container_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/containers/{container_id}/files/{file_id}/content,file_id,path,string,true,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/containers
openai,GET /v1/skills,limit,query,integer,false,,0,100,,,,DOCUMENTED,Number of items to retrieve,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills,order,query,string,false,,,,,,,DOCUMENTED,Sort order of results by timestamp. Use `asc` for ascending order or `desc` for descending order.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills,after,query,string,false,,,,,,,DOCUMENTED,Identifier for the last item from the previous pagination request,https://developers.openai.com/api/reference/resources/skills
openai,DELETE /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill to delete.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill to retrieve.,https://developers.openai.com/api/reference/resources/skills
openai,POST /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/content,skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill to download.,https://developers.openai.com/api/reference/resources/skills
openai,POST /v1/skills/{skill_id}/versions,skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill to version.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions,skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions,limit,query,integer,false,,0,100,,,,DOCUMENTED,Number of versions to retrieve.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions,order,query,string,false,,,,,,,DOCUMENTED,Sort order of results by version number.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions,after,query,string,false,,,,,,,DOCUMENTED,The skill version ID to start after.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions/{version},skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions/{version},version,path,string,true,,,,,,,DOCUMENTED,The version number to retrieve.,https://developers.openai.com/api/reference/resources/skills
openai,DELETE /v1/skills/{skill_id}/versions/{version},skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill.,https://developers.openai.com/api/reference/resources/skills
openai,DELETE /v1/skills/{skill_id}/versions/{version},version,path,string,true,,,,,,,DOCUMENTED,The skill version number.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions/{version}/content,skill_id,path,string,true,,,,,,,DOCUMENTED,The identifier of the skill.,https://developers.openai.com/api/reference/resources/skills
openai,GET /v1/skills/{skill_id}/versions/{version}/content,version,path,string,true,,,,,,,DOCUMENTED,The skill version number.,https://developers.openai.com/api/reference/resources/skills
openai,POST /v1/uploads,filename,body,string,true,,,,,,,DOCUMENTED,The name of the file to upload.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,purpose,body,string,true,,,,assistants|batch|fine-tune|vision,,,DOCUMENTED,"The intended purpose of the uploaded file.

See the [documentation on File
purposes](https://developers.openai.com/api/reference/resources/files/methods/create#%28resource%29%20files%20%3E%20%28method%29%20create%20%3E%20%28params%29%200%20%3E%20%28param%29%20purpose%20%3E%20%28schema%29).",https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,bytes,body,integer,true,,,,,,,DOCUMENTED,The number of bytes in the file you are uploading.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,mime_type,body,string,true,,,,,,,DOCUMENTED,"The MIME type of the file.


This must fall within the supported MIME types for your file purpose. See
the supported MIME types for assistants and vision.",https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,expires_after,body,object,false,,,,,,,DOCUMENTED,"The expiration policy for a file. By default, files with `purpose=batch` expire after 30 days and all other files are persisted until they are manually deleted. [schema: FileExpirationAfter]",https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,expires_after.anchor,body,string,true,,,,created_at,,,DOCUMENTED,Anchor timestamp after which the expiration policy applies. Supported anchors: `created_at`.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads,expires_after.seconds,body,integer,true,,3600,2592000,,,,DOCUMENTED,The number of seconds after the anchor time that the file will expire. Must be between 3600 (1 hour) and 2592000 (30 days).,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/cancel,upload_id,path,string,true,,,,,,,DOCUMENTED,The ID of the Upload.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/complete,upload_id,path,string,true,,,,,,,DOCUMENTED,The ID of the Upload.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/complete,part_ids,body,array<string>,true,,,,,,,DOCUMENTED,The ordered list of Part IDs.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/complete,md5,body,string,false,,,,,,,DOCUMENTED,The optional md5 checksum for the file contents to verify if the bytes uploaded matches what you expect.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/parts,upload_id,path,string,true,,,,,,,DOCUMENTED,The ID of the Upload.,https://developers.openai.com/api/reference/resources/uploads
openai,POST /v1/uploads/{upload_id}/parts,data,body(multipart),string,true,,,,,,,DOCUMENTED,The chunk of bytes for this Part.,https://developers.openai.com/api/reference/resources/uploads
openai,GET /v1/vector_stores,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores,before,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,file_ids,body,array<string>,false,,,,,,,DOCUMENTED,A list of [File](https://developers.openai.com/api/reference/resources/files) IDs that the vector store should use. Useful for tools like `file_search` that can access files.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,name,body,string,false,,,,,,,DOCUMENTED,The name of the vector store.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,description,body,string,false,,,,,,,DOCUMENTED,A description for the vector store. Can be used to describe the vector store's purpose.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,expires_after,body,object,false,,,,,,,DOCUMENTED,The expiration policy for a vector store. [schema: VectorStoreExpirationAfter],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,expires_after.anchor,body,string,true,,,,last_active_at,,,DOCUMENTED,Anchor timestamp after which the expiration policy applies. Supported anchors: `last_active_at`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,expires_after.days,body,integer,true,,1,365,,,,DOCUMENTED,The number of days after the anchor time that the vector store will expire.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,chunking_strategy,body,object,false,,,,,,,DOCUMENTED,"The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy. Only applicable if `file_ids` is non-empty.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,chunking_strategy.type,body,string,true,,,,static,,,DOCUMENTED,Always `static`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,chunking_strategy.static,body,object,true,,,,,,,DOCUMENTED,[schema: StaticChunkingStrategy],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,chunking_strategy.static.max_chunk_size_tokens,body,integer,true,,100,4096,,,,DOCUMENTED,The maximum number of tokens in each chunk. The default value is `800`. The minimum value is `100` and the maximum value is `4096`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,chunking_strategy.static.chunk_overlap_tokens,body,integer,true,,,,,,,DOCUMENTED,"The number of tokens that overlap between chunks. The default value is `400`.

Note that the overlap must not exceed half of `max_chunk_size_tokens`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores,metadata,body,object | null,false,,,,,,,DOCUMENTED,[schema: Metadata],https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store to retrieve.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store to modify.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},name,body,string,false,,,,,,,DOCUMENTED,The name of the vector store.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},expires_after,body,object,false,,,,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},expires_after.anchor,body,string,true,,,,last_active_at,,,DOCUMENTED,Anchor timestamp after which the expiration policy applies. Supported anchors: `last_active_at`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},expires_after.days,body,integer,true,,1,365,,,,DOCUMENTED,The number of days after the anchor time that the vector store will expire.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id},metadata,body,object | null,false,,,,,,,DOCUMENTED,[schema: Metadata],https://developers.openai.com/api/reference/resources/vector_stores
openai,DELETE /v1/vector_stores/{vector_store_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store to delete.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store for which to create a File Batch.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,file_ids,body,array<string>,false,,,,,,,DOCUMENTED,"A list of [File](https://developers.openai.com/api/reference/resources/files) IDs that the vector store should use. Useful for tools like `file_search` that can access files. If `attributes` or `chunking_strategy` are provided, they will be applied to all files in the batch. The maximum batch size is 2000 files. This endpoint is recommended for multi-file ingestion and helps reduce per-vector-store write request pressure. Mutually exclusive with `files`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files,body,array<object>,false,,,,,,,DOCUMENTED,A list of objects that each include a `file_id` plus optional `attributes` or `chunking_strategy`. Use this when you need to override metadata for specific files. The global `attributes` or `chunking_strategy` will be ignored and must be specified for each file. The maximum batch size is 2000 files. This endpoint is recommended for multi-file ingestion and helps reduce per-vector-store write request pressure. Mutually exclusive with `file_ids`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files[].file_id,body,string,true,,,,,,,DOCUMENTED,"A [File](https://developers.openai.com/api/reference/resources/files) ID that the vector store should use. Useful for tools like `file_search` that can access files. For multi-file ingestion, we recommend [`file_batches`](https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create) to minimize per-vector-store write requests.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files[].chunking_strategy,body,object,false,,,,,,,DOCUMENTED,"The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy. [schema: ChunkingStrategyRequestParam]",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files[].chunking_strategy.type,body,string,true,,,,static,,,DOCUMENTED,Always `static`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files[].chunking_strategy.static,body,object,true,,,,,,,DOCUMENTED,[schema: StaticChunkingStrategy],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,files[].attributes,body,object | null,false,,,,,,,DOCUMENTED,[schema: VectorStoreFileAttributes],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,chunking_strategy,body,object,false,,,,,,,DOCUMENTED,"The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy. [schema: ChunkingStrategyRequestParam]",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,chunking_strategy.type,body,string,true,,,,static,,,DOCUMENTED,Always `static`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,chunking_strategy.static,body,object,true,,,,,,,DOCUMENTED,[schema: StaticChunkingStrategy],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,chunking_strategy.static.max_chunk_size_tokens,body,integer,true,,100,4096,,,,DOCUMENTED,The maximum number of tokens in each chunk. The default value is `800`. The minimum value is `100` and the maximum value is `4096`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,chunking_strategy.static.chunk_overlap_tokens,body,integer,true,,,,,,,DOCUMENTED,"The number of tokens that overlap between chunks. The default value is `400`.

Note that the overlap must not exceed half of `max_chunk_size_tokens`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches,attributes,body,object | null,false,,,,,,,DOCUMENTED,[schema: VectorStoreFileAttributes],https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the file batch belongs to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id},batch_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file batch being retrieved.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the file batch belongs to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel,batch_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file batch to cancel.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the files belong to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,batch_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file batch that the files belong to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,before,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/file_batches/{batch_id}/files,filter,query,string,false,,,,in_progress|completed|failed|cancelled,,,DOCUMENTED,"Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the files belong to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,limit,query,integer,false,20,,,,,,DOCUMENTED,"A limit on the number of objects to be returned. Limit can range between 1 and 100, and the default is 20.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,order,query,string,false,desc,,,asc|desc,,,DOCUMENTED,Sort order by the `created_at` timestamp of the objects. `asc` for ascending order and `desc` for descending order.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,after,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `after` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, ending with obj_foo, your subsequent call can include after=obj_foo in order to fetch the next page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,before,query,string,false,,,,,,,DOCUMENTED,"A cursor for use in pagination. `before` is an object ID that defines your place in the list. For instance, if you make a list request and receive 100 objects, starting with obj_foo, your subsequent call can include before=obj_foo in order to fetch the previous page of the list.",https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files,filter,query,string,false,,,,in_progress|completed|failed|cancelled,,,DOCUMENTED,"Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store for which to create a File.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,file_id,body,string,true,,,,,,,DOCUMENTED,"A [File](https://developers.openai.com/api/reference/resources/files) ID that the vector store should use. Useful for tools like `file_search` that can access files. For multi-file ingestion, we recommend [`file_batches`](https://developers.openai.com/api/reference/resources/vector_stores/subresources/file_batches/methods/create) to minimize per-vector-store write requests.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,chunking_strategy,body,object,false,,,,,,,DOCUMENTED,"The chunking strategy used to chunk the file(s). If not set, will use the `auto` strategy. [schema: ChunkingStrategyRequestParam]",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,chunking_strategy.type,body,string,true,,,,static,,,DOCUMENTED,Always `static`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,chunking_strategy.static,body,object,true,,,,,,,DOCUMENTED,[schema: StaticChunkingStrategy],https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,chunking_strategy.static.max_chunk_size_tokens,body,integer,true,,100,4096,,,,DOCUMENTED,The maximum number of tokens in each chunk. The default value is `800`. The minimum value is `100` and the maximum value is `4096`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,chunking_strategy.static.chunk_overlap_tokens,body,integer,true,,,,,,,DOCUMENTED,"The number of tokens that overlap between chunks. The default value is `400`.

