[ { "record_type": "rate_limits", "provider": "anthropic", "generated_at": "2026-09-18", "status": [ "DOCUMENTED", "LIVE_DISCOVERED" ], "sources": [ { "url": "https://platform.claude.com/docs/en/api/rate-limits", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/api/service-tiers", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/build-with-claude/fast-mode", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/build-with-claude/token-counting", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/build-with-claude/files", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/api/overview", "retrieved_at": "2026-09-18" }, { "url": "https://platform.claude.com/docs/en/release-notes/overview", "retrieved_at": "2026-09-18" } ], "documented": { "principles": [ "Two limit kinds: spend limits (monthly USD cap per tier) and rate limits (RPM, ITPM, OTPM per model class).", "Limits are per organization; tier assigned automatically from usage history and account standing; new orgs may start in an Evaluation tier below the published numbers.", "Token-bucket algorithm: capacity replenishes continuously (60 RPM may be enforced as 1 request/second; bursts can 429).", "Rate limits are applied separately per model class; different models can be used up to their own limits simultaneously.", "Rate limits are shared across inference_geo values (us and global draw from the same pool).", "Acceleration limits: a sharp usage increase can trigger 429 rate_limit_error (since 2025-08-11; previously 529 overloaded_error). Ramp traffic gradually.", "All limits are maxima, not guaranteed minimums.", "History: tokens-per-minute replaced by ITPM + OTPM on 2024-11-20; dedicated 1M-context rate limits removed 2026-03-13; usage tiers (formerly Tier 1-4) consolidated into Start/Build/Scale on 2026-06-26 with Sonnet/Haiku raised to Opus levels." ], "tiers": { "Start": { "monthly_spend_cap_usd": 500 }, "Build": { "monthly_spend_cap_usd": 1000 }, "Scale": { "monthly_spend_cap_usd": 200000 }, "Custom": { "monthly_spend_cap_usd": null, "note": "no cap; arranged with account team; contact sales via Console Rate limits page" }, "Evaluation": { "note": "starting tier for new/low-history orgs, limits below Start; increases automatically" }, "Claude Platform on AWS": { "note": "same limits; placed on Start, moves up with paid AWS Marketplace invoices; no self-serve increase flow; no per-workspace limits; no fast mode" } }, "messages_api_per_model_class": { "unit": { "rpm": "requests/min", "itpm": "uncached input tokens/min", "otpm": "output tokens/min" }, "classes": { "Claude Fable 5.x (Fable 5.1 + Fable 5 combined; Mythos 5.1 + Mythos 5 share a separate combined limit on the same terms)": { "Start": [ 1000, 500000, 100000 ], "Build": [ 2000, 1500000, 300000 ], "Scale": [ 4000, 4000000, 800000 ] }, "Claude Opus 5 (own bucket)": { "Start": [ 1000, 2000000, 400000 ], "Build": [ 5000, 5000000, 1000000 ], "Scale": [ 10000, 10000000, 2000000 ] }, "Claude Opus 4.x (Opus 4.8 + 4.7 + 4.6 + 4.5 combined)": { "Start": [ 1000, 2000000, 400000 ], "Build": [ 5000, 5000000, 1000000 ], "Scale": [ 10000, 10000000, 2000000 ] }, "Claude Sonnet 5 (own bucket)": { "Start": [ 1000, 2000000, 400000 ], "Build": [ 5000, 5000000, 1000000 ], "Scale": [ 10000, 10000000, 2000000 ] }, "Claude Sonnet 4.x (Sonnet 4.6 + 4.5 combined)": { "Start": [ 1000, 2000000, 400000 ], "Build": [ 5000, 5000000, 1000000 ], "Scale": [ 10000, 10000000, 2000000 ] }, "Claude Haiku 4.5": { "Start": [ 1000, 2000000, 400000 ], "Build": [ 5000, 5000000, 1000000 ], "Scale": [ 10000, 10000000, 2000000 ] }, "Claude Haiku 3.5 (retired on Claude API; counts cache_read_input_tokens toward ITPM)": { "Start": [ 1000, 100000, 20000 ], "Build": [ 2000, 200000, 40000 ], "Scale": [ 4000, 400000, 80000 ] } }, "values_order": [ "rpm", "itpm", "otpm" ] }, "cache_aware_itpm": { "counts_toward_itpm": [ "input_tokens (tokens after the last cache breakpoint)", "cache_creation_input_tokens" ], "does_not_count": [ "cache_read_input_tokens (all models except Claude Haiku 3.5)" ], "estimation": "ITPM estimated at request start, adjusted to actual input tokens during the request", "otpm": "evaluated in real time on actual generated tokens; max_tokens does not count", "total_input_formula": "total_input_tokens = cache_read_input_tokens + cache_creation_input_tokens + input_tokens", "example": "2,000,000 ITPM with 80% cache hit rate ~= 10,000,000 total input tokens/min" }, "message_batches_api": { "note": "shared across all models; RPM applies to all Batches endpoints; queue limit counts individual batch requests not yet processed", "Start": { "rpm": 1000, "max_requests_in_processing_queue": 200000, "max_requests_per_batch": 100000 }, "Build": { "rpm": 2000, "max_requests_in_processing_queue": 300000, "max_requests_per_batch": 100000 }, "Scale": { "rpm": 4000, "max_requests_in_processing_queue": 500000, "max_requests_per_batch": 100000 }, "Custom": "contact sales" }, "token_counting_api": { "Start": { "rpm": 5000 }, "Build": { "rpm": 10000 }, "Scale": { "rpm": 20000 }, "note": "free; separate from Messages limits" }, "files_api": { "rpm_approx": 500, "note": "per organization, shared across upload/list/retrieve/download/delete; contact sales to raise" }, "managed_agents": { "create_endpoints_rpm": 300, "read_endpoints_rpm": 1200, "note": "per organization, separate from Messages API" }, "fast_mode": { "note": "dedicated limits separate from standard Opus limits when speed: fast on Opus 5 / Opus 4.8; 429 + retry-after when exceeded; Opus 4.7 -> error, Opus 4.6 -> standard speed", "headers": [ "anthropic-fast-input-tokens-limit", "anthropic-fast-input-tokens-remaining", "anthropic-fast-input-tokens-reset", "anthropic-fast-output-tokens-limit", "anthropic-fast-output-tokens-remaining", "anthropic-fast-output-tokens-reset" ] }, "programmatic_tool_calling": "each tool call from code execution counts as a separate invocation against the same limits as regular tool calls", "priority_tier": { "status": "no longer sold; existing commitments honoured to contract end", "burndown": { "cache_read": 0.1, "cache_write_5m": 1.25, "cache_write_1h": 2.0, "inference_geo_us_4_6_plus": 1.1, "other_tokens": 1.0 }, "assignment": "service_tier auto (default) uses priority capacity when ITPM and OTPM capacity are both sufficient, else standard; requests still draw from regular rate limits; standard_only opts out", "headers": [ "anthropic-priority-input-tokens-limit", "anthropic-priority-input-tokens-remaining", "anthropic-priority-input-tokens-reset", "anthropic-priority-output-tokens-limit", "anthropic-priority-output-tokens-remaining", "anthropic-priority-output-tokens-reset" ], "unsupported_models": [ "claude-fable-5-1", "claude-mythos-5-1", "claude-mythos-5", "claude-mythos-preview", "claude-opus-5", "claude-sonnet-5" ] }, "workspaces": { "note": "per-workspace spend and rate limits can be set below the org limit (not on the default workspace); per limiter type; org limits always apply; anthropic-workspace-id response header identifies the workspace", "rate_limits_api": "GET org/workspace configured limits programmatically (manage-claude/rate-limits-api; Admin API rate_limits list endpoints)" }, "response_headers": { "retry-after": "seconds to wait; earlier retries fail; NOT sent on the spend-cap 429", "anthropic-ratelimit-requests-limit|remaining|reset": "requests per period; reset in RFC 3339", "anthropic-ratelimit-tokens-limit|remaining|reset": "most restrictive token limit in effect (workspace limit if exceeded, else total input+output); remaining rounded to nearest thousand", "anthropic-ratelimit-input-tokens-limit|remaining|reset": "input tokens per period", "anthropic-ratelimit-output-tokens-limit|remaining|reset": "output tokens per period", "anthropic-priority-*": "Priority Tier only", "anthropic-fast-*": "fast mode only", "request-id": "unique request id (also in error bodies since 2025-08-19)", "anthropic-organization-id": "org id of the credential (since 2025-02-10)", "anthropic-workspace-id": "wrkspc_ id the credential resolved to (since 2026-08-11); absent on Admin API requests" }, "http_429_semantics": { "rate_limit_exceeded": { "status": 429, "error_type": "rate_limit_error", "retry_after": true, "message": "describes which limit was exceeded" }, "acceleration_limit": { "status": 429, "error_type": "rate_limit_error", "note": "sharp usage increase" }, "tier_spend_cap_reached": { "status": 429, "error_type": "rate_limit_error", "details.error_code": "enforced_spend_limit_reached", "retry_after": false, "resumes": "00:00 UTC first day of next month or on tier increase" }, "self_set_spend_limit_reached": { "status": 400, "error_type": "invalid_request_error", "message_prefix": "You have reached your specified API usage limits (or ...workspace API usage limits)" }, "claude_code_workspace_limit": { "status": 429, "retry_after": true, "note": "Claude Code workspace limits are checked separately" }, "overloaded": { "status": 529, "error_type": "overloaded_error", "note": "capacity; Priority Tier reduces these (see api/errors)" } }, "console": "Usage page shows Rate Limit - Input Tokens / Output Tokens charts (hourly max per minute vs limit, cache rate); Rate limits page shows tier and Request rate limit increase" }, "observed_for_this_key": { "note": "Values returned to OUR key on 2026-09-18/19 UTC; account-specific, not general limits. Do not treat as documentation.", "endpoint": "POST /v1/messages (claude-opus-5, claude-sonnet-5, claude-fable-5-1, claude-opus-4-8, claude-sonnet-4-6 all returned identical limit values)", "headers": { "anthropic-ratelimit-input-tokens-limit": "10000000", "anthropic-ratelimit-output-tokens-limit": "2000000", "anthropic-ratelimit-requests-limit": "10000", "anthropic-ratelimit-tokens-limit": "12000000", "anthropic-ratelimit-*-remaining": "present (e.g. requests-remaining 9999 after one call; tokens-remaining 12000000)", "anthropic-ratelimit-*-reset": "present, RFC 3339 (e.g. 2026-09-19T01:44:39Z)", "request-id": "present", "anthropic-organization-id": "present", "CF-RAY": "present (Cloudflare edge, e.g. ...-YUL)" }, "interpretation": "matches the documented Scale-tier Opus/Sonnet numbers (10,000 RPM / 10M ITPM / 2M OTPM); tokens-limit 12,000,000 = ITPM + OTPM as documented for the total-tokens header", "fable_5_1_note": "claude-fable-5-1 returned the same 10M/2M/10k headers as Opus 5 although the documented Fable Scale limits are 4,000 RPM / 4M ITPM / 800k OTPM - account-specific; recorded, not generalized", "GET /v1/models and GET /v1/models/{id}": "no anthropic-ratelimit-* headers returned", "POST /v1/messages/count_tokens": "no anthropic-ratelimit-* headers returned", "GET /v1/organizations/me (Admin API, observed earlier in this run by another agent)": { "anthropic-ratelimit-requests-limit": "100", "note": "only the requests-limit family observed on the Admin API" }, "retry-after": "not observed (no 429 triggered)" }, "_fragment": "generated/fragments/rate-limits/anthropic-rate-limits.json" }, { "record_type": "rate_limits", "provider": "gemini", "generated_at": "2026-09-18", "status": [ "DOCUMENTED", "LIVE_DISCOVERED" ], "sources": [ { "url": "https://ai.google.dev/gemini-api/docs/rate-limits", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/billing", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/priority-inference", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/flex-inference", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/batch-api", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/live-api/session", "retrieved_at": "2026-09-18" }, { "url": "https://ai.google.dev/gemini-api/docs/troubleshooting", "retrieved_at": "2026-09-18" }, { "url": "https://aistudio.google.com/rate-limit", "retrieved_at": null, "note": "per-model RPM/TPM/RPD tables live here (account-specific UI), not in the docs page" } ], "documented": { "principles": [ "Three dimensions: requests per minute (RPM), input tokens per minute (TPM), requests per day (RPD); exceeding any one of them returns 429 RESOURCE_EXHAUSTED.", "Limits are applied per Google Cloud project, not per API key. RPD quotas reset at midnight Pacific time.", "Model-specific extra dimensions exist: images per minute (IPM) for image-generation models, tokens per day (TPD) for some models.", "Experimental and preview models have more restrictive limits.", "Since 2026 the docs page no longer publishes the per-model RPM/TPM/RPD matrix; it points to the AI Studio rate-limit page. 