What each model accepts
Every cell is an observed API response (2026-11-10). "rejected" means HTTP 400 (or 404 for an unsupported endpoint).
Responses API
| Model | Reasoning effort levels | Default effort is none |
temperature / top_p |
text.verbosity |
Code interpreter |
|---|---|---|---|---|---|
gpt-6-sol |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-6.1-terra |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-6.1-sol |
low, medium, high, xhigh, max | no | never | low, medium, high | yes |
gpt-6-luna |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-5.6-terra |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-5.6-sol |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-6-astra |
low, medium, high, xhigh, max | no | never | low, medium, high | yes |
gpt-5.5 |
none, low, medium, high, xhigh | no | only with effort none |
low, medium, high | yes |
gpt-5.4 |
none, low, medium, high, xhigh | yes | only with effort none |
low, medium, high | yes |
gpt-5.4-mini |
none, low, medium, high, xhigh | yes | only with effort none |
low, medium, high | yes |
gpt-5.4-nano |
none, low, medium, high, xhigh | yes | only with effort none |
low, medium, high | yes |
gpt-5.3-codex |
none, low, medium, high, xhigh | yes | only with effort none |
low, medium, high | yes |
gpt-5.2 |
none, low, medium, high, xhigh | yes | only with effort none |
low, medium, high | yes |
gpt-5.6-luna |
none, low, medium, high, xhigh, max | no | only with effort none |
low, medium, high | yes |
gpt-5.1 |
none, low, medium, high | yes | only with effort none |
low, medium, high | yes |
gpt-5.2-pro |
medium, high, xhigh | no | never | low, medium, high | rejected |
gpt-5-mini |
minimal, low, medium, high | no | never | low, medium, high | yes |
gpt-5-nano |
minimal, low, medium, high | no | never | low, medium, high | yes |
gpt-5 |
minimal, low, medium, high | no | never | low, medium, high | yes |
gpt-4.1 |
not a reasoning model | — | always | medium | yes |
gpt-4.1-mini |
not a reasoning model | — | always | medium | yes |
gpt-4o |
not a reasoning model | — | always | medium | yes |
gpt-4o-mini |
not a reasoning model | — | always | medium | yes |
o3 |
low, medium, high | no | never | medium | yes, tool_choice: auto only |
gpt-5-pro |
high | no | never | low, medium, high | rejected |
chat-latest |
medium | no | never | medium | yes |
gpt-5.4-pro |
medium, high, xhigh | no | never | low, medium, high | yes |
gpt-5.5-pro |
medium, high, xhigh | no | never | low, medium, high | yes |
o3-pro |
low, medium, high | no | never | medium | yes, tool_choice: auto only |
Accepted on the Responses API by every probed model: instructions, system and developer roles, strict structured outputs (text.format json_schema), json_object, function tools with tool_choice auto / required / none / forced function, parallel_tool_calls, web search, image input, store: false, truncation, prompt_cache_key, prompt_cache_retention, metadata, max_tool_calls, background mode, include: ["reasoning.encrypted_content"]. max_output_tokens below 16 is rejected everywhere. top_logprobs is rejected by every reasoning model.
Chat Completions
| Model | Served | Legacy max_tokens |
Custom temperature |
stop |
Function tools | logprobs |
|---|---|---|---|---|---|---|
gpt-6-sol |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-6.1-terra |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-6.1-sol |
yes | rejected | never | rejected | rejected — use Responses | rejected |
gpt-6-luna |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-5.6-terra |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-5.6-sol |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-6-astra |
yes | rejected | never | rejected | rejected — use Responses | rejected |
gpt-5.5 |
yes | rejected | only with effort none |
rejected | yes | rejected |
gpt-5.4 |
yes | rejected | only with effort none |
rejected | yes | yes |
gpt-5.4-mini |
yes | rejected | only with effort none |
rejected | yes | yes |
gpt-5.4-nano |
yes | rejected | only with effort none |
rejected | yes | yes |
gpt-5.3-codex |
no — Responses only | — | — | — | — | — |
gpt-5.2 |
yes | rejected | only with effort none |
rejected | yes | yes |
gpt-5.6-luna |
yes | rejected | only with effort none |
rejected | only with reasoning_effort: "none" |
rejected |
gpt-5.1 |
yes | rejected | only with effort none |
rejected | yes | yes |
gpt-5.2-pro |
no — Responses only | — | — | — | — | — |
gpt-5-mini |
yes | rejected | never | rejected | yes | rejected |
gpt-5-nano |
yes | rejected | never | rejected | yes | rejected |
gpt-5 |
yes | rejected | never | rejected | yes | rejected |
gpt-4.1 |
yes | yes | always | yes | yes | yes |
gpt-4.1-mini |
yes | yes | always | yes | yes | yes |
gpt-4o |
yes | yes | always | yes | yes | yes |
gpt-4o-mini |
yes | yes | always | yes | yes | yes |
o3 |
yes | rejected | never | rejected | yes | rejected |
gpt-5-pro |
no — Responses only | — | — | — | — | — |
chat-latest |
yes | rejected | never | rejected | yes | rejected |
gpt-5.4-pro |
no — Responses only | — | — | — | — | — |
gpt-5.5-pro |
no — Responses only | — | — | — | — | — |
o3-pro |
no — Responses only | — | — | — | — | — |
Exact error texts