Note that the overlap must not exceed half of `max_chunk_size_tokens`.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,attributes,body,object | null,false,,,,,,,DOCUMENTED,[schema: VectorStoreFileAttributes],https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files/{file_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the file belongs to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file being retrieved.,https://developers.openai.com/api/reference/resources/vector_stores
openai,DELETE /v1/vector_stores/{vector_store_id}/files/{file_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store that the file belongs to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,DELETE /v1/vector_stores/{vector_store_id}/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file to delete.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files/{file_id},vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store the file belongs to.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files/{file_id},file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file to update attributes.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files/{file_id},attributes,body,object | null,true,,,,,,,DOCUMENTED,[schema: VectorStoreFileAttributes],https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files/{file_id}/content,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store.,https://developers.openai.com/api/reference/resources/vector_stores
openai,GET /v1/vector_stores/{vector_store_id}/files/{file_id}/content,file_id,path,string,true,,,,,,,DOCUMENTED,The ID of the file within the vector store.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,vector_store_id,path,string,true,,,,,,,DOCUMENTED,The ID of the vector store to search.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,query,body,string | array,true,,,,,,,DOCUMENTED,A query string for a search,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,rewrite_query,body,boolean,false,,,,,,,DOCUMENTED,Whether to rewrite the natural language query for vector search.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,max_num_results,body,integer,false,10,1,50,,,,DOCUMENTED,The maximum number of results to return. This number should be between 1 and 50 inclusive.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters,body,object | object,false,,,,,,,DOCUMENTED,A filter to apply based on file attributes.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.type,body,string,true,,,,and|or,,,DOCUMENTED,Type of operation: `and` or `or`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.key,body,string,true,,,,,,,DOCUMENTED,The key to compare against the value.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.value,body,string | number | boolean | array,true,,,,,,,DOCUMENTED,"The value to compare against the attribute key; supports string, number, or boolean types.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.filters,body,array<object>,true,,,,,,,DOCUMENTED,Array of filters to combine. Items can be `ComparisonFilter` or `CompoundFilter`.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.filters[].type,body,string,true,eq,,,eq|ne|gt|gte|lt|lte|in|nin,,,DOCUMENTED,"Specifies the comparison operator: `eq`, `ne`, `gt`, `gte`, `lt`, `lte`, `in`, `nin`.
- `eq`: equals
- `ne`: not equal
- `gt`: greater than
- `gte`: greater than or equal
- `lt`: less than
- `lte`: less than or equal
- `in`: in
- `nin`: not in",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.filters[].key,body,string,true,,,,,,,DOCUMENTED,The key to compare against the value.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,filters.filters[].value,body,string | number | boolean | array,true,,,,,,,DOCUMENTED,"The value to compare against the attribute key; supports string, number, or boolean types.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,ranking_options,body,object,false,,,,,,,DOCUMENTED,Ranking options for search.,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,ranking_options.ranker,body,string,false,auto,,,none|auto|default-2024-11-15,,,DOCUMENTED,"Enable re-ranking; set to `none` to disable, which can help reduce latency.",https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/search,ranking_options.score_threshold,body,number,false,,,1,,,,DOCUMENTED,,https://developers.openai.com/api/reference/resources/vector_stores
openai,POST /v1/vector_stores/{vector_store_id}/files,attributes.<key>,body,string | number | boolean,false,,,16,,,,DOCUMENTED,"Attribute map used by search filters: at most 16 keys, each key and string value <= 256 characters (Retrieval guide).",https://developers.openai.com/api/docs/guides/retrieval#attributes
openai,POST /v1/vector_stores/{vector_store_id}/search,ranking_options.hybrid_search.embedding_weight,body,number,false,,,,,,,DOCUMENTED|UNVERIFIED,Reciprocal-rank-fusion weight for semantic (embedding) matches (alias rrf_embedding_weight). At least one of the two weights must be > 0. Documented in the Retrieval guide only (not in the OpenAPI spec).,https://developers.openai.com/api/docs/guides/retrieval#ranking
openai,POST /v1/vector_stores/{vector_store_id}/search,ranking_options.hybrid_search.text_weight,body,number,false,,,,,,,DOCUMENTED|UNVERIFIED,Reciprocal-rank-fusion weight for sparse keyword matches (alias rrf_text_weight). Documented in the Retrieval guide only.,https://developers.openai.com/api/docs/guides/retrieval#ranking
openai,POST /v1/videos,prompt,body,string,true,,1,32000,,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED,"Text prompt (1-32000 chars) describing subject, camera, lighting, motion.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos,model,body,string,false,sora-2,,,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED,sora-2 (default) | sora-2-pro | dated snapshots. sora-2 alias -> sora-2-2025-12-08; sora-2-pro -> sora-2-pro-2025-10-06.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos,seconds,body,string,false,4,,,4|8|12|16 (guide)|20 (guide),sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED,"Clip duration as a string. Spec enum '4' | '8' | '12' (default '4'); guide/changelog (2026-03-12): both models also accept '16' and '20'; extensions doc lists 4, 8, 12, 16, 20.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos,size,body,string,false,720x1280,,,"720x1280|1280x720|1024x1792|1792x1024|1080x1920 (pro, guide)|1920x1080 (pro, guide)",sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED,"Resolution WIDTHxHEIGHT. Spec enum: 720x1280 (default), 1280x720, 1024x1792, 1792x1024. Guide + pricing add 1080x1920 / 1920x1080 (sora-2-pro only, $0.70/s). 1024p and 1080p are sora-2-pro only.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos,input_reference,body,"file (multipart) | {file_id?, image_url?} (JSON)",false,,,,,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED,"Optional first-frame image reference. Multipart: upload bytes (`-F input_reference=@img.jpeg;type=image/jpeg`). JSON/Batch: object with exactly one of file_id or image_url (data URL allowed, <= 20 MiB). Images with human faces are rejected.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos,characters,body,array<{id: string}>,false,,,2,,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,DOCUMENTED|DEPRECATED|UNVERIFIED,Reusable character ids from POST /v1/videos/characters (max 2 per video); mention the character name verbatim in the prompt. Combinable with input_reference; not supported by extensions.,https://developers.openai.com/api/docs/guides/video-generation
openai,GET /v1/videos,after,query,string,false,,,,,,,DOCUMENTED|DEPRECATED,Cursor: id of the last item of the previous page.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos,limit,query,integer,false,,,,,,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Page size.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos,order,query,string,false,,,,asc|desc,,,DOCUMENTED|DEPRECATED,asc | desc by timestamp.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos/{video_id},video_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Video job id (must start with 'video').,https://developers.openai.com/api/reference/resources/videos
openai,DELETE /v1/videos/{video_id},video_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,Video job id.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos/{video_id}/content,video_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Video job id.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos/{video_id}/content,variant,query,string,false,video,,,video|thumbnail|spritesheet,,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"video (default, MP4) | thumbnail (webp) | spritesheet (jpg).",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/{video_id}/remix,video_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,Completed source video id.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/{video_id}/remix,prompt,body,string,true,,,,,,,DOCUMENTED|DEPRECATED,Updated prompt for the remix.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/edits,video,body,{id: string} (JSON) | file (multipart),true,,,,,,,DOCUMENTED|DEPRECATED,Completed video to edit (id) or an uploaded MP4 (eligible customers only). Model is inferred from the id.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/edits,prompt,body,string,true,,1,32000,,,,DOCUMENTED|DEPRECATED,How to edit the source video (1-32000 chars). Keep to one clear adjustment.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/edits,model,body,string,false,,,,sora-2|sora-2-pro|sora-2-2025-10-06|sora-2-pro-2025-10-06|sora-2-2025-12-08,,,DOCUMENTED|DEPRECATED,Required only when uploading a new video (multipart).,https://developers.openai.com/api/docs/guides/video-generation
openai,POST /v1/videos/extensions,video,body,{id: string} (JSON) | file (multipart),true,,,,,,,DOCUMENTED|DEPRECATED,Completed video to extend.,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/extensions,prompt,body,string,true,,1,32000,,,,DOCUMENTED|DEPRECATED,How the scene continues (1-32000 chars).,https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/extensions,seconds,body,string,true,,,,4|8|12|16|20,,,DOCUMENTED|DEPRECATED,"Length of the new segment: 4, 8, 12, 16, 20 (spec enum only lists 4/8/12). Up to 6 extensions, 120 s total.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/characters,video,body,"file (multipart, video/mp4)",true,,,,,,,DOCUMENTED|DEPRECATED,"Short clip (2-4 s, 16:9 or 9:16, 720p-1080p) of a non-human subject.",https://developers.openai.com/api/reference/resources/videos
openai,POST /v1/videos/characters,name,body,string,true,,1,80,,,,DOCUMENTED|DEPRECATED,Display name (1-80 chars); must be used verbatim in prompts.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/videos/characters/{character_id},character_id,path,string,true,,,,,,,DOCUMENTED|DEPRECATED,Character id.,https://developers.openai.com/api/reference/resources/videos
openai,GET /v1/webhook_endpoints,limit,query,integer,false,,1,100,,,,DOCUMENTED,Maximum number of webhook endpoints to return. Defaults to 20.,https://developers.openai.com/api/reference/overview
openai,GET /v1/webhook_endpoints,after,query,string | null,false,,,,,,,DOCUMENTED,ID of the last webhook endpoint from the previous page.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints,name,body,string,true,,1,256,,,,DOCUMENTED,A human-readable name for the webhook endpoint.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints,url,body,string,true,,,2048,,,,DOCUMENTED,The HTTPS URL that receives webhook deliveries.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints,event_types,body,array<string>,true,,,,batch.completed|batch.failed|batch.expired|batch.cancelled|response.completed|response.failed|response.cancelled|response.incomplete|eval.run.succeeded|eval.run.failed|eval.run.canceled|fine_tuning.job.succeeded|fine_tuning.job.failed|fine_tuning.job.cancelled|realtime.call.incoming|video.completed|video.failed|safety.alert.created,,,DOCUMENTED,The event types that trigger deliveries to this endpoint.,https://developers.openai.com/api/reference/overview
openai,GET /v1/webhook_endpoints/{webhook_endpoint_id},webhook_endpoint_id,path,string,true,,,,,,,DOCUMENTED,The ID of the webhook endpoint to retrieve.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id},webhook_endpoint_id,path,string,true,,,,,,,DOCUMENTED,The ID of the webhook endpoint to update.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id},name,body,string,false,,1,256,,,,DOCUMENTED,A new human-readable name for the webhook endpoint.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id},url,body,string,false,,,2048,,,,DOCUMENTED,A new HTTPS URL that receives webhook deliveries.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id},event_types,body,array<string>,false,,,,batch.completed|batch.failed|batch.expired|batch.cancelled|response.completed|response.failed|response.cancelled|response.incomplete|eval.run.succeeded|eval.run.failed|eval.run.canceled|fine_tuning.job.succeeded|fine_tuning.job.failed|fine_tuning.job.cancelled|realtime.call.incoming|video.completed|video.failed|safety.alert.created,,,DOCUMENTED,The complete set of event types that should trigger deliveries.,https://developers.openai.com/api/reference/overview
openai,DELETE /v1/webhook_endpoints/{webhook_endpoint_id},webhook_endpoint_id,path,string,true,,,,,,,DOCUMENTED,The ID of the webhook endpoint to delete.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id}/rotate_secret,webhook_endpoint_id,path,string,true,,,,,,,DOCUMENTED,The ID of the webhook endpoint whose signing secret will be rotated.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id}/rotate_secret,keep_old_secret_active_for_24_hours,body,boolean,false,,,,,,,DOCUMENTED,"Whether to keep the previous signing secret valid for 24 hours after rotation. Defaults to false, which invalidates the previous secret immediately.",https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id}/test,webhook_endpoint_id,path,string,true,,,,,,,DOCUMENTED,The ID of the webhook endpoint to test.,https://developers.openai.com/api/reference/overview
openai,POST /v1/webhook_endpoints/{webhook_endpoint_id}/test,event_type,body,string,true,,,,batch.completed|batch.failed|batch.expired|batch.cancelled|response.completed|response.failed|response.cancelled|response.incomplete|eval.run.succeeded|eval.run.failed|eval.run.canceled|fine_tuning.job.succeeded|fine_tuning.job.failed|fine_tuning.job.cancelled|realtime.call.incoming|video.completed|video.failed|safety.alert.created,,,DOCUMENTED,,https://developers.openai.com/api/reference/overview
openai,POST https://auth.openai.com/oauth/token,grant_type,body,string,true,,,,urn:ietf:params:oauth:grant-type:token-exchange,,,DOCUMENTED,Must be urn:ietf:params:oauth:grant-type:token-exchange.,https://developers.openai.com/api/reference/workload-identity-federation
openai,POST https://auth.openai.com/oauth/token,subject_token_type,body,string,true,,,,urn:ietf:params:oauth:token-type:jwt|urn:ietf:params:oauth:token-type:id_token|urn:openai:params:oauth:token-type:x509,,,DOCUMENTED,JWT exchange: urn:ietf:params:oauth:token-type:jwt or urn:ietf:params:oauth:token-type:id_token. X.509 exchange: urn:openai:params:oauth:token-type:x509.,https://developers.openai.com/api/reference/workload-identity-federation
openai,POST https://auth.openai.com/oauth/token,subject_token,body,string,false,,,,,,,DOCUMENTED,"Externally issued OIDC JWT or SPIFFE JWT-SVID (required for JWT exchange; must be OMITTED for X.509 exchange). Must carry kid+alg header and iss, aud, sub, exp, iat claims matching the provider.",https://developers.openai.com/api/reference/workload-identity-federation
openai,POST https://auth.openai.com/oauth/token,identity_provider_id,body,string,true,,,,,,,DOCUMENTED,OpenAI Workload Identity Provider ID configured for the external issuer.,https://developers.openai.com/api/reference/workload-identity-federation
openai,POST https://auth.openai.com/oauth/token,service_account_id,body,string,true,,,,,,,DOCUMENTED,OpenAI service account ID resolved against the provider's service-account mappings (exactly one enabled mapping must match).,https://developers.openai.com/api/reference/workload-identity-federation
openai,POST https://auth.openai.com/oauth/token,scope,body,string,false,,,,,,,DOCUMENTED,Ignored for authorization: a scope value in the request does not grant access; effective permissions come from the mapping.,https://developers.openai.com/api/reference/workload-identity-federation
xai,POST /v1/chat/completions,deferred,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set to `true`, the request returns a `request_id`. You can then get the deferred response by GET `/v1/chat/deferred-completion/{request_id}`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,frequency_penalty,body,number | null,false,0,-2,2,,grok-4.20-0309-non-reasoning,,DOCUMENTED,"(Not supported by reasoning models) Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,logit_bias,body,object | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"(Unsupported) A JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,logprobs,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. Not supported by models `grok-4.20` and newer; the field will be silently ignored if set.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,max_completion_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An upper bound for the number of tokens that can be generated for a completion, only applies to visible output tokens (i.e. does not apply to tokens used for reasoning or function calls). Defaults to 128,000 when unset; set a larger value to allow longer generations.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,max_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,\[DEPRECATED\] The maximum number of tokens that can be generated in the chat completion. Deprecated in favor of `max_completion_tokens`.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages,body,array<Message>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of messages that make up the chat conversation. Different models support different message types, such as image and text.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content,body,Content,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,System prompt content.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].file,body,null | FileRef,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].file.file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The file ID from the Files API.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].file.url,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Public URL for a file attachment.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].image_url,body,null | ImageUrl,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].image_url.file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"File ID from the xAI Files API. Mutually exclusive with `url`.
The file must be an image (JPEG, PNG, or WebP) and fully uploaded.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].image_url.url,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Public URL or base64-encoded data URL of the image (JPEG, PNG, or WebP).
Also accepts `image_url` for compatibility.
Required when `file_id` is not set.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].text,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Text prompt.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the content part. Can be `text`, `image_url`, `text_file` or `file`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].name,body,string | null,false,,,,,*,,DOCUMENTED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].role,body,string,true,,,,system,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content,body,Content,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,System prompt content.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].name,body,string | null,false,,,,,*,,DOCUMENTED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].role,body,string,true,,,,user,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content,body,null | Content,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].name,body,string | null,false,,,,,*,,DOCUMENTED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].reasoning_content,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Assistant reasoning content.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].role,body,string,true,,,,assistant,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An array of tool calls available to the model on your machine.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].function,body,Function,true,,,,,*,,DOCUMENTED,Function to call for the tool call.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].function.arguments,body,string,true,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].function.name,body,string,true,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].id,body,string,true,,,,,*,,DOCUMENTED,"A unique ID of the tool call generated by xAI. After performing tool call's function, user provides this ID with tool call's result in the subsequent request to xAI. xAI can then match the tool call result sent with tool call request.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].index,body,integer | null,false,,,,,*,,DOCUMENTED,Index of the tool call.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_calls[].type,body,string | null,false,,,,,*,,DOCUMENTED,"Type of tool call, should be `""function""` or `""web_search_call""` or `""x_search_call""` or `""code_interpreter_call""` or `""mcp_call""`",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content,body,Content,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Content of the tool call result.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].role,body,string,true,,,,tool,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].tool_call_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The ID of the tool call received from assistant message response.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].content,body,Content,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Content of the tool call result.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].name,body,string | null,false,,,,,*,,DOCUMENTED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,messages[].role,body,string,true,,,,function,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,model,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Model name for the model to use. Obtainable from <https://console.x.ai/team/default/models> or <https://docs.x.ai/docs/models>.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,n,body,integer | null,false,1,1,,,*,,DOCUMENTED|LIVE_VERIFIED,How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,parallel_tool_calls,body,boolean | null,false,true,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set to false, the model can perform maximum one tool call.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,presence_penalty,body,number | null,false,0,-2,2,,grok-4.20-0309-non-reasoning,,DOCUMENTED|LIVE_VERIFIED,"(Not supported by `grok-3` and reasoning models) Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,prompt_cache_key,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A stable cache key for best-effort sticky routing / prompt-cache hits
across requests sharing a prompt prefix. Plumbed to `x-grok-conv-id`,
same as on `/v1/responses`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,reasoning_effort,body,string | null,false,,,,"low|medium|high|xhigh|none (LIVE_DISCOVERED, grok-4.3)",grok-4.6|grok-4.5|grok-4.3|grok-4.20-0309-reasoning|grok-4.20-multi-agent-0309|grok-build-0.1,,DOCUMENTED|LIVE_VERIFIED,Constrains how hard a reasoning model thinks before responding. Higher efforts use more reasoning tokens for deeper thinking. The supported values and the default depend on the model.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,response_format,body,null | ResponseFormat,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,response_format.type,body,string,true,,,,text,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,response_format.type,body,string,true,,,,json_object,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,response_format.json_schema,body,any,true,,,,,*,,DOCUMENTED,"A json schema representing the desired response schema. Includes the schema as a `""schema""` field.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,response_format.type,body,string,true,,,,json_schema,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,safety_identifier,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Supplied by the API client to identify the end user behind this request. A stable string that uniquely identifies each of your users; hash your internal user id or username rather than sending an email or name. Stored with the request metadata so a usage-policy violation can be attributed to that user rather than to the API key.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters,body,null | SearchParameters,false,,,,,*,,DOCUMENTED|RETIRED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.from_date,body,string | null,false,,,,,*,,DOCUMENTED|RETIRED,"Date from which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.max_search_results,body,integer | null,false,15,1,30,,*,,DOCUMENTED|RETIRED,Maximum number of search results to use.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.mode,body,string | null,false,auto,,,,*,,DOCUMENTED|RETIRED,"Choose the mode to query realtime data:
* `off`: no search performed and no external will be considered.
* `on` (default): the model will search in every sources for relevant data.
* `auto`: the model choose whether to search data or not and where to search the data.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.return_citations,body,boolean | null,false,true,,,,*,,DOCUMENTED|RETIRED,Whether to return citations in the response or not.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources,body,array | null,false,,,,,*,,DOCUMENTED|RETIRED,"List of sources to search in. If no sources specified, the model will look over the web and X by default.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].excluded_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"List of X handles to exclude from the search results. X posts returned will not include
any posts authored by these handles.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].included_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"NOTE: `included_x_handles` and `x_handles` are the same parameter.
`included_x_handles` is the new name but we keep both for backward compatibility.