'Specified rate limits are not guaranteed and actual capacity may vary.'", "Tier upgrades Free -> Tier 1 are instant once billing is linked; later upgrades take effect within ~10 minutes; upgrade can be denied 'based on other factors'." ], "usage_tiers": [ { "tier": "Free", "qualification": "Active project or free trial", "billing_tier_spend_cap": null, "spend_rate_limit_per_10_min": null, "notes": "Free of charge on many models; content may be used to improve Google products (terms: Unpaid Services). Pro models (gemini-3.1-pro-preview, gemini-2.5-pro) and paid-only media models are not available (observed limit 0)." }, { "tier": "Tier 1", "qualification": "Set up and link an active Cloud Billing account", "billing_tier_spend_cap": "$250", "spend_rate_limit_per_10_min": "$10" }, { "tier": "Tier 2", "qualification": "Paid $100 cumulative on Google Cloud + 3 days since first successful payment", "billing_tier_spend_cap": "$2,000", "spend_rate_limit_per_10_min": "$50" }, { "tier": "Tier 3", "qualification": "Paid $1,000 cumulative + 30 days since first successful payment", "billing_tier_spend_cap": "$20,000 - $100,000+", "spend_rate_limit_per_10_min": "$200" } ], "spend_based_rate_limits": "Rolling 10-minute window per tier ($10 / $50 / $200); exceeding it returns 429 RESOURCE_EXHAUSTED; whether it applies depends on billing history and account standing.", "priority_inference": { "rate_limit": "0.3x the standard rate limit for each model and tier", "overflow": "graceful server-side downgrade to Standard processing instead of failing", "pricing": "75-100% more than Standard (tables show 1.8x)" }, "flex_inference": { "rate_limit": "own limits; best-effort, sheddable capacity", "latency": "minutes (1-15 min target)", "errors": "429 RESOURCE_EXHAUSTED when capacity is shed; clients must retry (docs: 'Implement retries', per-request timeouts)", "pricing": "50% discount", "supported_models": [ "gemini-3.8-flash", "gemini-3.7-flash", "gemini-3.6-flash", "gemini-3.5-flash-lite", "gemini-3.5-flash", "gemini-3.1-flash-lite", "gemini-3.1-pro-preview", "gemini-3-flash-preview", "gemini-2.5-pro", "gemini-2.5-flash" ] }, "batch_api": { "concurrent_batch_requests": 100, "input_file_size_limit": "2 GB", "file_storage_limit": "20 GB", "turnaround": "target 24 hours (usually much faster)", "pricing": "50% of standard", "enqueued_tokens_per_model": { "note": "Maximum tokens enqueued across all active batch jobs for a given model.", "tier_1": { "gemini-3.1-pro-preview": 5000000, "gemini-3.5-flash-lite": 10000000, "gemini-3.8-flash": 3000000, "gemini-3.7-flash": 3000000, "gemini-3.1-flash-lite": 10000000, "gemini-3.1-flash-lite-preview": 10000000, "gemini-3.6-flash": 3000000, "gemini-3.5-flash": 3000000, "gemini-2.5-pro": 5000000, "gemini-2.5-pro-preview-tts": 25000, "gemini-2.5-flash": 3000000, "gemini-2.5-flash-preview": 3000000, "gemini-2.5-flash-image-preview": 3000000, "gemini-2.5-flash-preview-tts": 100000, "gemini-2.5-flash-lite": 10000000, "gemini-2.5-flash-lite-preview": 10000000, "gemini-2.0-flash": 10000000, "gemini-2.0-flash-image": 3000000, "gemini-2.0-flash-lite": 10000000, "gemini-3.1-flash-image-preview": 1000000, "gemini-3.1-flash-lite-image": 2000000, "gemini-3-pro-image-preview": 2000000, "gemini-embedding": 500000 }, "tier_2": { "gemini-3.1-pro-preview": 500000000, "gemini-3.5-flash-lite": 500000000, "gemini-3.1-flash-lite": 500000000, "gemini-3.1-flash-lite-preview": 500000000, "gemini-3.8-flash": 400000000, "gemini-3.7-flash": 400000000, "gemini-3.6-flash": 400000000, "gemini-3.5-flash": 400000000, "gemini-2.5-pro": 500000000, "gemini-2.5-pro-preview-tts": 100000, "gemini-2.5-flash": 400000000, "gemini-2.5-flash-preview": 400000000, "gemini-2.5-flash-image-preview": 400000000, "gemini-2.5-flash-preview-tts": 100000, "gemini-2.5-flash-lite": 500000000, "gemini-2.5-flash-lite-preview": 500000000, "gemini-2.0-flash": 1000000000, "gemini-2.0-flash-image": 400000000, "gemini-2.0-flash-lite": 1000000000, "gemini-3.1-flash-image-preview": 250000000, "gemini-3.1-flash-lite-image": 270000000, "gemini-3-pro-image-preview": 270000000, "gemini-embedding": 5000000 }, "tier_3": { "gemini-3.1-pro-preview": 1000000000, "gemini-3.5-flash-lite": 1000000000, "gemini-3.1-flash-lite": 1000000000, "gemini-3.1-flash-lite-preview": 1000000000, "gemini-3.8-flash": 1000000000, "gemini-3.7-flash": 1000000000, "gemini-3.6-flash": 1000000000, "gemini-3.5-flash": 1000000000, "gemini-2.5-pro": 1000000000, "gemini-2.5-pro-preview-tts": 1000000, "gemini-2.5-flash": 1000000000, "gemini-2.5-flash-preview": 1000000000, "gemini-2.5-flash-image-preview": 1000000000, "gemini-2.5-flash-preview-tts": 4000000, "gemini-2.5-flash-lite": 1000000000, "gemini-2.5-flash-lite-preview": 1000000000, "gemini-2.0-flash": 5000000000, "gemini-2.0-flash-image": 1000000000, "gemini-2.0-flash-lite": 5000000000, "gemini-3.1-flash-image-preview": 750000000, "gemini-3.1-flash-lite-image": 1000000000, "gemini-3-pro-image-preview": 1000000000, "gemini-embedding": 10000000 }, "docs_note": "The batch table still names shut-down models (2.0 Flash, 2.5 previews) and preview image ids; the GA image ids (gemini-3.1-flash-image, gemini-3-pro-image) are not listed - assume the preview row applies." } }, "live_api_sessions": { "session_lifetime": "audio-only 15 minutes, audio+video 2 minutes without context-window compression; with contextWindowCompression sessions can run longer", "connection_lifetime": "~10 minutes per WebSocket connection; use sessionResumption handles (valid 2 h after the last session termination) to continue", "ephemeral_tokens": "short-lived tokens for client-to-server WebSocket connections (POST /v1beta/auth_tokens); Live API only", "note": "Live API concurrency/session-count limits are not published on the rate-limits page." }, "grounding_free_quotas": { "google_search_gemini_3": "5,000 free search requests per month shared across all Gemini 3.x models, then $14 per 1,000 requests", "google_search_gemini_2_5": "1,500 RPD free (shared Flash/Flash-Lite; free tier 500 RPD), then $35 per 1,000 grounded prompts", "google_maps_gemini_3": "5,000 prompts per month free shared, then $14 per 1,000 search queries", "google_maps_gemini_2_5": "1,500 RPD free (10,000 RPD for Pro), then $25 per 1,000 grounded prompts" }, "model_parameter_ranges_troubleshooting_page": { "candidateCount": "1-8", "temperature": "0.0-1.0 (troubleshooting page; live Model.maxTemperature=2 for most Gemini models, 1 for image models)", "topP": "0.0-1.0" }, "increase_requests": "Paid-tier increase form https://forms.gle/ETzX94k8jf7iSotH9 ('no guarantees')." }, "error_semantics_429": { "http_status": 429, "status": "RESOURCE_EXHAUSTED", "message_pattern": "You exceeded your current quota, please check your plan and billing details. ... * Quota exceeded for metric: generativelanguage.googleapis.com/, limit: , model: ... Please retry in s.", "details": [ { "@type": "type.googleapis.com/google.rpc.Help", "links": [ { "description": "Learn more about Gemini API quotas", "url": "https://ai.google.dev/gemini-api/docs/rate-limits" } ] }, { "@type": "type.googleapis.com/google.rpc.QuotaFailure", "violations": [ { "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_input_token_count", "quotaId": "GenerateContentInputTokensPerModelPerDay-FreeTier", "quotaDimensions": { "location": "global", "model": "gemini-3.1-pro" } }, { "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_requests", "quotaId": "GenerateRequestsPerDayPerProjectPerModel-FreeTier" }, { "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_requests", "quotaId": "GenerateRequestsPerMinutePerProjectPerModel-FreeTier" } ] } ], "quota_metrics_observed": [ "generativelanguage.googleapis.com/generate_content_free_tier_input_token_count", "generativelanguage.googleapis.com/generate_content_free_tier_requests" ], "quota_ids_observed": [ "GenerateContentInputTokensPerModelPerDay-FreeTier", "GenerateRequestsPerDayPerProjectPerModel-FreeTier", "GenerateRequestsPerMinutePerProjectPerModel-FreeTier" ], "retry_hint": "The retry delay is only in the message text ('Please retry in 54.22098241s.'); no Retry-After header was returned.", "retryable": "yes with exponential backoff + jitter (docs); Python SDK retries 429/5xx up to 4 attempts, initial ~1 s, max 60 s" }, "headers": { "documented_rate_limit_headers": "none - the Gemini docs do not document any x-ratelimit-* response headers", "observed_on_generateContent_2026_09_19": [ "Server-Timing: gfet4t7; dur=", "X-Gemini-Service-Tier: standard (also present on 429 responses)", "Alt-Svc", "Vary", "X-Content-Type-Options", "X-Frame-Options", "X-XSS-Protection", "Accept-Ranges", "Transfer-Encoding", "Content-Type: application/json; charset=UTF-8", "Date", "Server: scaffolding on HTTPServer2" ], "observed_on_openai_compat": [ "Server-Timing", "Set-Cookie", "Cache-Control", "Expires", "P3P", "Vary", "X-Content-Type-Options", "X-Frame-Options", "X-XSS-Protection" ], "conclusion": "Rate-limit state (remaining RPM/TPM/RPD) is NOT exposed in response headers; only the 429 body carries quota metric/limit/retry seconds. Usage is visible in AI Studio (Projects / rate-limit pages)." }, "observed_for_this_key": { "note": "Account-specific observations from 2026-09-19 UTC probes; not documentation.", "tier": "Free tier (no billing): 429 with 'limit: 0' on generate_content_free_tier_* metrics for model gemini-3.1-pro (requested as gemini-3.1-pro-preview and via gemini-pro-latest).", "successful_calls": [ "gemini-3.5-flash-lite", "gemini-3.5-flash", "gemini-3.8-flash", "gemma-4-26b-a4b-it", "gemini-flash-latest (-> gemini-3.8-flash)", "OpenAI-compat chat.completions gemini-3.5-flash-lite" ], "restricted_calls": [ "gemini-2.5-flash-lite -> 404 NOT_FOUND 'no longer available to new users'", "gemini-3.1-pro-preview / gemini-pro-latest -> 429 RESOURCE_EXHAUSTED free-tier limit 0" ], "usage_metadata_shape": { "promptTokenCount": 5, "candidatesTokenCount": 2, "totalTokenCount": 7, "promptTokensDetails": [ { "modality": "TEXT", "tokenCount": 5 } ], "thoughtsTokenCount": "present when the model thought (5 on 3.5/3.8 Flash and Gemma 4 with maxOutputTokens=8 -> finishReason MAX_TOKENS, empty text)", "serviceTier": "standard" } }, "_fragment": "generated/fragments/rate-limits/gemini-rate-limits.json" }, { "provider": "openai", "retrieved_at": "2026-09-18", "sources": [ "https://developers.openai.com/api/docs/guides/rate-limits", "https://developers.openai.com/api/docs/models/ (Rate limits section)", "https://developers.openai.com/api/docs/guides/batch", "https://developers.openai.com/api/docs/guides/fast-mode" ], "concepts": [ { "metric": "RPM", "meaning": "requests per minute", "scope": "organization and project, per model (or shared-limit group)" }, { "metric": "RPD", "meaning": "requests per day", "scope": "some models / free tier" }, { "metric": "TPM", "meaning": "tokens per minute (max(max_tokens, estimated prompt tokens) counted per request)", "scope": "per model" }, { "metric": "TPD", "meaning": "tokens per day", "scope": "some models" }, { "metric": "IPM", "meaning": "images per minute", "scope": "image models (gpt-image-*)" }, { "metric": "minutes-of-audio per minute", "meaning": "audio duration admitted per minute", "scope": "streaming audio models (gpt-realtime-whisper, gpt-realtime-translate)" }, { "metric": "concurrent sessions", "meaning": "simultaneous live sessions", "scope": "gpt-live-1 (v1/live/sessions)" }, { "metric": "batch queue limit", "meaning": "total input tokens queued across pending batch jobs for a model; released when the batch completes", "scope": "Batch API, per model" }, { "metric": "long-context limit", "meaning": "separate RPM/TPM/batch-queue table for requests >272K input tokens (GPT-5.4/5.5/5.6/6 Astra)", "scope": "long-context models; visible in the console" }, { "metric": "shared limits", "meaning": "some model families share one limit pool (listed under 'shared limit' in the console)", "scope": "organization" }, { "metric": "project-scoped token limit", "meaning": "optional per-project token limit exposed via x-ratelimit-*-project-tokens headers", "scope": "project" }, { "metric": "monthly usage limit", "meaning": "approved monthly spend ceiling per organization, separate from configurable spend limits", "scope": "organization" }, { "metric": "vector store ingestion", "meaning": "/vector_stores/{id}/files and /file_batches share 300 requests per minute per vector store", "scope": "per vector store" }, { "metric": "ramp-rate (slow_down)", "meaning": "429 rate_limit_error/slow_down when traffic increases too quickly even below RPM/TPM; above ~1M TPM ramp at most +50% every 15 minutes", "scope": "per model" } ], "usage_tiers": [ { "tier": "Free", "qualification": "allowed geography", "usage_limit_usd_per_month": 100 }, { "tier": "Tier 1", "qualification": "$5 paid", "usage_limit_usd_per_month": 100 }, { "tier": "Tier 2", "qualification": "$50 paid", "usage_limit_usd_per_month": 500 }, { "tier": "Tier 3", "qualification": "$100 paid", "usage_limit_usd_per_month": 1000 }, { "tier": "Tier 4", "qualification": "$250 paid", "usage_limit_usd_per_month": 5000 }, { "tier": "Tier 5", "qualification": "$1,000 paid", "usage_limit_usd_per_month": 200000 } ], "headers": [ { "name": "Retry-After", "sample": "56", "description": "minimum seconds to wait before retrying a temporary 429 (slow_down / rate limit) or 503 (server_is_overloaded)" }, { "name": "x-ratelimit-limit-requests", "sample": "60", "description": "max requests before exhausting the rate limit" }, { "name": "x-ratelimit-limit-tokens", "sample": "150000", "description": "max tokens before exhausting the rate limit" }, { "name": "x-ratelimit-remaining-requests", "sample": "59", "description": "remaining requests" }, { "name": "x-ratelimit-remaining-tokens", "sample": "149984", "description": "remaining tokens" }, { "name": "x-ratelimit-reset-requests", "sample": "1s", "description": "time until request limit resets" }, { "name": "x-ratelimit-reset-tokens", "sample": "6m0s", "description": "time until token limit resets" }, { "name": "x-ratelimit-limit-project-tokens", "sample": "60000", "description": "project