max_output_8: Invalid 'max_output_tokens': integer below minimum value. Expected a value >= 16, but got 8 instead.chat_temperature: Unsupported value: 'temperature' does not support 0.5 with this model. Only the default (1) value is supported.chat_logprobs: Unsupported parameter: 'logprobs' is not supported with this model.chat_max_tokens_legacy: Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.top_p: Unsupported parameter: 'top_p' is not supported with this model.chat_stop: Unsupported parameter: 'stop' is not supported with this model.chat_tools: Function tools with reasoning_effort are not supported for in /v1/chat/completions. To use function tools, use /v1/responses or set reasoning_effort to 'none'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', 'high', 'xhigh', and 'max'.temperature: Unsupported parameter: 'temperature' is not supported with this model.top_logprobs: logprobs are not supported with reasoning models.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'low', 'medium', 'high', 'xhigh', and 'max'.effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'low', 'medium', 'high', 'xhigh', and 'max'.chat_tools_effort_none: Unsupported value: 'reasoning_effort' does not support 'none' with this model. Supported values are: 'low', 'medium', 'high', and 'xhigh'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', 'high', and 'xhigh'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', 'high', and 'xhigh'.chat_basic: This model is not supported in the v1/chat/completions endpoint. Use the v1/responses endpoint instead.chat_logprobs: You are not allowed to request logprobs from this modelservice_tier_flex: Flex is not available for this model.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', and 'high'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', and 'high'.effort_xhigh: Unsupported value: 'xhigh' is not supported with the '' model. Supported values are: 'none', 'low', 'medium', and 'high'.chat_developer_role: This is not a chat model and thus not supported in the v1/chat/completions endpoint. Did you mean to use v1/completions?effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'medium', 'high', and 'xhigh'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'medium', 'high', and 'xhigh'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'medium', 'high', and 'xhigh'.effort_low: Unsupported value: 'low' is not supported with the '' model. Supported values are: 'medium', 'high', and 'xhigh'.code_interpreter: Tool 'code_interpreter' is not supported with .effort_xhigh: Unsupported value: 'xhigh' is not supported with the '' model. Supported values are: 'minimal', 'low', 'medium', and 'high'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'minimal', 'low', 'medium', and 'high'.effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'minimal', 'low', 'medium', and 'high'.chat_tools_effort_none: Unsupported value: 'reasoning_effort' does not support 'none' with this model. Supported values are: 'minimal', 'low', 'medium', and 'high'.verbosity_low: Unsupported value: 'low' is not supported with the '' model. Supported values are: 'medium'.effort_low: Unsupported parameter: 'reasoning.effort' is not supported with this model.verbosity_high: Unsupported value: 'high' is not supported with the '' model. Supported values are: 'medium'.chat_reasoning_effort: Unrecognized request argument supplied: reasoning_effortservice_tier_flex: Invalid service_tier argumentservice_tier_flex: There was an issue with your request. Please check your inputs and try againeffort_xhigh: Unsupported value: 'xhigh' is not supported with the '' model. Supported values are: 'low', 'medium', and 'high'.effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'low', 'medium', and 'high'.code_interpreter: Tool choices other than 'auto' are not supported with model '' and the following tool types: 'code_interpreter'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'low', 'medium', and 'high'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'low', 'medium', and 'high'.web_search: Tool choices other than 'auto' are not supported with model '' and the following tool types: 'web_search_preview'.chat_temperature: This model is only supported in v1/responses and not in v1/chat/completions.effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'high'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'high'.effort_xhigh: Unsupported value: 'xhigh' is not supported with the '' model. Supported values are: 'high'.effort_low: Unsupported value: 'low' is not supported with the '' model. Supported values are: 'high'.effort_medium: Unsupported value: 'medium' is not supported with the '' model. Supported values are: 'high'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'high'.chat_reasoning_effort: Unsupported value: 'reasoning_effort' does not support 'low' with this model. Supported values are: 'medium'.effort_minimal: Unsupported value: 'minimal' is not supported with the '' model. Supported values are: 'medium'.effort_xhigh: Unsupported value: 'xhigh' is not supported with the '' model. Supported values are: 'medium'.effort_none: Unsupported value: 'none' is not supported with the '' model. Supported values are: 'medium'.effort_max: Unsupported value: 'max' is not supported with the '' model. Supported values are: 'medium'.chat_temperature_effort_none: Unsupported value: 'reasoning_effort' does not support 'none' with this model. Supported values are: 'medium'.