X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].post_favorite_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum favorite count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].post_view_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum view count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].type,body,string,true,,,,x,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"DEPRECATED in favor of `included_x_handles`. Use `included_x_handles` instead.
X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].allowed_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to allow in the search results. This parameter act as a whitelist
where only those websites can be selected. A maximum of 5 websites can be selected.

Note 1: If no relevant information is found on those websites, the number of results
returned might be smaller than `max_search_results`.

Note 2: This parameter cannot be set with `excluded_websites`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.

Note 2: This parameter cannot be set with `allowed_websites`",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].type,body,string,true,,,,web,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].type,body,string,true,,,,news,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].links,body,array<string>,true,,,,,*,,DOCUMENTED,Links of the RSS feeds.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.sources[].type,body,string,true,,,,rss,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,search_parameters.to_date,body,string | null,false,,,,,*,,DOCUMENTED|RETIRED,"Date up to which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,seed,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same `seed` and parameters should return the same result. Determinism is not guaranteed, and you should refer to the `system_fingerprint` response parameter to monitor changes in the backend.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,service_tier,body,null | ServiceTier,false,,,,default|priority,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,stop,body,array | null,false,,,,,grok-4.20-0309-non-reasoning,,DOCUMENTED|LIVE_VERIFIED,(Not supported by reasoning models) Up to 4 sequences where the API will stop generating further tokens.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,stream,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set, partial message deltas will be sent. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,stream_options,body,null | StreamOptions,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,stream_options.include_usage,body,boolean,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Set an additional chunk to be streamed before the `data: [DONE]` message. The other chunks will return `null` in `usage` field.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED|LIVE_VERIFIED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tool_choice,body,null | ToolChoice,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tool_choice.function,body,null | FunctionChoice,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tool_choice.function.name,body,string,true,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tool_choice.type,body,string,true,,,,,*,,DOCUMENTED,"Type is always `""function""`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of tools the model may call in JSON-schema. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 350 functions are supported.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function,body,FunctionDefinition,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Definition of tool call available to the model.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function.defer_loading,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"When true, the definition is hidden from the model's prompt but stays
callable, loaded via a `tool_search` step (`/v1/responses` only).",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function.description,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A description of the function to indicate to the model when to call it.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function.name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The name of the function. If the model calls the function, this name is used in the
response.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function.parameters,body,any,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A JSON schema describing the function parameters. The model _should_ follow the schema,
however, this is not enforced at the moment.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].function.strict,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Not supported. Only maintained for compatibility reasons.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].type,body,string,true,,,,function,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources,body,array<SearchSource>,true,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].excluded_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"List of X handles to exclude from the search results. X posts returned will not include
any posts authored by these handles.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].included_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"NOTE: `included_x_handles` and `x_handles` are the same parameter.
`included_x_handles` is the new name but we keep both for backward compatibility.

X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].post_favorite_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum favorite count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].post_view_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum view count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].type,body,string,true,,,,x,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"DEPRECATED in favor of `included_x_handles`. Use `included_x_handles` instead.
X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].allowed_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to allow in the search results. This parameter act as a whitelist
where only those websites can be selected. A maximum of 5 websites can be selected.

Note 1: If no relevant information is found on those websites, the number of results
returned might be smaller than `max_search_results`.

Note 2: This parameter cannot be set with `excluded_websites`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.

Note 2: This parameter cannot be set with `allowed_websites`",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].type,body,string,true,,,,web,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].type,body,string,true,,,,news,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].links,body,array<string>,true,,,,,*,,DOCUMENTED,Links of the RSS feeds.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].sources[].type,body,string,true,,,,rss,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,tools[].type,body,string,true,,,,live_search,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,top_logprobs,body,integer | null,false,,0,8,,*,,DOCUMENTED|LIVE_VERIFIED,"An integer between 0 and 8 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used. Not supported by models `grok-4.20` and newer; the field will be silently ignored if set.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,top_p,body,number | null,false,1,,1,,*,,DOCUMENTED,"An alternative to sampling with `temperature`, called nucleus sampling, where the model considers the results of the tokens with `top_p` probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. It is generally recommended to alter this or `temperature` but not both.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,user,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,web_search_options,body,null | WebSearchOptions,false,,,,,*,,DOCUMENTED|RETIRED,,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,web_search_options.filters,body,any,false,,,,,*,,DOCUMENTED|RETIRED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,web_search_options.search_context_size,body,string | null,false,medium,,,,*,,DOCUMENTED|RETIRED,This field included for compatibility reason with OpenAI's API. It is mapped to `max_search`.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,web_search_options.user_location,body,any,false,,,,,*,,DOCUMENTED|RETIRED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,store,body,boolean,false,,,,,*,,LIVE_DISCOVERED,"OpenAI field; NOT in the xAI OpenAPI spec. Live: accepted silently (200), no stored-completion retrieval endpoints exist on xAI (GET /v1/chat/completions/{id} not offered).",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,metadata,body,object,false,,,,,*,,LIVE_DISCOVERED,OpenAI field; not in xAI spec. Live: silently accepted on chat completions (contrast: /v1/responses rejects it with 400).,https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/chat/completions,x-grok-conv-id,header,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Conversation/sticky-routing key for prompt caching (same effect as body prompt_cache_key).,https://docs.x.ai/developers/advanced-api-usage/prompt-caching/maximizing-cache-hits
xai,GET /v1/chat/deferred-completion/{request_id},request_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The deferred request id returned by a previous `deferred: true` chat request. 202 (empty body) while pending, 200 ChatResponse when done; result stays fetchable (a second GET also returned 200; docs say exactly once within 24 h).",https://docs.x.ai/developers/rest-api-reference/inference/chat-completions
xai,POST /v1/files,expires_after,multipart,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Optional TTL in seconds (measured from upload time). Must be between
3600 (1 hour) and 2592000 (30 days). If unset the file does not expire.