token limit (present when a project-scoped limit applies)" }, { "name": "x-ratelimit-remaining-project-tokens", "sample": "57000", "description": "remaining project tokens" }, { "name": "x-ratelimit-reset-project-tokens", "sample": "3s", "description": "time until project token limit resets" } ], "errors": [ { "http_status": 429, "type": "rate_limit_error", "code": "slow_down", "meaning": "request rate increased too quickly", "action": "honour Retry-After, reduce rate, ramp gradually" }, { "http_status": 429, "type": "rate_limit_error", "code": "rate_limit_exceeded", "meaning": "RPM/TPM/RPD/TPD/IPM exhausted", "action": "exponential backoff with jitter; batch requests; reduce max_tokens" }, { "http_status": 429, "type": "insufficient_quota", "code": "insufficient_quota", "meaning": "monthly usage / spend limit reached (429 also returned when a hard spend limit is hit)", "action": "do not retry; raise limits" }, { "http_status": 503, "type": "service_unavailable_error", "code": "server_is_overloaded", "meaning": "model temporarily overloaded", "action": "honour Retry-After, retry with increasing delay" } ], "fine_tuning_limits_endpoint": "GET /v1/fine_tuning/model_limits", "capacity_products": [ "Scale Tier (pay-as-you-go traffic routinely hitting ramp limits)", "Reserved Tier (GPT-5.6 and later)", "Ultrafast mode (limited preview, GPT-5.6 Sol, announced 2026-08-13)" ], "per_model_documented_tiers": { "babbage-002": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 10000, "Batch queue limit": 100000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 40000, "Batch queue limit": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 80000, "Batch queue limit": 5000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 300000, "Batch queue limit": 30000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 1000000, "Batch queue limit": 150000000 } } } }, "chat-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "chatgpt-4o-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "chatgpt-image-latest": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "codex-mini-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 1000, "TPM": 100000, "Batch queue limit": 1000000 }, "Tier 2": { "RPM": 2000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "computer-use-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 3": { "RPM": 3000, "TPM": 20000000, "Batch queue limit": 450000000 }, "Tier 4": { "RPM": 3000, "TPM": 20000000, "Batch queue limit": 450000000 }, "Tier 5": { "RPM": 3000, "TPM": 20000000, "Batch queue limit": 450000000 } } } }, "dall-e-3": { "default": { "metrics": [ "RPM" ], "note": null, "tiers": { "Tier free": { "RPM": "1 img/min" }, "Tier 1": { "RPM": "500 img/min" }, "Tier 2": { "RPM": "2500 img/min" }, "Tier 3": { "RPM": "5000 img/min" }, "Tier 4": { "RPM": "7500 img/min" }, "Tier 5": { "RPM": "10000 img/min" } } } }, "davinci-002": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 10000, "Batch queue limit": 100000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 40000, "Batch queue limit": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 80000, "Batch queue limit": 5000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 300000, "Batch queue limit": 30000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 1000000, "Batch queue limit": 150000000 } } } }, "gpt-3.5-turbo": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 5000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 50000000, "Batch queue limit": 10000000000 } } } }, "gpt-4": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 10000, "Batch queue limit": 100000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 40000, "Batch queue limit": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 80000, "Batch queue limit": 5000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 300000, "Batch queue limit": 30000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 1000000, "Batch queue limit": 150000000 } } } }, "gpt-4-turbo": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 600000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 800000, "Batch queue limit": 80000000 }, "Tier 5": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 300000000 } } } }, "gpt-4-turbo-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 600000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 800000, "Batch queue limit": 80000000 }, "Tier 5": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 300000000 } } } }, "gpt-4.1": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "128k input tokens", "tiers": { "Tier 1": { "RPM": 100, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 250, "TPM": 500000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 1000, "TPM": 5000000, "Batch queue limit": 100000000 }, "Tier 5": { "RPM": 4000, "TPM": 10000000, "Batch queue limit": 1000000000 } } } }, "gpt-4.1-mini": { "Standard": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "RPD": null, "TPM": 150000000, "Batch queue limit": 15000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "128k input tokens", "tiers": { "Tier 1": { "RPM": 200, "TPM": 400000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 3": { "RPM": 1000, "TPM": 2000000, "Batch queue limit": 80000000 }, "Tier 4": { "RPM": 2000, "TPM": 10000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 8000, "TPM": 20000000, "Batch queue limit": 2000000000 } } } }, "gpt-4.1-nano": { "Standard": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "RPD": null, "TPM": 150000000, "Batch queue limit": 15000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "128k input tokens", "tiers": { "Tier 1": { "RPM": 200, "TPM": 400000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 3": { "RPM": 1000, "TPM": 2000000, "Batch queue limit": 80000000 }, "Tier 4": { "RPM": 2000, "TPM": 10000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 8000, "TPM": 20000000, "Batch queue limit": 2000000000 } } } }, "gpt-4.5-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 1000, "TPM": 125000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 250000, "Batch queue limit": 500000 }, "Tier 3": { "RPM": 5000, "TPM": 500000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 1000000, "Batch queue limit": 100000000 }, "Tier 5": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 5000000000 } } } }, "gpt-4o": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "gpt-4o-audio-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "gpt-4o-mini": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "RPD": null, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "gpt-4o-mini-audio-preview": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "free": { "RPM": 