Accepts either a plain integer or the OpenAI SDK deepObject form
(`expires_after[anchor]=created_at` + `expires_after[seconds]=N`)
as separate multipart fields. The anchor+seconds form must arrive
before the `file` part.",https://docs.x.ai/developers/rest-api-reference/files/upload
xai,POST /v1/files,file,multipart,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The file to upload. The filename from the multipart
`Content-Disposition: filename=` header is recorded as the file's
`filename`.",https://docs.x.ai/developers/rest-api-reference/files/upload
xai,POST /v1/files,purpose,multipart,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Optional purpose label, accepted for OpenAI SDK compatibility. xAI
does not enforce or interpret this field. Setting `""assistants""`
is the conventional choice.",https://docs.x.ai/developers/rest-api-reference/files/upload
xai,GET /v1/files,limit,query,integer,false,100,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Max objects per page (1-100, default 100).",https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files,order,query,string,false,desc,,,asc|desc,*,,DOCUMENTED|LIVE_VERIFIED,asc | desc (default desc).,https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files,sort_by,query,string,false,created_at,,,created_at|filename|size,*,,DOCUMENTED|LIVE_VERIFIED,created_at | filename | size (default created_at).,https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files,pagination_token,query,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Token from the previous page. Live: null when the page is the last one.,https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files,after,query,string,false,,,,,*,,DOCUMENTED,Only included for OpenAI compatibility. Use pagination_token instead.,https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files,filter,query,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"AIP-160 filter (name/file_name, file_id, size_bytes, content_type, created_at, expires_at, upload_status, user_defined_id, public_url). Live: `public_url != null` matched the file with an active public URL.",https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"File id → File metadata (404 after delete: {code:'not-found', error:'File not found'}).",https://docs.x.ai/developers/rest-api-reference/files/manage
xai,DELETE /v1/files/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"File id → {id, deleted:true, object:'file'}.",https://docs.x.ai/developers/rest-api-reference/files/manage
xai,GET /v1/files/{file_id}/content,file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"File id → raw bytes (200, 200 bytes matched the upload).",https://docs.x.ai/developers/rest-api-reference/files/download
xai,GET /v1/files/{file_id}/content,format,query,string,false,,,,original|text,*,,DOCUMENTED|FAILED_VERIFICATION,original | text. Live: format=text on a text/plain file → 404 {error:'Failed to retrieve file'} (text extraction not available for that type).,https://docs.x.ai/developers/rest-api-reference/files/download
xai,POST /v1/files/{file_id}/public-url,file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,File id (content type must be png/jpeg/gif/webp/mp4/webm/pdf; text/plain → 400).,https://docs.x.ai/developers/files/public-urls
xai,POST /v1/files/{file_id}/public-url/revoke,file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"File id → {id, revoked:true, public_url} or {id, revoked:false} when nothing to revoke.",https://docs.x.ai/developers/files/public-urls
xai,POST /v1/files/{file_id}/public-url,expires_after,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Seconds from now until the public URL expires. Must be between `3600` (1
hour) and `2592000` (30 days). Omit to inherit the file's expiry (if it
has one) or to make the URL valid indefinitely.",https://docs.x.ai/developers/files/public-urls
xai,POST /v1/documents/search,filter,body,string | null,false,,,,,*,,DOCUMENTED,"Optional metadata filter string to apply to search results.
Uses AIP-160 filter syntax for querying document metadata.
Supports comparison operators: `=`, `!=`, `>`, `>=`, `<`, `<=`
Supports logical operators: `AND`, `OR`
Supports range syntax: `field:10..20` (inclusive)
Examples: `author = ""John""` or `year > 2020 AND category = ""finance""`",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by,body,null | GroupBy,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate,body,null | GroupByAggregate,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.min_k,body,object,true,,,,,*,,DOCUMENTED,Keep the k records with the smallest values (best scores).,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.min_k.k,body,integer,true,,,,,*,,DOCUMENTED,Number of records to keep per group. Must be >= 1.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.min_k.keys,body,array<string>,true,,,,,*,,DOCUMENTED,"Key(s) to sort by within each group. Use ""#score"" for relevance score.",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.max_k,body,object,true,,,,,*,,DOCUMENTED,Keep the k records with the largest values.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.max_k.k,body,integer,true,,,,,*,,DOCUMENTED,Number of records to keep per group. Must be >= 1.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.aggregate.max_k.keys,body,array<string>,true,,,,,*,,DOCUMENTED,Key(s) to sort by within each group.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,group_by.keys,body,array<string>,true,,,,,*,,DOCUMENTED,"Metadata key(s) to group results by (e.g. ""category"", ""department"").
At least one key is required.",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,instructions,body,string | null,false,,,,,*,,DOCUMENTED,User-defined instructions to be included in the search query. Defaults to generic search instructions.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,limit,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The number of chunks to return.
Will always return the top matching chunks.
Optional, defaults to 10.",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,query,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The query to search for which will be embedded using the
same embedding model as the one used for the source to query.",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,ranking_metric,body,null | RankingMetric,false,,,,RANKING_METRIC_UNKNOWN|RANKING_METRIC_L2_DISTANCE|RANKING_METRIC_COSINE_SIMILARITY,*,,DOCUMENTED|DEPRECATED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode,body,null | RetrievalMode,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker,body,null | HybridReranker,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.instructions,body,string | null,false,,,,,*,,DOCUMENTED,Instructions for the reranking model. Defaults to generic reranking instructions.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.model,body,string | null,false,,,,,*,,DOCUMENTED,The model to use for reranking. Defaults to standard reranker model.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.type,body,string,true,,,,reranker_model,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.embedding_weight,body,number,false,,,,,*,,DOCUMENTED,Weight for embedding (dense) search results. Should be between 0 and 1. Defaults to 0.5.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.k,body,integer | null,false,,,,,*,,DOCUMENTED,The RRF constant k used in the reciprocal rank fusion formula. Defaults to 60.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.text_weight,body,number,false,,,,,*,,DOCUMENTED,Weight for keyword (sparse) search results. Should be between 0 and 1. Defaults to 0.5.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.type,body,string,true,,,,rrf,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.search_multiplier,body,integer | null,false,,,,,*,,DOCUMENTED,"Additional multiplier to requested search limit.
Valid range is [1, 100]. Defaults to 1 when unset.",https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.type,body,string,true,,,,hybrid,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker,body,null | RerankerModel,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.instructions,body,string | null,false,,,,,*,,DOCUMENTED,Instructions for the reranking model. Defaults to generic reranking instructions.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker.model,body,string | null,false,,,,,*,,DOCUMENTED,The model to use for reranking. Defaults to standard reranker model.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.type,body,string,true,,,,semantic,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.reranker,body,null | RerankerModel,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,retrieval_mode.type,body,string,true,,,,keyword,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,source,body,DocumentsSource,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The source to query.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,source.collection_ids,body,array<string>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The collection IDs to search in.,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/documents/search,source.rag_pipeline,body,null | RagPipeline,false,,,,chroma_db|es,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/collections/search
xai,POST /v1/collections,collection_name,body,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Name of the new collection.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,team_id,body,string,false,,,,,*,,DOCUMENTED,Owning team; derived from credentials if omitted.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,collection_description,body,string,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Human-friendly description.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,index_configuration.model_name,body,string,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Embedding model (live default echoed: grok-embedding-small).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration,body,object,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,"One of chars_configuration | tokens_configuration | markdown_tokens_configuration | markdown_chars_configuration | code_tokens_configuration | code_chars_configuration | table_configuration | bytes_configuration | ast_configuration (deprecated) + strip_whitespace, inject_name_into_chunks. Live default echoed: {config:{BytesConfiguration:{max_chunk_size_bytes:4000, chunk_overlap_bytes:800}}, strip_whitespace:true, inject_name_into_chunks:false}.",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.tokens_configuration.max_chunk_size_tokens,body,integer,false,,,,,*,,DOCUMENTED,Max chunk length in tokens.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.tokens_configuration.chunk_overlap_tokens,body,integer,false,,,,,*,,DOCUMENTED,Overlap in tokens (both sides).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.tokens_configuration.encoding_name,body,string,false,,,,,*,,DOCUMENTED,"Tokenizer encoding, e.g. o200k_base.",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.chars_configuration.max_chunk_size_chars,body,integer,false,,,,,*,,DOCUMENTED,Max chunk length in characters.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.chars_configuration.chunk_overlap_chars,body,integer,false,,,,,*,,DOCUMENTED,Overlap in characters.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.bytes_configuration.max_chunk_size_bytes,body,integer,false,,,,,*,,DOCUMENTED,Max chunk length in bytes.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.bytes_configuration.chunk_overlap_bytes,body,integer,false,,,,,*,,DOCUMENTED,Overlap in bytes.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.strip_whitespace,body,boolean,false,,,,,*,,DOCUMENTED,Remove leading/trailing whitespace.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,chunk_configuration.inject_name_into_chunks,body,boolean,false,,,,,*,,DOCUMENTED,Inject document name into chunks.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,metric_space,body,string,false,,,,HNSW_METRIC_UNKNOWN|HNSW_METRIC_COSINE|HNSW_METRIC_EUCLIDEAN|HNSW_METRIC_INNER_PRODUCT,*,,DOCUMENTED,HNSW distance space.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,version,body,integer,false,,,,,*,,DOCUMENTED,Internal only.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions,body,array<object>,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,"Metadata field definitions. Live (api.x.ai): required; may be []. Each item must carry key, required, inject_into_chunk, unique (422 'missing field inject_into_chunk' otherwise) and optionally description.",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions[].key,body,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Field key.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions[].required,body,boolean,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Required for every document.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions[].inject_into_chunk,body,boolean,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Prepend value to each chunk (contextual retrieval).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions[].unique,body,boolean,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Unique across documents.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections,field_definitions[].description,body,string,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Optional description.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,limit,query,integer,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Max 100 (default 100).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,order,query,string,false,,,,ORDERING_UNKNOWN|ORDERING_ASCENDING|ORDERING_DESCENDING,*,,DOCUMENTED,Sort order.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,sort_by,query,string,false,,,,COLLECTIONS_SORT_BY_NAME|COLLECTIONS_SORT_BY_AGE,*,,DOCUMENTED,Sort field.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,pagination_token,query,string,false,,,,,*,,DOCUMENTED,Next-page token.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,filter,query,string,false,,,,,*,,DOCUMENTED,"Filter on collection_id, collection_name (partial), created_at, documents_count.",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections,team_id,query,string,false,,,,,*,,DOCUMENTED,Owning team.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,DELETE /v1/collections/{collection_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id → 200 {}.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|FAILED_VERIFICATION,Collection id (path). Live (api.x.ai): body must ALSO contain collection_id and field_definition_updates (422 otherwise).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},collection_name,body,string,false,,,,,*,,DOCUMENTED,New name.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},collection_description,body,string,false,,,,,*,,DOCUMENTED,New description.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},chunk_configuration,body,object,false,,,,,*,,DOCUMENTED,Same shape as on create.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},field_definition_updates,body,array<object>,false,,,,,*,,DOCUMENTED,"[{field_definition:{key,…}, operation: FIELD_DEFINITION_ADD|FIELD_DEFINITION_DELETE}]. Live: required on api.x.ai (may be []).",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PUT /v1/collections/{collection_id},field_definition_updates[].operation,body,string,false,,,,FIELD_DEFINITION_ADD|FIELD_DEFINITION_DELETE,*,,DOCUMENTED,Add/update or delete (cascade) a field definition.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections/{collection_id}/documents/{file_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id (path). Live (api.x.ai): the JSON body must repeat collection_id AND file_id (gRPC-gateway transcoding) — 422 'missing field collection_id/file_id' otherwise.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections/{collection_id}/documents/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Files-API file id to attach.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections/{collection_id}/documents/{file_id},fields,body,object,false,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,User-defined metadata fields (must match field_definitions).,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/collections/{collection_id}/documents/{file_id},team_id,body,string,false,,,,,*,,DOCUMENTED,Owning team.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,limit,query,integer,false,,,,,*,,DOCUMENTED,Max 100.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,order,query,string,false,,,,ORDERING_UNKNOWN|ORDERING_ASCENDING|ORDERING_DESCENDING,*,,DOCUMENTED,Sort order.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,sort_by,query,string,false,,,,DOCUMENTS_SORT_BY_NAME|DOCUMENTS_SORT_BY_SIZE|DOCUMENTS_SORT_BY_AGE,*,,DOCUMENTED,Sort field.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,pagination_token,query,string,false,,,,,*,,DOCUMENTED,Next-page token.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,filter,query,string,false,,,,,*,,DOCUMENTED,"Filter on name, content_type, size_bytes, created_at, status, fields.{key}.",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents,name,query,string,false,,,,,*,,DOCUMENTED|DEPRECATED,DEPRECATED: use filter name:value.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents/{file_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Document (file) id. Live: `status` came back as the proto enum NUMBER 1 (= DOCUMENT_STATUS_PROCESSING) instead of the documented string.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PATCH /v1/collections/{collection_id}/documents/{file_id},collection_id,path,string,true,,,,,*,,DOCUMENTED,Regenerate indices for the document.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,PATCH /v1/collections/{collection_id}/documents/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED,Document id.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,DELETE /v1/collections/{collection_id}/documents/{file_id},collection_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Collection id. Live: 200 {} and the underlying Files-API file was DELETED too (GET /v1/files/{id} → 404 afterwards) — even when the document had never been added.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,DELETE /v1/collections/{collection_id}/documents/{file_id},file_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_DISCOVERED|LIVE_VERIFIED,Document id.,https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,GET /v1/collections/{collection_id}/documents:batchGet,file_ids,query,array<string>,true,,,,,*,,DOCUMENTED|FAILED_VERIFICATION,"Documented as required query array. Live: `file_ids=<id>` → 400 'invalid type: string, expected a sequence'; `file_ids[]=<id>` → 400 'missing field file_ids' (query encoding unresolved).",https://docs.x.ai/developers/rest-api-reference/collections/collection
xai,POST /v1/batches,name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Human-readable batch name.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches,input_file_id,body,string,false,,,,,*,,DOCUMENTED,"Files-API id of a JSONL file (custom_id/method/url/body per line, ≤200 MB, ≤50,000 requests). File-based batches are sealed (no AddBatchRequests).",https://docs.x.ai/developers/advanced-api-usage/batch-api
xai,GET /v1/batches,limit,query,integer,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Page size.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches,pagination_token,query,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Next-page token (null on last page).,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id},batch_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Batch id.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Batch id.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests,body,array<object>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Requests to add (each payload ≤25 MB; ≤1000 add calls / 30 s / team).,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request_id,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Caller id, unique within the batch (UUID generated if omitted).",https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request,body,object,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Exactly one of chat_get_completion | responses | image_generation | image_edit | video_generation | video_extension.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.chat_get_completion,body,ChatRequest,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Full /v1/chat/completions body (max 128 functions per docs).,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.responses,body,ModelRequest,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Full /v1/responses body (server-side tools allowed). Live: accepted, but the result came back under `chat_get_completion` (chat.completion object) and the request metadata endpoint reported `xai_api.Chat/GetCompletion`.",https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.image_generation,body,GenerateImageRequest,false,,,,,*,,DOCUMENTED,Image generation body (see images agent).,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.image_edit,body,EditImageRequest,false,,,,,*,,DOCUMENTED,Image edit body.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.video_generation,body,GenerateVideoRequest,false,,,,,*,,DOCUMENTED,Video generation/edit body.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}/requests,batch_requests[].batch_request.video_extension,body,ExtendVideoRequest,false,,,,,*,,DOCUMENTED,Video extension body.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/requests,batch_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Batch id.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/requests,limit,query,integer,false,,,,,*,,DOCUMENTED,Max 1000.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/requests,pagination_token,query,string,false,,,,,*,,DOCUMENTED,Next-page token.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/results,batch_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Batch id.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/results,limit,query,integer,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Max 1000.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,GET /v1/batches/{batch_id}/results,pagination_token,query,string,false,,,,,*,,DOCUMENTED,Next-page token.,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/batches/{batch_id}:cancel,batch_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Batch id → Batch with cancel_time set (works even after completion).,https://docs.x.ai/developers/rest-api-reference/inference/batches
xai,POST /v1/images/generations,model,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"Image model id or alias (grok-imagine-image, grok-imagine-image-2.0, grok-imagine-image-quality → served by 2.0/low from 2026-11-02). Effectively required.",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,prompt,body,string,true,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"Prompt (max_prompt_length: 16000 chars on grok-imagine-image / -quality, 64000 on 2.0 per the models endpoint). Required.",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,n,body,integer|null,false,1,1,10,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Number of images (1–10).,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,response_format,body,string|null,false,url,,,url|b64_json,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"`url` (ephemeral hosted URL) or `b64_json` (bare base64, no data-URI prefix).",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,resolution,body,string|null,false,1k,,,1k|1.5k|2k,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Output resolution; grok-imagine models only. Pricing on 2.0 rises with resolution (1k < 1.5k < 2k).,https://docs.x.ai/developers/model-capabilities/images/generation
xai,POST /v1/images/generations,quality,body,string|null,false,auto,,,low|medium|auto,grok-imagine-image-2.0,,DOCUMENTED|LIVE_VERIFIED,"Quality tier for grok-imagine-image-2.0: low, medium or auto (default; currently low for generation, medium for editing). Billed at the quality served.",https://docs.x.ai/developers/model-capabilities/images/generation
xai,POST /v1/images/generations,user,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,Opaque end-user id for abuse monitoring.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,storage_options,body,object,false,,,,,,,DOCUMENTED,Persist the output to the Files API and return `file_output` alongside the ephemeral URL/b64.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,storage_options.filename,body,string,true,,,,,,,DOCUMENTED,Filename for the stored file (required when storage_options is present).,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,storage_options.expires_after,body,integer|null,false,,,2592000,,,,DOCUMENTED,Seconds until the stored file auto-expires; max 2592000 (30 days); omitted = no expiry.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,storage_options.public_url,body,boolean|object,false,,,,,,,DOCUMENTED,Also create a permanent shareable public URL (`file_output.public_url`); object form accepts an `expires_after`.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/generations,aspect_ratio,body,string|null,false,auto,,,1:1|3:4|4:3|9:16|16:9|2:3|3:2|9:19.5|19.5:9|9:20|20:9|1:2|2:1|21:9|5:2|auto,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Aspect ratio; `auto` (default) lets the model pick. 21:9 and 5:2 added 2026-08.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,model,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"Image model id or alias (grok-imagine-image, grok-imagine-image-2.0, grok-imagine-image-quality → served by 2.0/low from 2026-11-02). Effectively required.",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,prompt,body,string,true,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"Prompt (max_prompt_length: 16000 chars on grok-imagine-image / -quality, 64000 on 2.0 per the models endpoint). Required.",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,n,body,integer|null,false,1,1,10,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Number of images (1–10).,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,response_format,body,string|null,false,url,,,url|b64_json,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,"`url` (ephemeral hosted URL) or `b64_json` (bare base64, no data-URI prefix).",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,resolution,body,string|null,false,1k,,,1k|1.5k|2k,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Output resolution; grok-imagine models only. Pricing on 2.0 rises with resolution (1k < 1.5k < 2k).,https://docs.x.ai/developers/model-capabilities/images/generation
xai,POST /v1/images/edits,quality,body,string|null,false,auto,,,low|medium|auto,grok-imagine-image-2.0,,DOCUMENTED|LIVE_VERIFIED,"Quality tier for grok-imagine-image-2.0: low, medium or auto (default; currently low for generation, medium for editing). Billed at the quality served.",https://docs.x.ai/developers/model-capabilities/images/generation
xai,POST /v1/images/edits,user,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,Opaque end-user id for abuse monitoring.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,storage_options,body,object,false,,,,,,,DOCUMENTED,Persist the output to the Files API and return `file_output` alongside the ephemeral URL/b64.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,storage_options.filename,body,string,true,,,,,,,DOCUMENTED,Filename for the stored file (required when storage_options is present).,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,storage_options.expires_after,body,integer|null,false,,,2592000,,,,DOCUMENTED,Seconds until the stored file auto-expires; max 2592000 (30 days); omitted = no expiry.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,storage_options.public_url,body,boolean|object,false,,,,,,,DOCUMENTED,Also create a permanent shareable public URL (`file_output.public_url`); object form accepts an `expires_after`.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,aspect_ratio,body,string|null,false,,,,1:1|3:4|4:3|9:16|16:9|2:3|3:2|9:19.5|19.5:9|9:20|20:9|1:2|2:1|21:9|5:2|auto,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,"Only for multi-image edits (`images`): output ratio, otherwise auto-detected from the first input. Do not set for single-image edits.",https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,image,body,object|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Single source image. Mutually exclusive with `images`.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,image.url,body,string,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED|LIVE_VERIFIED,Public URL or base64 data URL (JPEG/PNG/WebP). Also accepted as `image_url`. Required when file_id is absent. The docs examples add `type: 'image_url'` (ignored/optional).,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,image.file_id,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,Files API file id (must be a fully uploaded image). Mutually exclusive with url.,https://docs.x.ai/developers/model-capabilities/imagine/files/inputs
xai,POST /v1/images/edits,images,body,array<object>,false,,,5,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,"2–5 reference images (was 3 before 2026-08); refer to them as <IMAGE_0>, <IMAGE_1>… in the prompt. Each item {url | file_id}. Mutually exclusive with `image`.",https://docs.x.ai/developers/model-capabilities/images/multi-image-editing
xai,POST /v1/images/edits,images[].url,body,string,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,Public URL or base64 data URL.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,POST /v1/images/edits,images[].file_id,body,string|null,false,,,,,grok-imagine-image|grok-imagine-image-2.0|grok-imagine-image-quality,,DOCUMENTED,Files API file id.,https://docs.x.ai/developers/rest-api-reference/inference/images
xai,GET /v1/image-generation-models/{model_id},model_id,path,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Model id or alias.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,POST /v1/completions,best_of,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,(Unsupported) Generates multiple completions internally and returns the top-scoring one. Not functional yet.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,echo,body,boolean | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Option to include the original prompt in the response along with the generated completion.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,frequency_penalty,body,number | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"(Unsupported) Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,logit_bias,body,object | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"(Unsupported) Accepts a JSON object that maps tokens to an associated bias value from -100 to 100. You can use this tokenizer tool to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,logprobs,body,boolean | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"Include the log probabilities on the `logprobs` most likely output tokens, as well the chosen tokens. For example, if `logprobs` is 5, the API will return a list of the 5 most likely tokens. The API will always return the logprob of the sampled token, so there may be up to `logprobs+1` elements in the response. Not supported by models `grok-4.20` and newer; the field will be silently ignored if set.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,max_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Limits the number of tokens that can be produced in the output. Ensure the sum of prompt tokens and `max_tokens` does not exceed the model's context limit.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,model,body,string,true,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Specifies the model to be used for the request.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,n,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Determines how many completion sequences to produce for each prompt. Be cautious with its use due to high token consumption; adjust `max_tokens` and stop sequences accordingly.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,presence_penalty,body,number | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"(Not supported by `grok-3` and reasoning models) Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,prompt,body,SampleContent,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"Input for generating completions, which can be a string, list of strings, token list, or list of token lists. `<|endoftext|>` is used as a document separator, implying a new context start if omitted.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,seed,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result. Determinism is not guaranteed, and you should refer to the system_fingerprint response parameter to monitor changes in the backend.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,stop,body,array | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,(Not supported by reasoning models) Up to 4 sequences where the API will stop generating further tokens.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,stream,body,boolean | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"Whether to stream back partial progress. If set, tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,stream_options,body,null | StreamOptions,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,stream_options.include_usage,body,boolean,true,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Set an additional chunk to be streamed before the `data: [DONE]` message. The other chunks will return `null` in `usage` field.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,suffix,body,string | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,(Unsupported) Optional string to append after the generated text.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED|LEGACY|RETIRED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or `top_p` but not both.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,top_p,body,number | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with `top_p` probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/completions,user,body,string | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,max_tokens_to_sample,body,integer,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,The maximum number of tokens to generate before stopping.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,metadata,body,null | MessageMetadata,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,metadata.user_id,body,string | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,model,body,string,true,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Model to use for completion.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,prompt,body,string,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,Prompt for the model to perform completion on.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,stop_sequences,body,array | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,(Not supported by reasoning models) Up to 4 sequences where the API will stop generating further tokens.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,stream,body,boolean | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"(Unsupported) If set, partial message deltas will be sent. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED|LEGACY|RETIRED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,top_k,body,integer | null,false,,,,,*,,DOCUMENTED|LEGACY|RETIRED,"(Unsupported) When generating next tokens, randomly selecting the next token from the k most likely options.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/complete,top_p,body,number | null,false,1,,1,,*,,DOCUMENTED|LEGACY|RETIRED,"An alternative to sampling with `temperature`, called nucleus sampling, where the model considers the results of the tokens with `top_p` probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. It is generally recommended to alter this or `temperature` but not both.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/embeddings,dimensions,body,integer | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,The number of dimensions the resulting output embeddings should have.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,encoding_format,body,string | null,false,,,,float|base64,*,,DOCUMENTED|ACCOUNT_RESTRICTED,The format to return the embeddings in. Can be either `float` or `base64`.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,input,body,EmbeddingInput,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,"Input text to embed, encoded as a string or list of tokens. To embed multiple inputs in a single request, pass a list of strings or list of token lists. The total tokens in the input must not exceed the context window for the model and cannot be an empty string. If the input is a list, the maximum length of the list is 128.",https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,input.String,body,string,true,,,,,*,,DOCUMENTED,"A strings to be embedded. For best performance, prepend ""query: "" in front of query content and prepend ""passage: "" in front of passage/text",https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,input.StringArray,body,array<string>,true,,,,,*,,DOCUMENTED,An array of strings to be embedded,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,input.Ints,body,array<integer>,true,,,,,*,,DOCUMENTED,A token in integer to be embedded,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,input.IntsArray,body,array<array<integer>>,true,,,,,*,,DOCUMENTED,An array of tokens in integers to be embedded,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,model,body,string,true,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,ID of the model to use.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,preview,body,boolean | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,Flag to use the new format of the API.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/embeddings,user,body,string | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/tokenize-text,model,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The model to tokenize with.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/tokenize-text,text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text content to be tokenized.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/tokenize-text,user,body,string | null,false,,,,,*,,DOCUMENTED,Optional user identifier.,https://docs.x.ai/developers/rest-api-reference/inference/other
xai,POST /v1/messages,max_tokens,body,integer,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,The maximum number of tokens to generate before stopping. The model may stop before the max_tokens when it reaches the stop sequence.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages,body,array<MessageBody>,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Input messages.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content,body,MessageContent,true,,,,,*,,DOCUMENTED,The content message.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].text,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Text prompt.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,text,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source,body,MessageImageContent,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Image source.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source.data,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Base64 encoded image string.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source.media_type,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"Media type of the image source. Available options: `image/jpeg`, `image/png`, `image/webp`.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source.type,body,string,true,,,,base64,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source.type,body,string,true,,,,url,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].source.url,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,URL of the image.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,image,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].id,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,ID of the tool call.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].input,body,any,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Input for tool call.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].name,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Name of the tool call.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,tool_use,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content,body,ToolResultContent,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Result content of the tool call. Can be a string or an array of content blocks.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].text,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,The text content.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].type,body,string,true,,,,text,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source,body,MessageImageContent,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Image source.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source.data,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Base64 encoded image string.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source.media_type,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"Media type of the image source. Available options: `image/jpeg`, `image/png`, `image/webp`.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source.type,body,string,true,,,,base64,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source.type,body,string,true,,,,url,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].source.url,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,URL of the image.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].content[].type,body,string,true,,,,image,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].is_error,body,boolean | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Whether the tool call returns an error.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].tool_use_id,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,ID of the tool call given by the model.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,tool_result,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].data,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED,Encrypted data of the redacted thinking.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,redacted_thinking,*,,DOCUMENTED|DEPRECATED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].signature,body,string | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Signature of the thinking block (required by Anthropic SDK for round-tripping).,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].thinking,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Thinking.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].content[].type,body,string,true,,,,thinking,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,messages[].role,body,string,true,,,,,*,,DOCUMENTED,"The role that the message belongs to, `""system""` for system prompt, `""user""` for user prompt, and `""assistant""` for response from the model.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,metadata,body,null | MessageMetadata,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,metadata.user_id,body,string | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,model,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Model name for the model to use.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,stop_sequences,body,array | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Not supported by reasoning models) Up to 4 sequences where the API will stop generating further tokens.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,stream,body,boolean | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"If set, partial message deltas will be sent. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,system,body,null | SystemMessageContent,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,system[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,system[].text,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,System prompt text.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,system[].type,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"Type of the object. This is always `""text""`.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. It may not work well with reasoning models.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tool_choice,body,null | MessageToolChoice,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tool_choice.type,body,string,true,,,,auto,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tool_choice.type,body,string,true,,,,any,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tool_choice.name,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Name of the tool to use.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tool_choice.type,body,string,true,,,,tool,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools,body,array | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"A list of tools the model may call in JSON-schema. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 350 functions are supported.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].cache_control,body,any,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,(Unsupported) Cache control.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].description,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Description of the tool.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].input_schema,body,MessageToolInputSchema,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Input schema allowed by the tool.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].input_schema.properties,body,any,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,JSON-object of the tool input schema.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].input_schema.required,body,array | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"Required properties of the tool input schema, if any.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].input_schema.type,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"Type of the schema. This is always `""object""`.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,tools[].name,body,string,true,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,Name of the tool.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,top_k,body,integer | null,false,,,,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"(Unsupported) When generating next tokens, randomly selecting the next token from the k most likely options.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,top_p,body,number | null,false,1,,1,,*,,DOCUMENTED|DEPRECATED|LIVE_VERIFIED,"An alternative to sampling with `temperature`, called nucleus sampling, where the model considers the results of the tokens with `top_p` probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. It is generally recommended to alter this or `temperature` but not both.",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,thinking,body,object,false,,,,,*,,LIVE_DISCOVERED,"Anthropic `thinking` config. NOT in the xAI spec. Live: {type:'enabled', budget_tokens:1024} accepted silently; no thinking block appeared in the non-streamed text answer (thinking blocks appear when a tool_use is emitted and in streams regardless of this field).",https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,service_tier,body,string,false,,,,,*,,LIVE_DISCOVERED,Anthropic field; not in the xAI spec; silently accepted (unknown fields are ignored on /v1/messages).,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/messages,anthropic-version,header,string,false,,,,,*,,LIVE_VERIFIED,Not required by xAI; sending anthropic-version + x-api-key alongside Authorization: Bearer works (200). Auth is Bearer only.,https://docs.x.ai/developers/rest-api-reference/inference/legacy
xai,POST /v1/responses,background,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,(Unsupported) Whether to process the response asynchronously in the background.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,context_management,body,array | null,false,,,,,*,,DOCUMENTED,Optional context-management directives (e.g. compaction). Parsed but not yet executed.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,include,body,array | null,false,,,,reasoning.encrypted_content|web_search_call.action.sources|code_interpreter_call.outputs|file_search_call.results|no_inline_citations|message.output_text.logprobs (ignored),*,,DOCUMENTED|LIVE_VERIFIED,"What additional output data to include in the response. Supported values include
`reasoning.encrypted_content` (encrypted reasoning tokens) and tool-output options.
OpenAI's `message.output_text.logprobs` is accepted for compatibility but silently ignored.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input,body,ModelInput,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The input passed to the model. Can be text, image or file.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content,body,ModelInputContent,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image or audio input.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Text input.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,output_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].image_url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A public URL of image prompt, only available for vision models.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,input_image,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].file_data,body,string | null,false,,,,,*,,DOCUMENTED,"Inline file bytes (base64 encoded).

When set, the file content is provided directly in the request and does NOT need
to be uploaded to the Files API first.

Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The file ID from the Files API.

Set this to reference a previously-uploaded file.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].file_url,body,string | null,false,,,,,*,,DOCUMENTED,"Public URL for a file attachment.

When set, the file content is fetched from this URL.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].filename,body,string | null,false,,,,,*,,DOCUMENTED,"Filename for inline uploads.

Required when `file_data` is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].mime_type,body,string | null,false,,,,,*,,DOCUMENTED,"Optional MIME type for inline uploads (e.g. ""application/pdf"").",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,input_file,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].name,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse. Only supported for `user` messages.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].role,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the message. Possible values are `user`, `assistant`, `system` and `developer`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the message, which is always `message`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the function tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output,body,ModelInputContent,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The output of the function tool call. Can be a plain string or a list
of content items (`input_text`, `input_image`, `input_file`).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Text input.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].type,body,string,true,,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].type,body,string,true,,,,output_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].image_url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A public URL of image prompt, only available for vision models.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].type,body,string,true,,,,input_image,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].file_data,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Inline file bytes (base64 encoded).

When set, the file content is provided directly in the request and does NOT need
to be uploaded to the Files API first.

Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The file ID from the Files API.

Set this to reference a previously-uploaded file.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].file_url,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Public URL for a file attachment.

When set, the file content is fetched from this URL.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].filename,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Filename for inline uploads.

Required when `file_data` is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].mime_type,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Optional MIME type for inline uploads (e.g. ""application/pdf"").",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].type,body,string,true,,,,input_file,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the function tool call, which is always `function_call_output`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the shell tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The maximum length of the shell command output. Generated by the model
and should be passed back with the raw output.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output,body,array<ShellCallOutputResult>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An array of shell call output contents.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].outcome,body,ShellCallOutcome,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The outcome of the command execution.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].outcome.exit_code,body,integer,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The exit code of the command.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].outcome.type,body,string,true,,,,exit,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].outcome.type,body,string,true,,,,timeout,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].stderr,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The standard error output that was captured.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output[].stdout,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The standard output that was captured.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the output item, which is always `shell_call_output`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].encrypted_content,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The encrypted content of the compacted conversation.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the compaction item (e.g. `cmp_<uuid>`).,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the item. Always `""compaction""`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,instructions,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"An alternate way to specify the system prompt. Note that this cannot be used alongside `previous_response_id`, where the system prompt of the previous message will be used.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,logprobs,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message. Not supported by models `grok-4.20` and newer; the field will be silently ignored if set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,max_output_tokens,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Max number of tokens that can be generated in a response. This includes both output and reasoning tokens. Defaults to 128,000 when unset; set a larger value to allow longer generations.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,max_turns,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Maximum number of agentic tool calling turns allowed for this request.
If not set, defaults to the server's global cap.
This parameter will be ignored for any non-agentic requests, and for
agentic SLOP requests that have neither a server-side tool nor a file
attachment.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,metadata,body,any,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Not supported. Only maintained for compatibility reasons.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,min_p,body,number | null,false,,0,1,,*,,DOCUMENTED|LIVE_VERIFIED,Min-p sampling: tokens whose probability is below `min_p` times the probability of the most likely token are excluded from sampling. Disabled when unset.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,model,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Model name for the model to use. Obtainable from <https://console.x.ai/team/default/models> or <https://docs.x.ai/docs/models>.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,parallel_tool_calls,body,boolean | null,false,true,,,,*,,DOCUMENTED,Whether to allow the model to run parallel tool calls.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,previous_response_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The ID of the previous response from the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,prompt_cache_key,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Plumbed to x-grok-conv-id for Open Responses compatibility, used for routing.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,reasoning,body,null | ReasoningConfiguration,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,reasoning.effort,body,string | null,false,,,,low|medium|high|xhigh,grok-4.6|grok-4.5|grok-4.3|grok-4.20-0309-reasoning|grok-4.20-multi-agent-0309|grok-build-0.1,,DOCUMENTED|LIVE_VERIFIED,Constrains how hard a reasoning model thinks before responding. Higher efforts use more reasoning tokens for deeper thinking. The supported values and the default depend on the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,reasoning.generate_summary,body,string | null,false,,,,,*,,DOCUMENTED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,reasoning.summary,body,string | null,false,,,,auto|concise|detailed,*,,DOCUMENTED|LIVE_VERIFIED,"A summary of the model's reasoning process. Possible values are `auto`, `concise` and `detailed`. Only included for compatibility. The model shall always return `detailed`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,reasoning_effort,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,reasoning_effort alternative to reasoning configuration. This is a non-standard field meant to ease user experience. We only look at this if the reasoning field is unset.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,safety_identifier,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Supplied by the API client to identify the end user behind this request. A stable string that uniquely identifies each of your users; hash your internal user id or username rather than sending an email or name. Stored with the request metadata so a usage-policy violation can be attributed to that user rather than to the API key.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters,body,null | SearchParameters,false,,,,,*,,DOCUMENTED|RETIRED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.from_date,body,string | null,false,,,,,*,,DOCUMENTED|RETIRED,"Date from which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.max_search_results,body,integer | null,false,15,1,30,,*,,DOCUMENTED|RETIRED,Maximum number of search results to use.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.mode,body,string | null,false,auto,,,,*,,DOCUMENTED|RETIRED,"Choose the mode to query realtime data:
* `off`: no search performed and no external will be considered.
* `on` (default): the model will search in every sources for relevant data.
* `auto`: the model choose whether to search data or not and where to search the data.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.return_citations,body,boolean | null,false,true,,,,*,,DOCUMENTED|RETIRED,Whether to return citations in the response or not.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources,body,array | null,false,,,,,*,,DOCUMENTED|RETIRED,"List of sources to search in. If no sources specified, the model will look over the web and X by default.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].excluded_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"List of X handles to exclude from the search results. X posts returned will not include
any posts authored by these handles.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].included_x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"NOTE: `included_x_handles` and `x_handles` are the same parameter.
`included_x_handles` is the new name but we keep both for backward compatibility.