3, "RPD": 200, "TPM": 40000, "Batch queue limit": null }, "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "RPD": null, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "gpt-4o-mini-realtime-preview": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-4o-mini-search-preview": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "free": { "RPM": 3, "RPD": 200, "TPM": 40000, "Batch queue limit": null }, "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "RPD": null, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "gpt-4o-mini-transcribe": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 50000 }, "Tier 2": { "RPM": 2000, "TPM": 150000 }, "Tier 3": { "RPM": 5000, "TPM": 600000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 8000000 } } } }, "gpt-4o-mini-tts": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 50000 }, "Tier 2": { "RPM": 2000, "TPM": 150000 }, "Tier 3": { "RPM": 5000, "TPM": 600000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 8000000 } } } }, "gpt-4o-realtime-preview": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-4o-search-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 100, "TPM": 30000, "Batch queue limit": 0 }, "Tier 2": { "RPM": 500, "TPM": 45000, "Batch queue limit": 0 }, "Tier 3": { "RPM": 500, "TPM": 80000, "Batch queue limit": 0 }, "Tier 4": { "RPM": 1000, "TPM": 200000, "Batch queue limit": 0 }, "Tier 5": { "RPM": 1000, "TPM": 3000000, "Batch queue limit": 0 } } } }, "gpt-4o-transcribe": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 10000 }, "Tier 2": { "RPM": 2000, "TPM": 100000 }, "Tier 3": { "RPM": 5000, "TPM": 400000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 6000000 } } } }, "gpt-4o-transcribe-diarize": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 10000 }, "Tier 2": { "RPM": 5000, "TPM": 100000 }, "Tier 3": { "RPM": 5000, "TPM": 400000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 6000000 } } } }, "gpt-5": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5-chat-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5-codex": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 10000000, "Batch queue limit": 15000000000 } } } }, "gpt-5-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5-nano": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5-pro": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "gpt-5.1": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.1-chat-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.1-codex": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.1-codex-max": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.1-codex-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.2": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.2-chat-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.2-codex": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.2-pro": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "128k input tokens", "tiers": { "Tier 1": { "RPM": 100, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 250, "TPM": 500000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 1000, "TPM": 5000000, "Batch queue limit": 100000000 }, "Tier 5": { "RPM": 4000, "TPM": 10000000, "Batch queue limit": 1000000000 } } } }, "gpt-5.3-chat-latest": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 50000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.3-codex": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.4": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "272K input tokens", "tiers": { "Tier 1": { "RPM": 200, "TPM": 400000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 3": { "RPM": 1000, "TPM": 2000000, "Batch queue limit": 80000000 }, "Tier 4": { "RPM": 2000, "TPM": 10000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 8000, "TPM": 20000000, "Batch queue limit": 2000000000 } } } }, "gpt-5.4-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.4-nano": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.4-pro": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 50, "TPM": 50000, "Batch queue limit": 900000 }, "Tier 2": { "RPM": 500, "TPM": 100000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 500, "TPM": 200000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 1000, "TPM": 400000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 1500, "TPM": 4000000, "Batch queue limit": 15000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "272K input tokens", "tiers": { "Tier 1": { "RPM": 20, "TPM": 40000, "Batch queue limit": 2000000 }, "Tier 2": { "RPM": 50, "TPM": 100000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 100, "TPM": 200000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 200, "TPM": 1000000, "Batch queue limit": 100000000 }, "Tier 5": { "RPM": 800, "TPM": 2000000, "Batch queue limit": 1000000000 } } } }, "gpt-5.5": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } }, "Long Context": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": "272K input tokens", "tiers": { "Tier 1": { "RPM": 200, "TPM": 400000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 500, "TPM": 1000000, "Batch queue limit": 40000000 }, "Tier 3": { "RPM": 1000, "TPM": 2000000, "Batch queue limit": 80000000 }, "Tier 4": { "RPM": 2000, "TPM": 10000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 8000, "TPM": 20000000, "Batch queue limit": 2000000000 } } } }, "gpt-5.5-pro": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 50, "TPM": 50000, "Batch queue limit": 500000 }, "Tier 2": { "RPM": 500, "TPM": 200000, "Batch queue limit": 1000000 }, "Tier 3": { "RPM": 500, "TPM": 500000, "Batch queue limit": 10000000 }, "Tier 4": { "RPM": 1000, "TPM": 1000000, "Batch queue limit": 20000000 }, "Tier 5": { "RPM": 2000, "TPM": 4000000, "Batch queue limit": 1500000000 } } } }, "gpt-5.6-cyber": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.6-luna": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 5000000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 180000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.6-sol": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-5.6-terra": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-6-astra": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-audio": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "gpt-audio-1.5": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "gpt-audio-mini": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000 }, "Tier 2": { "RPM": 5000, "TPM": 450000 }, "Tier 3": { "RPM": 5000, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000 } } } }, "gpt-daybreak-blue-latest": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-daybreak-red-latest": { "Standard": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 500000, "Batch queue limit": 1500000 }, "Tier 2": { "RPM": 5000, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 3": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 15000, "TPM": 40000000, "Batch queue limit": 15000000000 } } } }, "gpt-image-1": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "gpt-image-1-mini": { "default": { "metrics": [ "IPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "IPM": 5, "TPM": 100000 }, "Tier 2": { "IPM": 20, "TPM": 250000 }, "Tier 3": { "IPM": 50, "TPM": 800000 }, "Tier 4": { "IPM": 150, "TPM": 3000000 }, "Tier 5": { "IPM": 250, "TPM": 8000000 } } } }, "gpt-image-1.5": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "gpt-image-2": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "gpt-image-2.5-flare": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "gpt-image-2.5-sunburst": { "default": { "metrics": [ "TPM", "IPM" ], "note": null, "tiers": { "Tier 1": { "TPM": 100000, "IPM": 5 }, "Tier 2": { "TPM": 250000, "IPM": 20 }, "Tier 3": { "TPM": 800000, "IPM": 50 }, "Tier 4": { "TPM": 3000000, "IPM": 150 }, "Tier 5": { "TPM": 8000000, "IPM": 250 } } } }, "gpt-live-1": { "Concurrent sessions": { "metrics": [ "Concurrent sessions" ], "note": null, "tiers": { "Tier 1": { "Concurrent sessions": 25 }, "Tier 2": { "Concurrent sessions": 50 }, "Tier 3": { "Concurrent sessions": 200 }, "Tier 4": { "Concurrent sessions": 300 }, "Tier 5": { "Concurrent sessions": 500 } } } }, "gpt-live-transcribe": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 60000 }, "Tier 2": { "RPM": 2000, "TPM": 210000 }, "Tier 3": { "RPM": 5000, "TPM": 390000 }, "Tier 4": { "RPM": 10000, "TPM": 600000 }, "Tier 5": { "RPM": 10000, "TPM": 780000 } } } }, "gpt-oss-120b": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 2": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 3": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 4": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 5": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 } } } }, "gpt-oss-20b": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 2": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 3": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 4": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 }, "Tier 5": { "RPM": 0, "TPM": 0, "Batch queue limit": 0 } } } }, "gpt-realtime": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-realtime-1.5": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-realtime-2": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-realtime-2.1": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "RPD": 1000, "TPM": 40000 }, "Tier 2": { "RPM": 400, "RPD": null, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "RPD": null, "TPM": 15000000 } } } }, "gpt-realtime-2.1-mini": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "TPM": 40000 }, "Tier 2": { "RPM": 400, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "TPM": 15000000 } } } }, "gpt-realtime-mini": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 200, "TPM": 40000 }, "Tier 2": { "RPM": 400, "TPM": 200000 }, "Tier 3": { "RPM": 5000, "TPM": 800000 }, "Tier 4": { "RPM": 10000, "TPM": 4000000 }, "Tier 5": { "RPM": 20000, "TPM": 15000000 } } } }, "gpt-realtime-translate": { "default": { "metrics": [ "Minutes-of-audio per minute" ], "note": null, "tiers": { "Tier 1": { "Minutes-of-audio per minute": 50 }, "Tier 2": { "Minutes-of-audio per minute": 200 }, "Tier 3": { "Minutes-of-audio per minute": 400 }, "Tier 4": { "Minutes-of-audio per minute": 650 }, "Tier 5": { "Minutes-of-audio per minute": 850 } } } }, "gpt-realtime-whisper": { "default": { "metrics": [ "Minutes-of-audio per minute" ], "note": null, "tiers": { "Tier 1": { "Minutes-of-audio per minute": 100 }, "Tier 2": { "Minutes-of-audio per minute": 350 }, "Tier 3": { "Minutes-of-audio per minute": 650 }, "Tier 4": { "Minutes-of-audio per minute": 1000 }, "Tier 5": { "Minutes-of-audio per minute": 1300 } } } }, "gpt-transcribe": { "default": { "metrics": [ "RPM", "TPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 200000 }, "Tier 2": { "RPM": 5000, "TPM": 2000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000 } } } }, "o1": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "o1-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 200000, "Batch queue limit": null }, "Tier 2": { "RPM": 5000, "TPM": 2000000, "Batch queue limit": null }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "o1-preview": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": null }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": null }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "o1-pro": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "o3": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "o3-deep-research": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 200000, "Batch queue limit": 200000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 300000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 500000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 2000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 10000000 } } } }, "o3-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 1000, "TPM": 100000, "Batch queue limit": 1000000 }, "Tier 2": { "RPM": 2000, "TPM": 200000, "Batch queue limit": 2000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "o3-pro": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 500, "TPM": 30000, "Batch queue limit": 90000 }, "Tier 2": { "RPM": 5000, "TPM": 450000, "Batch queue limit": 1350000 }, "Tier 3": { "RPM": 5000, "TPM": 800000, "Batch queue limit": 50000000 }, "Tier 4": { "RPM": 10000, "TPM": 2000000, "Batch queue limit": 200000000 }, "Tier 5": { "RPM": 10000, "TPM": 30000000, "Batch queue limit": 5000000000 } } } }, "o4-mini": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 1000, "TPM": 100000, "Batch queue limit": 1000000 }, "Tier 2": { "RPM": 2000, "TPM": 2000000, "Batch queue limit": 2000000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 40000000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 1000000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000, "Batch queue limit": 15000000000 } } } }, "o4-mini-deep-research": { "default": { "metrics": [ "RPM", "TPM", "Batch queue limit" ], "note": null, "tiers": { "Tier 1": { "RPM": 1000, "TPM": 200000, "Batch queue limit": 200000 }, "Tier 2": { "RPM": 2000, "TPM": 2000000, "Batch queue limit": 300000 }, "Tier 3": { "RPM": 5000, "TPM": 4000000, "Batch queue limit": 500000 }, "Tier 4": { "RPM": 10000, "TPM": 10000000, "Batch