X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].post_favorite_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum favorite count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].post_view_count,body,integer | null,false,,,,,*,,DOCUMENTED,The minimum view count of the X posts to consider.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].type,body,string,true,,,,x,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].x_handles,body,array | null,false,,,,,*,,DOCUMENTED,"DEPRECATED in favor of `included_x_handles`. Use `included_x_handles` instead.
X Handles of the users from whom to consider the posts. Only available if mode is `auto`, `on` or `x`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].allowed_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to allow in the search results. This parameter act as a whitelist
where only those websites can be selected. A maximum of 5 websites can be selected.

Note 1: If no relevant information is found on those websites, the number of results
returned might be smaller than `max_search_results`.

Note 2: This parameter cannot be set with `excluded_websites`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.

Note 2: This parameter cannot be set with `allowed_websites`",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].type,body,string,true,,,,web,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].country,body,string | null,false,,,,,*,,DOCUMENTED,"ISO alpha-2 code of the country. If the country is set, only data coming from this country
will be considered. See <https://en.wikipedia.org/wiki/ISO_3166-2>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].excluded_websites,body,array | null,false,,,,,*,,DOCUMENTED,"List of website to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].safe_search,body,boolean | null,false,,,,,*,,DOCUMENTED,"If set to true, mature content won't be considered during the search. Default to `true`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].type,body,string,true,,,,news,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].links,body,array<string>,true,,,,,*,,DOCUMENTED,Links of the RSS feeds.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.sources[].type,body,string,true,,,,rss,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,search_parameters.to_date,body,string | null,false,,,,,*,,DOCUMENTED|RETIRED,"Date up to which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,service_tier,body,null | ServiceTier,false,,,,default|priority,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,store,body,boolean | null,false,true,,,,*,,DOCUMENTED|LIVE_VERIFIED,Whether to store the input message(s) and model response for later retrieval.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,stream,body,boolean | null,false,false,,,,*,,DOCUMENTED|LIVE_VERIFIED,"If set, partial message deltas will be sent. Tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,temperature,body,number | null,false,1,0,2,,*,,DOCUMENTED,"What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text,body,null | ModelResponseConfiguration,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format,body,null | ModelResponseFormat,false,,,,text|json_object|json_schema,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.type,body,string,true,,,,text,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.type,body,string,true,,,,json_object,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.description,body,string | null,false,,,,,*,,DOCUMENTED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.name,body,string | null,false,,,,,*,,DOCUMENTED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.schema,body,any,true,,,,,*,,DOCUMENTED,A json schema representing the desired response schema.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.strict,body,boolean | null,false,,,,,*,,DOCUMENTED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,text.format.type,body,string,true,,,,json_schema,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tool_choice,body,null | ModelToolChoice,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tool_choice.name,body,string,true,,,,,*,,DOCUMENTED,Name of the function to use.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tool_choice.type,body,string,true,,,,,*,,DOCUMENTED,"Type is always `""function""`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A list of tools the model may call in JSON-schema. Currently, only functions and web search are supported as tools. A max of 350 tools are supported.`web_search_preview` tool, if specified, will be overridden by `search_parameters`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].defer_loading,body,boolean | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,"When true, the definition is hidden from the model's prompt but stays
callable, loaded via a `tool_search` step (`/v1/responses` only).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].description,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A description of the function to indicate to the model when to call it.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The name of the function. If the model calls the function, this name is used in the
response.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].parameters,body,any,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A JSON schema describing the function parameters. The model _should_ follow the schema,
however, this is not enforced at the moment.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].strict,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Not supported. Only maintained for compatibility reasons.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,function,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].filters,body,null | WebSearchFilters,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].filters.allowed_domains,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of website domains (without protocol specification or subdomains)
to restrict search results to (e.g., [""example.com""]). A maximum of 5 websites can be allowed.
Use this as a whitelist to limit results to only these specific sites; no other websites will
be considered. If no relevant information is found on these websites, the number of results
returned might be smaller than `max_search_results` set in `SearchParameters`. Note: This
parameter cannot be set together with `excluded_domains`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].filters.excluded_domains,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of website domains (without protocol specification or subdomains) to exclude from search results (e.g., [""example.com""]).
Use this to prevent results from unwanted sites. A maximum of 5 websites can be excluded.
This parameter cannot be set together with `allowed_domains`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].search_context_size,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"High level guidance for the amount of context window space to use for the
search. Available values are `low`, `medium`, or `high`.
For OpenAI API compatibility ONLY. Request will be rejected if this field is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location,body,null | WebSearchUserLocation,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location.city,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,City of the user's location.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location.country,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Two-letter ISO 3166-1 alpha-2 country code, like US, GB, etc.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location.region,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Region of the user's location.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location.timezone,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Timezone of the user's location, IANA timezone like America/Chicago, Europe/London, etc.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].user_location.type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Type is always `""approximate""`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].allowed_domains,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of website domains to allow in the search results. This parameter act as a whitelist
where only those websites can be selected. A maximum of 5 websites can be selected.

Note: This parameter cannot be set with `excluded_domains`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].enable_image_search,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"When true, activates the image search server-side tool, enabling
the model to search for images alongside the standard web search.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].enable_image_understanding,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Enable image understanding during web search.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].excluded_domains,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of website domains to exclude from the search results without protocol specification or
subdomains. A maximum of 5 websites can be excluded.

Note: This parameter cannot be set with `allowed_domains`",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].external_web_access,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Control whether the web search tool fetches live content or uses only cached content.
For OpenAI API compatibility ONLY. Request will be rejected if this field is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,web_search,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].allowed_x_handles,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of X Handles of the users from whom to consider the posts.

Note: This parameter cannot be set with `excluded_x_handles`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].enable_image_understanding,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Enable image understanding during X search.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].enable_video_understanding,body,boolean | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Enable video understanding during X search.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].excluded_x_handles,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"List of X Handles of the users from whom to exclude the posts.

Note: This parameter cannot be set with `allowed_x_handles`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].from_date,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Date from which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].to_date,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Date up to which to consider the results in ISO-8601 YYYY-MM-DD. See
<https://en.wikipedia.org/wiki/ISO_8601>.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,x_search,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].action,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Which image capabilities to expose to the model. One of `auto`
(the default; both generation and editing), `generate`
(text-to-image only), or `edit` (image editing only).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,image_generation,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].filters,body,any,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A filter to apply.
For OpenAI API compatibility ONLY. Request will be rejected if this field is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].max_num_results,body,integer | null,false,,1,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].ranking_options,body,any,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Ranking options for search.
For OpenAI API compatibility ONLY. Request will be rejected if this field is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,file_search,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].vector_store_ids,body,array<string>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,List of vector store IDs to search within.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].container,body,any,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The code interpreter container. Can be a container ID or an object that specifies
uploaded file IDs to make available to your code.
For OpenAI API compatibility ONLY. Request will be rejected if this field is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,code_interpreter,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].allowed_tools,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].authorization,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].connector_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].defer_loading,body,boolean | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,"When true, this server's tool definitions are hidden from the
model's prompt but stay callable, loaded via a `tool_search` step.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].headers,body,object | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].require_approval,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].server_description,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].server_label,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].server_url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,mcp,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment,body,ShellEnvironment,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The environment configuration for the shell tool.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment.skills,body,array<LocalShellSkill>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An optional list of skills available in the local environment.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment.skills[].description,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,A description of what the skill does.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment.skills[].name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The name of the skill.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment.skills[].path,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The path to the directory containing the skill (with a SKILL.md file).,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].environment.type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The type of the environment. Currently only `local` is supported.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,shell,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].execution,body,string | null,false,,,,,*,,DOCUMENTED|ACCOUNT_RESTRICTED,"Where the search runs. Only `""server""` (or unset) is supported.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,tools[].type,body,string,true,,,,tool_search,*,,DOCUMENTED|ACCOUNT_RESTRICTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,top_k,body,integer | null,false,,1,,,*,,DOCUMENTED|LIVE_VERIFIED,Top-k sampling: only the `top_k` most probable tokens are considered at each sampling step. Disabled when unset.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,top_logprobs,body,integer | null,false,,0,8,,*,,DOCUMENTED|LIVE_VERIFIED,"An integer between 0 and 8 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used. Not supported by models `grok-4.20` and newer; the field will be silently ignored if set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,top_p,body,number | null,false,1,,1,,*,,DOCUMENTED,"An alternative to sampling with `temperature`, called nucleus sampling, where the model considers the results of the tokens with `top_p` probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. It is generally recommended to alter this or `temperature` but not both.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,truncation,body,string | null,false,,,,,*,,DOCUMENTED,Not supported. Only maintained for compatibility reasons.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,user,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,GET /v1/responses/{response_id},response_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Stored response id. 200 ModelResponse; after DELETE → 404.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,DELETE /v1/responses/{response_id},response_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Stored response id → {id, object:'response', deleted:true}.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,GET /v1/responses/{response_id}/input_items,response_id,path,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Stored response id.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,GET /v1/responses/{response_id}/input_items,limit,query,integer,false,20,,,,*,,DOCUMENTED,"Maximum number of items to return (1-100, default 20).",https://docs.x.ai/openapi.json
xai,GET /v1/responses/{response_id}/input_items,order,query,string,false,asc,,,asc|desc,*,,DOCUMENTED,Sort order: asc or desc. Default asc.,https://docs.x.ai/openapi.json
xai,GET /v1/responses/{response_id}/input_items,after,query,string,false,,,,,*,,DOCUMENTED,Cursor for pagination. Returns items after this item ID (ids look like `item_0`).,https://docs.x.ai/openapi.json
xai,POST /v1/responses/compact,input,body,ModelInput,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The full input window to compact.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content,body,ModelInputContent,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Text, image or audio input.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Text input.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].type,body,string,true,,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].type,body,string,true,,,,output_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].image_url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A public URL of image prompt, only available for vision models.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].type,body,string,true,,,,input_image,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].file_data,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Inline file bytes (base64 encoded).

When set, the file content is provided directly in the request and does NOT need
to be uploaded to the Files API first.

Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The file ID from the Files API.

Set this to reference a previously-uploaded file.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].file_url,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Public URL for a file attachment.

When set, the file content is fetched from this URL.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].filename,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Filename for inline uploads.

Required when `file_data` is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].mime_type,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Optional MIME type for inline uploads (e.g. ""application/pdf"").",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].content[].type,body,string,true,,,,input_file,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].name,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A unique identifier representing your end-user, which can help xAI to monitor and detect abuse. Only supported for `user` messages.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].role,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the message. Possible values are `user`, `assistant`, `system` and `developer`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].type,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the message, which is always `message`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].call_id,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the function tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output,body,ModelInputContent,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The output of the function tool call. Can be a plain string or a list
of content items (`input_text`, `input_image`, `input_file`).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Text input.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].type,body,string,true,,,,input_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].type,body,string,true,,,,output_text,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Only included for compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].image_url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"A public URL of image prompt, only available for vision models.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].type,body,string,true,,,,input_image,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].file_data,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Inline file bytes (base64 encoded).

When set, the file content is provided directly in the request and does NOT need
to be uploaded to the Files API first.

Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].file_id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The file ID from the Files API.

Set this to reference a previously-uploaded file.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].file_url,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Public URL for a file attachment.

When set, the file content is fetched from this URL.
Exactly one of `file_id`, `file_data`, or `file_url` should be set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].filename,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Filename for inline uploads.

Required when `file_data` is set.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].mime_type,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Optional MIME type for inline uploads (e.g. ""application/pdf"").",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].type,body,string,true,,,,input_file,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the function tool call, which is always `function_call_output`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].call_id,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the shell tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The maximum length of the shell command output. Generated by the model
and should be passed back with the raw output.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output,body,array<ShellCallOutputResult>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An array of shell call output contents.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].outcome,body,ShellCallOutcome,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The outcome of the command execution.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].outcome.exit_code,body,integer,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The exit code of the command.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].outcome.type,body,string,true,,,,exit,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].outcome.type,body,string,true,,,,timeout,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].stderr,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The standard error output that was captured.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].output[].stdout,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The standard output that was captured.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the output item, which is always `shell_call_output`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].encrypted_content,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The encrypted content of the compacted conversation.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].id,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the compaction item (e.g. `cmp_<uuid>`).,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the item. Always `""compaction""`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses/compact,model,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Model to use for compaction summarization (required).,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content,body,array<OutputMessageContent>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Content of the output message.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations,body,array<Annotation>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Citations.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations[].end_index,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The end index of the annotation.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations[].start_index,body,integer | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The summary of the annotation.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations[].title,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The title of the annotation.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The type of the annotation. Only supported type currently is `url_citation`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].annotations[].url,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The URL of the web resource.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs,body,array<TokenLogProb>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The log probabilities of each output token returned in the content of message.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].bytes,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The ASCII encoding of the output character.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].logprob,body,number,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The log probability of returning this token.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].token,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The token.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].top_logprobs,body,array<TopLogProb>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,An array of the most likely tokens to return at this token position.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].top_logprobs[].bytes,body,array | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The ASCII encoding of the output character.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].top_logprobs[].logprob,body,number,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The log probability of returning this token.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].logprobs[].top_logprobs[].token,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The token.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The text output from the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,refusal,*,,DOCUMENTED|LIVE_VERIFIED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].refusal,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The reason for the refusal.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the output message.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].role,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The role of the output message, which can be `assistant` or `tool`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Status of the item. One of `completed`, `in_progress` or `incomplete`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the output message, which is always `message`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].arguments,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The arguments to pass to the function, as a JSON string.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the function tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the function tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].name,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The name of the function.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Status of the item. One of `completed`, `in_progress` or `incomplete`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the function tool call, which can be `""function_call""` for client-side tool calls,
and `""web_search_call""` or `""x_search_call""` or `""code_interpreter_call""` or `""mcp_call""` for server-side tool calls.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content,body,array<ReasoningText>,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The reasoning text contents.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Reasoning done by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].content[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the object, which is always `reasoning_text`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].encrypted_content,body,string | null,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The enrypted reasoning. Returned when `reasoning.encrypted_content` is passed in `include`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The unique ID of the reasoning content.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"Status of the item. One of `completed`, `in_progress` or `incomplete`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].summary,body,array<SummaryText>,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,The summarized reasoning text contents.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].summary[].text,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,Summary of the reasoning done by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].summary[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the object, which is always `summary_text`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED|LIVE_VERIFIED,"The type of the object, which is always `reasoning`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action,body,WebSearchAction,true,,,,,*,,DOCUMENTED,"An object describing the specific action taken in this web search call.
Includes details on how the model used the web (search, open_page, find).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.query,body,string,true,,,,,*,,DOCUMENTED,The search query.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.sources,body,array<WebSearchSource>,false,,,,,*,,DOCUMENTED,The sources used in the search.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.sources[].type,body,string,true,,,,,*,,DOCUMENTED,The type of source.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.sources[].url,body,string,false,,,,,*,,DOCUMENTED,The URL of the source.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.type,body,string,true,,,,find_in_page,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.url,body,string,true,,,,,*,,DOCUMENTED,The URL of the page being searched within.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.pattern,body,string,true,,,,,*,,DOCUMENTED,The pattern or text to search for within the page.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the web search tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,The status of the web search tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,The type of the web search tool call. Always `web_search_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the file search tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].queries,body,array<string>,true,,,,,*,,DOCUMENTED,The queries used to search for files.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].results,body,array<FileSearchResult>,true,,,,,*,,DOCUMENTED,The results of the file search tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].results[].file_id,body,string,true,,,,,*,,DOCUMENTED,The file ID of the file search result.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].results[].filename,body,string,true,,,,,*,,DOCUMENTED,The filename of the file search result.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].results[].score,body,number,false,,,,,*,,DOCUMENTED,"The score of the file search result.
Proto3 omits float fields with value 0.0; default to 0.0 when absent.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].results[].text,body,string,true,,,,,*,,DOCUMENTED,The text of the file search result.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,The status of the file search tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,The type of the file search tool call. Always `file_search_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].code,body,string,false,,,,,*,,DOCUMENTED,The code of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].outputs,body,array<CodeInterpreterOutput>,true,,,,,*,,DOCUMENTED,The outputs of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].outputs[].logs,body,string,true,,,,,*,,DOCUMENTED,The output of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].outputs[].type,body,string,true,,,,image,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].outputs[].url,body,string,true,,,,,*,,DOCUMENTED,The error of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,The status of the code interpreter tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,The type of the code interpreter tool call. Always `code_interpreter_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].arguments,body,string,true,,,,,*,,DOCUMENTED,A JSON string of the arguments passed to the tool.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].error,body,string,false,,,,,*,,DOCUMENTED,The error message of the MCP tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the MCP tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].name,body,string,true,,,,,*,,DOCUMENTED,The name of the tool that was run.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].output,body,string,true,,,,,*,,DOCUMENTED,The output of the MCP tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].server_label,body,string,true,,,,,*,,DOCUMENTED,The label of the MCP server running the tool.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,The status of the MCP tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,The type of the MCP tool call. Always `mcp_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the function tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,true,,,,,*,,DOCUMENTED,The status of the custom tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].input,body,string,false,,,,,*,,DOCUMENTED,"The unique ID of the custom tool call,",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].name,body,string,true,,,,,*,,DOCUMENTED,An identifier used to map this custom tool call to a tool call output.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,"Status of the item. One of `completed`, `in_progress` or `incomplete`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,The input for the custom tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action,body,ShellCallAction,true,,,,,*,,DOCUMENTED,The shell commands and limits that describe how to run the tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.commands,body,array<string>,true,,,,,*,,DOCUMENTED,The commands to run.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.max_output_length,body,integer | null,false,,,,,*,,DOCUMENTED,Optional maximum number of characters to return from each command.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.timeout_ms,body,integer | null,false,,,,,*,,DOCUMENTED,Optional timeout in milliseconds for the commands.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].action.type,body,string | null,false,,,,,*,,DOCUMENTED,"The type of the action, which is always `exec`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the shell tool call generated by the model.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment,body,null | ShellEnvironment,false,,,,,*,,DOCUMENTED,,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment.skills,body,array<LocalShellSkill>,false,,,,,*,,DOCUMENTED,An optional list of skills available in the local environment.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment.skills[].description,body,string,true,,,,,*,,DOCUMENTED,A description of what the skill does.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment.skills[].name,body,string,true,,,,,*,,DOCUMENTED,The name of the skill.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment.skills[].path,body,string,true,,,,,*,,DOCUMENTED,The path to the directory containing the skill (with a SKILL.md file).,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].environment.type,body,string,true,,,,,*,,DOCUMENTED,The type of the environment. Currently only `local` is supported.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the shell call item.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,"Status of the item. One of `completed`, `in_progress` or `incomplete`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,,*,,DOCUMENTED,"The type of the output item, which is always `shell_call`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,false,,,,,*,,DOCUMENTED,The unique ID of the image generation tool call.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].prompt,body,string | null,false,,,,,*,,DOCUMENTED,The prompt used to generate the image.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].result,body,string | null,false,,,,,*,,DOCUMENTED,"The generated image encoded in base64 (OpenAI-compatible: bare
base64, no data-URL prefix), or `null` while the call is still in
progress or has failed. The image format can be determined from the
decoded magic bytes (typically JPEG or PNG).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,"The status of the image generation tool call. One of `in_progress`,
`generating`, `completed` or `failed`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,image_generation_call,*,,DOCUMENTED,The type of the image generation tool call. Always `image_generation_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].arguments,body,any,false,,,,,*,,DOCUMENTED,"The arguments of the tool search call: `{""query"": ..., ""limit"": ...}`.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string | null,false,,,,,*,,DOCUMENTED,Always `null`; present for OpenAI compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].execution,body,string,false,,,,,*,,DOCUMENTED,Where the search runs. Always `server`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the tool search call item.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,"Status of the item. One of `in_progress`, `completed` or
`incomplete` (a failed search).",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,tool_search_call,*,,DOCUMENTED,The type of the item. Always `tool_search_call`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].call_id,body,string | null,false,,,,,*,,DOCUMENTED,Always `null`; present for OpenAI compatibility.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].execution,body,string,false,,,,,*,,DOCUMENTED,Where the search ran. Always `server`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].id,body,string,true,,,,,*,,DOCUMENTED,The unique ID of the tool search output item.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].status,body,string,false,,,,,*,,DOCUMENTED,Status of the item. Always `completed`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].tools,body,array<any>,false,,,,,*,,DOCUMENTED,"The loaded tool definitions, echoed in the shape they were supplied in.",https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,POST /v1/responses,input[].type,body,string,true,,,,tool_search_output,*,,DOCUMENTED,The type of the item. Always `tool_search_output`.,https://docs.x.ai/developers/rest-api-reference/inference/responses
xai,GET /v1/skills,limit,query,integer|null,false,100,1,100,,,,DOCUMENTED|ACCOUNT_RESTRICTED,Page size 1–100.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,GET /v1/skills,after,query,string|null,false,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,Cursor: id of the last item of the previous page.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,GET /v1/skills,order,query,string|null,false,desc,,,asc|desc,,,DOCUMENTED|ACCOUNT_RESTRICTED,asc | desc.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,POST /v1/skills,files,body,array<binary>,true,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,"Skill zip, or the field repeated once per file of a directory upload. SKILL.md frontmatter (name, description) is parsed server-side.",https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,GET /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,Skill id.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,DELETE /v1/skills/{skill_id},skill_id,path,string,true,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,Skill id.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,GET /v1/skills/{skill_id}/content,skill_id,path,string,true,,,,,,,DOCUMENTED|ACCOUNT_RESTRICTED,Skill id.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,POST /v1/responses (tools[] shell.environment),environment.skills[],body,array<LocalShellSkill>,false,,,,,,,DOCUMENTED,"Skills available in a local shell environment: {name, description, path (directory with SKILL.md)}. `environment.type` is 'local'. This is the only place the OpenAPI wires skills into inference (ShellCall.environment).",https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,POST /v1/videos/generations,model,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,"grok-imagine-video ($0.05/s; text/image/video inputs) or grok-imagine-video-1.5 ($0.08/s; text/image/audio inputs, native 1080p, reference audio, last_frame).",https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,prompt,body,string,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,"Required for text-to-video; optional when `image`, `reference_images` or `last_frame` is present.",https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,duration,body,integer|null,false,8,1,15,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,"Output length in seconds, 1–15 (also accepts `seconds` and string values for OpenAI compatibility).",https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,aspect_ratio,body,string|null,false,16:9,,,1:1|16:9|9:16|4:3|3:4|3:2|2:3,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,Ignored for image-to-video (derived from the first frame — guide says it can override and stretch).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,resolution,body,string|null,false,480p,,,480p|720p|1080p,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,"480p (default, fastest), 720p, 1080p (grok-imagine-video-1.5 T2V/I2V only; reference-to-video capped at 720p).",https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,image,body,object|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,First-frame image for image-to-video ({url | file_id}; also `input_reference`). With reference_images/reference_audios/last_frame on 1.5 it pins the first frame.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,image.url,body,string,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Public URL or base64 data URL (JPEG/PNG/WebP). Also `image_url`.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,image.file_id,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Files API file id.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,reference_images,body,array<object>,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,"Reference-to-video: style/content references ({url | file_id}), tagged <IMAGE_0>… in the prompt.",https://docs.x.ai/developers/model-capabilities/video/reference-to-video
xai,POST /v1/videos/generations,reference_audios,body,array<object>,false,,,3,,grok-imagine-video-1.5,,DOCUMENTED,"Up to 3 voice references ({voice_id} preset from GET /v1/tts/voices, or {url} clip ≤ 15 s — own-clip uploads are trusted-partner only). Tag <AUDIO_0>… in the prompt. grok-imagine-video-1.5 only.",https://docs.x.ai/developers/model-capabilities/video/reference-to-video
xai,POST /v1/videos/generations,reference_audios[].voice_id,body,string|null,false,,,,,grok-imagine-video-1.5,,DOCUMENTED,Preset voice id (case-insensitive); unknown → 400 with the list of voices.,https://docs.x.ai/developers/model-capabilities/video/reference-to-video
xai,POST /v1/videos/generations,reference_audios[].url,body,string|null,false,,,,,grok-imagine-video-1.5,,DOCUMENTED,"data:audio/wav;base64,… or http(s) clip ≤ 15 s (trusted partners).",https://docs.x.ai/developers/model-capabilities/video/reference-to-video
xai,POST /v1/videos/generations,last_frame,body,object,false,,,,,grok-imagine-video-1.5,,DOCUMENTED,Pin the exact last frame ({url | file_id}); alone or with `image` (first frame → interpolation). grok-imagine-video-1.5 only; classic grok-imagine-video rejects it.,https://docs.x.ai/developers/model-capabilities/video/generation
xai,POST /v1/videos/generations,generate_audio,body,boolean,false,true,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Videos include an audio track by default; false → silent video.,https://docs.x.ai/developers/model-capabilities/video/generation
xai,POST /v1/videos/generations,output,body,object|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Bring-your-own storage: {upload_url} presigned URL the service PUTs the MP4 to (used by Grok Build under ZDR).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,output.upload_url,body,string,true,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Signed URL for HTTP PUT of the generated video.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,user,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Opaque end-user id.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,storage_options,body,object,false,,,,,,,DOCUMENTED,Persist the output to the Files API and return `file_output` alongside the ephemeral URL/b64.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,storage_options.filename,body,string,true,,,,,,,DOCUMENTED,Filename for the stored file (required when storage_options is present).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,storage_options.expires_after,body,integer|null,false,,,2592000,,,,DOCUMENTED,Seconds until the stored file auto-expires; max 2592000 (30 days); omitted = no expiry.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/generations,storage_options.public_url,body,boolean|object,false,,,,,,,DOCUMENTED,Also create a permanent shareable public URL (`file_output.public_url`); object form accepts an `expires_after`.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,model,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Video model (docs examples use grok-imagine-video).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,prompt,body,string,true,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Edit instruction.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,video,body,object,true,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED|LIVE_VERIFIED,"Source video {url | file_id}; .mp4 (H.264/H.265/AV1…), ≤ 8.7 s, output capped at 720p.",https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,video.url,body,string,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Public .mp4 URL or base64 data URL.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,video.file_id,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Files API file id (video).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,output,body,object|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,{upload_url} presigned PUT target.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,user,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Opaque end-user id.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,storage_options,body,object,false,,,,,,,DOCUMENTED,Persist the output to the Files API and return `file_output` alongside the ephemeral URL/b64.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,storage_options.filename,body,string,true,,,,,,,DOCUMENTED,Filename for the stored file (required when storage_options is present).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,storage_options.expires_after,body,integer|null,false,,,2592000,,,,DOCUMENTED,Seconds until the stored file auto-expires; max 2592000 (30 days); omitted = no expiry.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/edits,storage_options.public_url,body,boolean|object,false,,,,,,,DOCUMENTED,Also create a permanent shareable public URL (`file_output.public_url`); object form accepts an `expires_after`.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,model,body,string|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Video model.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,prompt,body,string,true,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,What happens next.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,video,body,object,true,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Source video {url | file_id}.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,duration,body,integer|null,false,6,2,10,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,Length of the ADDED segment in seconds (2–10); total output = original + extension.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,output,body,object|null,false,,,,,grok-imagine-video|grok-imagine-video-1.5,,DOCUMENTED,{upload_url} presigned PUT target.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,storage_options,body,object,false,,,,,,,DOCUMENTED,Persist the output to the Files API and return `file_output` alongside the ephemeral URL/b64.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,storage_options.filename,body,string,true,,,,,,,DOCUMENTED,Filename for the stored file (required when storage_options is present).,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,storage_options.expires_after,body,integer|null,false,,,2592000,,,,DOCUMENTED,Seconds until the stored file auto-expires; max 2592000 (30 days); omitted = no expiry.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,POST /v1/videos/extensions,storage_options.public_url,body,boolean|object,false,,,,,,,DOCUMENTED,Also create a permanent shareable public URL (`file_output.public_url`); object form accepts an `expires_after`.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,GET /v1/videos/{request_id},request_id,path,string (UUID),true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Deferred request id from a generations/edits/extensions call.,https://docs.x.ai/developers/rest-api-reference/inference/videos
xai,GET /v1/video-generation-models/{model_id},model_id,path,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,Model id or alias.,https://docs.x.ai/openapi.json (local copy sources/xai/openapi/openapi.json)
xai,POST /v1/realtime/client_secrets,expires_after.seconds,body,integer,false,600,,3600,,,,DOCUMENTED|LIVE_VERIFIED,TTL of the secret; max 3600; default 600.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/client_secrets,session,body,object|null,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Initial session config stored with the secret and applied when the WebSocket opens.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/client_secrets,session.model,body,string,false,,,,grok-voice-latest|grok-voice-think-fast-2.0,,,DOCUMENTED|LIVE_VERIFIED,Voice model bound to the secret.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/client_secrets,session.reasoning.effort,body,string,false,high,,,high|none,,,DOCUMENTED|LIVE_VERIFIED,high (default) or none.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime,model,query,string,false,grok-voice-latest,,,grok-voice-latest|grok-voice-think-fast-2.0,grok-voice-latest|grok-voice-think-fast-2.0,,DOCUMENTED|LIVE_VERIFIED,Voice model; ignored with call_id.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime,reasoning.effort,query,string,false,high,,,high|none,,,DOCUMENTED,Enable/disable reasoning for the session.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime,call_id,query,string,false,,,,,,,DOCUMENTED,SIP call id from a realtime.call.incoming webhook; binds the socket to that call (API key auth only).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime,conversation_id,query,string,false,,,,,,,DOCUMENTED,Resume a cached conversation (requires session.resumption.enabled on both sessions; 30 min inactivity expiry).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime,Authorization,header,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Bearer <api key> (server) or Bearer <ephemeral secret>.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime,Sec-WebSocket-Protocol,header,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Browser auth: `xai-client-secret.<ephemeral token>`; replaces the Authorization header.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime session.update,session.model,body,string,false,,,,grok-voice-latest|grok-voice-think-fast-2.0,,,DOCUMENTED,Model (also settable by query).,https://docs.x.ai/voice-realtime.ws.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.instructions,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,"System prompt (second person, fixed section order recommended).",https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.reasoning.effort,body,string,false,high,,,high|none,,,DOCUMENTED|LIVE_VERIFIED,high (default) or none.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.voice,body,string,false,xai_ara (observed in session.created; docs say eve),,,,,,DOCUMENTED|LIVE_VERIFIED,Built-in voice id (GET /v1/tts/voices) or 8-char custom voice id.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.turn_detection.type,body,string|null,false,,,,server_vad|,,,DOCUMENTED|LIVE_VERIFIED,`server_vad` (auto turns) or null (manual: commit + response.create).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.turn_detection.threshold,body,number,false,0.85,0.1,0.9,,,,DOCUMENTED,VAD activation threshold 0.1–0.9.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.turn_detection.silence_duration_ms,body,number,false,,0,10000,,,,DOCUMENTED,Silence before the turn ends (0–10000).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.turn_detection.prefix_padding_ms,body,number,false,333,0,10000,,,,DOCUMENTED,Audio kept before detected speech start (0–10000).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.turn_detection.idle_timeout_ms,body,number|null,false,,,,,,,DOCUMENTED,Proactive re-engagement timer after each response (emits input_audio_buffer.timeout_triggered).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.resumption.enabled,body,boolean,false,false,,,,,,DOCUMENTED,Cache turns keyed by conversation_id for replay on reconnect (xAI extension).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.format.type,body,string,false,audio/pcm,,,audio/pcm|audio/pcmu|audio/pcma|audio/opus,,,DOCUMENTED|LIVE_VERIFIED,Input codec.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.format.rate,body,integer,false,24000,,,8000|11025|16000|22050|24000|32000|44100|48000,,,DOCUMENTED|LIVE_VERIFIED,PCM sample rate (Hz).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.transport,body,string,false,json,,,json|binary,,,DOCUMENTED|LIVE_VERIFIED,json (base64 in input_audio_buffer.append) or binary (raw frames; dual-accept).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.transcription.language_hint,body,string,false,,,,,,,DOCUMENTED,BCP-47 hint for ASR (es/pt need a regional variant).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.transcription.keyterms,body,array<string>,false,,,100,,,,DOCUMENTED,"≤ 100 terms, ≤ 50 chars each.",https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.input.transcription.model,body,string,false,,,,grok-transcribe,,,DOCUMENTED,Set to `grok-transcribe` to receive conversation.item.input_audio_transcription.updated events (live captions).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.output.format.type,body,string,false,audio/pcm,,,audio/pcm|audio/pcmu|audio/pcma|audio/opus,,,DOCUMENTED|LIVE_VERIFIED,Output codec.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.output.format.rate,body,integer,false,24000,,,8000|11025|16000|22050|24000|32000|44100|48000,,,DOCUMENTED|LIVE_VERIFIED,PCM sample rate (Hz).,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.output.transport,body,string,false,json,,,json|binary,,,DOCUMENTED|LIVE_VERIFIED,json or binary; strict (output only on the configured path); changes apply at the next response boundary.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.audio.output.speed,body,number,false,1.0,0.7,1.5,,,,DOCUMENTED,Playback speed 0.7–1.5.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.replace,body,object,false,,,,,,,DOCUMENTED,Phrase → spoken substitution map applied before TTS (transcript unchanged); echoed on session.updated.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.tools,body,array<object>,false,,,,,,,DOCUMENTED,"file_search {vector_store_ids, max_num_results} · web_search {allowed_domains|excluded_domains ≤5, enable_image_understanding, location} · x_search {allowed_x_handles|excluded_x_handles ≤20, from_date, to_date, enable_image_understanding, enable_video_understanding} · mcp {server_url, server_label, server_description, allowed_tools, authorization, headers} · function {name, description, parameters}.",https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime session.update,session.tool_choice,body,string,false,auto,,,,,,LIVE_DISCOVERED,Observed in session.updated ('auto'); not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.enable_noise_suppression,body,boolean,false,false,,,,,,LIVE_DISCOVERED,Observed in session.updated (false); not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.enable_phonetic_spelling,body,boolean,false,false,,,,,,LIVE_DISCOVERED,Observed in session.updated (false); not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.keep_context,body,boolean,false,false,,,,,,LIVE_DISCOVERED,Observed in session.updated (false); not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.temperature,body,number,false,-1.0,,,,,,LIVE_DISCOVERED,Observed in session.updated as -1.0 (unset); not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime session.update,session.max_response_output_tokens,body,string|integer,false,inf,,,,,,LIVE_DISCOVERED,Observed in session.updated as 'inf'; not documented.,tmp-live/xai-media/voice-ws-session.json
xai,WS wss://api.x.ai/v1/realtime response.create,response.instructions,body,string,false,,,,,,,DOCUMENTED,Per-response override of the session system prompt.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime response.create,response.metadata,body,object|null,false,,,,,,,DOCUMENTED,Developer key-values echoed on response.done (schema).,https://docs.x.ai/voice-realtime.ws.json
xai,WS wss://api.x.ai/v1/realtime conversation.item.create,item.type,body,string,false,,,,message|function_call|function_call_output|force_message,,,DOCUMENTED|LIVE_VERIFIED,"message (user/assistant, content input_text|input_audio|output_text) · function_call (seed history) · function_call_output {call_id, output} · force_message (xAI: verbatim TTS line, {role:'assistant', interruptible, content:[{type:'output_text', text}]}).",https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,WS wss://api.x.ai/v1/realtime conversation.item.create,item.interruptible,body,boolean,false,true,,,,,,DOCUMENTED,force_message only: false drops caller audio until playback completes.,https://docs.x.ai/developers/model-capabilities/audio/speech-to-speech
xai,POST /v1/tts,text,body,string,true,,,60000,,,,DOCUMENTED|LIVE_VERIFIED,"≤ 60,000 chars; speech tags [pause] [laugh] … and wrappers <whisper> <soft> …",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,voice_id,body,string,false,eve,,,,,,DOCUMENTED|LIVE_VERIFIED,Built-in (GET /v1/tts/voices) or custom voice id.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,language,body,string,true,,,,,,,DOCUMENTED|LIVE_VERIFIED,"BCP-47 (en, zh, pt-BR…) or auto. REQUIRED.",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,output_format.codec,body,string,false,mp3,,,mp3|wav|pcm|mulaw|alaw,,,DOCUMENTED|LIVE_VERIFIED,Audio codec.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,output_format.sample_rate,body,integer|null,false,24000,,,8000|16000|22050|24000|44100|48000,,,DOCUMENTED|LIVE_VERIFIED,Hz.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,output_format.bit_rate,body,integer|null,false,128000,,,32000|64000|96000|128000|192000,,,DOCUMENTED,"bps, mp3 only.",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,speed,body,number,false,1.0,0.7,1.5,,,,DOCUMENTED,0.7–1.5.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,optimize_streaming_latency,body,string|integer,false,0,,,0|1|2,,,DOCUMENTED,0 (quality) | 1 (smaller first chunk); the guide also lists 2.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,text_normalization,body,boolean,false,false,,,,,,DOCUMENTED,Expand numbers/abbreviations to spoken form.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,with_timestamps,body,boolean,false,false,,,,,,DOCUMENTED|LIVE_VERIFIED,true → JSON envelope with per-character timings (adds latency).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/tts,replace,body,object,false,,,200,,,,DOCUMENTED,"Phrase → respelling or /IPA/ map; ≤ 200 entries, keys ≤ 100 chars (letters/digits/apostrophes/spaces), values ≤ 128; expanded text ≤ 240,000 chars.",https://docs.x.ai/developers/model-capabilities/audio/text-to-speech
xai,WS wss://api.x.ai/v1/tts,voice,query,string,false,eve,,,,,,DOCUMENTED,Voice id.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,language,query,string,false,,,,,,,DOCUMENTED,BCP-47 or auto (required).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,codec,query,string,false,mp3,,,mp3|wav|pcm|mulaw|alaw,,,DOCUMENTED,Output codec.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,sample_rate,query,integer,false,24000,,,,,,DOCUMENTED,Hz.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,bit_rate,query,integer,false,128000,,,,,,DOCUMENTED,bps (mp3).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,optimize_streaming_latency,query,integer,false,0,,,0|1,,,DOCUMENTED,0 or 1.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,speed,query,number,false,1.0,,,,,,DOCUMENTED,0.7–1.5.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,text_normalization,query,boolean,false,false,,,,,,DOCUMENTED,Normalize text.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/tts,with_timestamps,query,boolean,false,false,,,,,,DOCUMENTED,audio_timestamps on each audio.delta.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,file,body,binary,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Audio ≤ 500 MB; must be the LAST multipart field. Either file or url.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,url,body,string,false,,,,,,,DOCUMENTED,Server-side download URL (alternative to file).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,audio_format,body,string,false,,,,pcm|mulaw|alaw|wav|mp3|ogg|opus|flac|aac|mp4|m4a|mkv,,,DOCUMENTED,Only for raw formats.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,sample_rate,body,string,false,,,,8000|16000|22050|24000|44100|48000,,,DOCUMENTED,Required for raw formats (also sample_rate_hertz).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,language,body,string,false,,,,,,,DOCUMENTED|LIVE_VERIFIED,Language code; with format=true enables inverse text normalization.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,format,body,string,false,false,,,true|false,,,DOCUMENTED,'true' enables formatting (requires language).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,multichannel,body,string,false,false,,,true|false,,,DOCUMENTED,Per-channel transcription → channels[].,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,channels,body,integer,false,,2,8,,,,DOCUMENTED,"2–8, required for multichannel raw audio.",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,diarize,body,string,false,false,,,true|false,,,DOCUMENTED,Speaker diarization → words[].speaker.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,keyterm,body,array<string>,false,,,100,,,,DOCUMENTED,"Repeatable bias terms, ≤ 100 × 50 chars.",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,filler_words,body,string,false,false,,,true|false,,,DOCUMENTED,Keep uh/um.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,vad_threshold,body,number,false,0.5,0,1,,,,DOCUMENTED,Speech-probability gate 0–1; 0 disables.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/stt,model,body,string,false,grok-voice-transcribe-2.0,,,grok-voice-transcribe-1.0|grok-voice-transcribe-2.0,,,DOCUMENTED,grok-voice-transcribe-1.0 | 2.0 (default 2.0 per the guide; the REST page does not list it for batch).,https://docs.x.ai/developers/model-capabilities/audio/voice
xai,WS wss://api.x.ai/v1/stt,sample_rate,query,integer,false,16000,,,8000|16000|22050|24000|44100|48000,,,DOCUMENTED,Hz (ignored for opus).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,encoding,query,string,false,pcm,,,pcm|mulaw|alaw|opus,,,DOCUMENTED,"pcm s16le | mulaw | alaw | opus (one packet per binary frame, mono).",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,interim_results,query,boolean,false,false,,,,,,DOCUMENTED,Emit is_final=false partials ~every 500 ms.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,endpointing,query,integer,false,400,,,,,,DOCUMENTED,Silence ms before speech_final (0–5000).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,language,query,string,false,,,,,,,DOCUMENTED,Enables inverse text normalization.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,model,query,string,false,grok-voice-transcribe-2.0,,,grok-voice-transcribe-1.0|grok-voice-transcribe-2.0,,,DOCUMENTED,Transcribe model.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,multichannel,query,boolean,false,false,,,,,,DOCUMENTED,Requires channels ≥ 2; not with opus.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,channels,query,integer,false,1,,,,,,DOCUMENTED,1–8.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,diarize,query,boolean,false,false,,,,,,DOCUMENTED,Speaker field on words.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,keyterm,query,string (repeatable),false,,,,,,,DOCUMENTED,Bias terms.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,filler_words,query,boolean,false,false,,,,,,DOCUMENTED,Keep fillers.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,smart_turn,query,number,false,,,,,,,DOCUMENTED,End-of-turn confidence threshold 0–1 (adds end_of_turn_confidence to partials).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,smart_turn_timeout,query,integer,false,,,,,,,DOCUMENTED,Max silence ms before forced speech_final (1–5000).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,WS wss://api.x.ai/v1/stt,vad_threshold,query,number,false,0.08,,,,,,DOCUMENTED,Speech gate 0–1.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,file,body,binary,true,,,,,,,DOCUMENTED,Reference clip ≤ 120 s (90–120 s recommended).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,name,body,string,false,,,,,,,DOCUMENTED,Display name.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,description,body,string,false,,,,,,,DOCUMENTED,Free text.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,gender,body,string,false,,,,male|female|neutral,,,DOCUMENTED,Label.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,accent,body,string,false,,,,,,,DOCUMENTED,Free text.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,age,body,string,false,,,,young|middle-aged|old,,,DOCUMENTED,Label.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,language,body,string,false,,,,,,,DOCUMENTED,ISO 639 / BCP-47 (region uppercase).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,use_case,body,string,false,,,,conversational|narration|characters|educational|advertisement|social_media|entertainment,,,DOCUMENTED,Label.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/custom-voices,tone,body,string,false,,,,warm|casual|professional|friendly|authoritative|expressive|calm,,,DOCUMENTED,Label.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,GET /v1/custom-voices,limit,query,integer,false,100,1,1000,,,,DOCUMENTED|LIVE_VERIFIED,1–1000.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,GET /v1/custom-voices,pagination_token,query,string,false,,,,,,,DOCUMENTED,Next-page token from the previous response.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,GET /v1/custom-voices/{voice_id},voice_id,path,string,true,,,,,,,DOCUMENTED,8-char lowercase alphanumeric id.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,origin,body,string,true,,,,xai_provisioned|byo_trunk,,,DOCUMENTED,byo_trunk for customer-owned Direct SIP numbers; xai_provisioned not available via API per the SIP guide.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,name,body,string,true,,,,,,,DOCUMENTED,Display name.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,agent_id,body,string,false,,,,,,,DOCUMENTED,Route to an agent (mutually exclusive with webhook).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,phone_number,body,string,false,,,,,,,DOCUMENTED,E.164 number (byo_trunk).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,area_code,body,string,false,,,,,,,DOCUMENTED,3-digit US area code filter (xai_provisioned).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,sip_auth.auth_username,body,string,false,,,,,,,DOCUMENTED,SIP digest username (with auth_password).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,sip_auth.auth_password,body,string,false,,,,,,,DOCUMENTED,"Stored encrypted, never returned.",https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,sip_auth.allowed_addresses,body,array<string>,false,,,,,,,DOCUMENTED,Source CIDRs allowed to INVITE.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,webhook.url,body,string,true,,,,,,,DOCUMENTED,Receives signed realtime.call.incoming POSTs.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,webhook.name,body,string,false,,,,,,,DOCUMENTED,Optional display name.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,webhook.auth_url,body,string,false,,,,,,,DOCUMENTED,OAuth token-exchange URL (with auth_token).,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v2/phone-numbers,webhook.auth_token,body,string,false,,,,,,,DOCUMENTED,Bearer credential or OAuth client credential.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/calls/{call_id}/refer,call_id,path,string,true,,,,,,,DOCUMENTED,From realtime.call.incoming.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/calls/{call_id}/refer,target_uri,body,string,true,,,,,,,DOCUMENTED,tel:+E.164 or sip:user@host.,https://docs.x.ai/developers/rest-api-reference/inference/voice
xai,POST /v1/realtime/calls/{call_id}/hangup,call_id,path,string,true,,,,,,,DOCUMENTED,From realtime.call.incoming.,https://docs.x.ai/developers/rest-api-reference/inference/voice