queue limit": 2000000 }, "Tier 5": { "RPM": 30000, "TPM": 150000000, "Batch queue limit": 10000000 } } } }, "omni-moderation-latest": { "default": { "metrics": [ "RPM", "RPD", "TPM" ], "note": null, "tiers": { "free": { "RPM": 250, "RPD": 5000, "TPM": 10000 }, "Tier 1": { "RPM": 500, "RPD": 10000, "TPM": 10000 }, "Tier 2": { "RPM": 500, "RPD": null, "TPM": 20000 }, "Tier 3": { "RPM": 1000, "RPD": null, "TPM": 50000 }, "Tier 4": { "RPM": 2000, "RPD": null, "TPM": 250000 }, "Tier 5": { "RPM": 5000, "RPD": null, "TPM": 500000 } } } }, "sora-2": { "Standard RPM": { "metrics": [ "RPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 25 }, "Tier 2": { "RPM": 50 }, "Tier 3": { "RPM": 125 }, "Tier 4": { "RPM": 200 }, "Tier 5": { "RPM": 375 } } } }, "sora-2-pro": { "Standard RPM": { "metrics": [ "RPM" ], "note": null, "tiers": { "Tier 1": { "RPM": 10 }, "Tier 2": { "RPM": 25 }, "Tier 3": { "RPM": 50 }, "Tier 4": { "RPM": 75 }, "Tier 5": { "RPM": 150 } } } }, "text-embedding-3-large": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "free": { "RPM": 100, "RPD": 2000, "TPM": 40000, "Batch queue limit": null }, "Tier 1": { "RPM": 3000, "RPD": null, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 1000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 5000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 5000000, "Batch queue limit": 500000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 4000000000 } } } }, "text-embedding-3-small": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "free": { "RPM": 100, "RPD": 2000, "TPM": 40000, "Batch queue limit": null }, "Tier 1": { "RPM": 3000, "RPD": null, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 1000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 5000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 5000000, "Batch queue limit": 500000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 4000000000 } } } }, "text-embedding-ada-002": { "default": { "metrics": [ "RPM", "RPD", "TPM", "Batch queue limit" ], "note": null, "tiers": { "free": { "RPM": 100, "RPD": 2000, "TPM": 40000, "Batch queue limit": null }, "Tier 1": { "RPM": 3000, "RPD": null, "TPM": 1000000, "Batch queue limit": 3000000 }, "Tier 2": { "RPM": 5000, "RPD": null, "TPM": 1000000, "Batch queue limit": 20000000 }, "Tier 3": { "RPM": 5000, "RPD": null, "TPM": 5000000, "Batch queue limit": 100000000 }, "Tier 4": { "RPM": 10000, "RPD": null, "TPM": 5000000, "Batch queue limit": 500000000 }, "Tier 5": { "RPM": 10000, "RPD": null, "TPM": 10000000, "Batch queue limit": 4000000000 } } } }, "tts-1": { "default": { "metrics": [ "RPM", "RPD" ], "note": null, "tiers": { "free": { "RPM": 3, "RPD": 200 }, "Tier 1": { "RPM": 500, "RPD": null }, "Tier 2": { "RPM": 2500, "RPD": null }, "Tier 3": { "RPM": 5000, "RPD": null }, "Tier 4": { "RPM": 7500, "RPD": null }, "Tier 5": { "RPM": 10000, "RPD": null } } } }, "tts-1-hd": { "default": { "metrics": [ "RPM" ], "note": null, "tiers": { "free": { "RPM": null }, "Tier 1": { "RPM": 500 }, "Tier 2": { "RPM": 2500 }, "Tier 3": { "RPM": 5000 }, "Tier 4": { "RPM": 7500 }, "Tier 5": { "RPM": 10000 } } } }, "whisper-1": { "default": { "metrics": [ "RPM", "RPD" ], "note": null, "tiers": { "free": { "RPM": 3, "RPD": 200 }, "Tier 1": { "RPM": 500, "RPD": null }, "Tier 2": { "RPM": 2500, "RPD": null }, "Tier 3": { "RPM": 5000, "RPD": null }, "Tier 4": { "RPM": 7500, "RPD": null }, "Tier 5": { "RPM": 10000, "RPD": null } } } } }, "per_model_notes": { "gpt-live-1": [ "Rate limits are measured in concurrent sessions.", "Unsupported usage tiers: Free." ] }, "observed": [ { "date": "2026-09-18", "request": "POST /v1/responses (this atlas key)", "headers": { "x-ratelimit-limit-requests": "30000", "x-ratelimit-limit-tokens": "180000000" }, "note": "observed for OUR key/org/project on 2026-09-18 — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-5.6-luna", "headers": { "x-ratelimit-limit-requests": "30000", "x-ratelimit-limit-tokens": "180000000", "x-ratelimit-remaining-requests": "29999", "x-ratelimit-remaining-tokens": "180000000", "x-ratelimit-reset-requests": "2ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-5.6-sol", "headers": { "x-ratelimit-limit-requests": "15000", "x-ratelimit-limit-tokens": "40000000", "x-ratelimit-remaining-requests": "14999", "x-ratelimit-remaining-tokens": "40000000", "x-ratelimit-reset-requests": "4ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-5.6-terra", "headers": { "x-ratelimit-limit-requests": "15000", "x-ratelimit-limit-tokens": "40000000", "x-ratelimit-remaining-requests": "14999", "x-ratelimit-remaining-tokens": "40000000", "x-ratelimit-reset-requests": "4ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-6-astra", "headers": { "x-ratelimit-limit-requests": "15000", "x-ratelimit-limit-tokens": "40000000", "x-ratelimit-remaining-requests": "14999", "x-ratelimit-remaining-tokens": "40000000", "x-ratelimit-reset-requests": "4ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-5.5", "headers": { "x-ratelimit-limit-requests": "15000", "x-ratelimit-limit-tokens": "40000000", "x-ratelimit-remaining-requests": "14999", "x-ratelimit-remaining-tokens": "40000000", "x-ratelimit-reset-requests": "4ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" }, { "date": "2026-09-19", "request": "POST /v1/responses model=gpt-5.4-mini", "headers": { "x-ratelimit-limit-requests": "30000", "x-ratelimit-limit-tokens": "180000000", "x-ratelimit-remaining-requests": "29999", "x-ratelimit-remaining-tokens": "180000000", "x-ratelimit-reset-requests": "2ms", "x-ratelimit-reset-tokens": "0s" }, "note": "observed for OUR key — account-specific, do not generalize" } ], "_fragment": "generated/fragments/rate-limits/openai-rate-limits.json" }, { "record_type": "rate_limits", "provider": "xai", "generated_at": "2026-09-18", "status": [ "DOCUMENTED", "LIVE_DISCOVERED" ], "sources": [ { "url": "https://docs.x.ai/developers/rate-limits", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/management-api-guide", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/debugging", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/pricing", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/advanced-api-usage/batch-api", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.6", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.5", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.3", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.20-0309-reasoning", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-build-0.1", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-4.20-multi-agent-0309", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-imagine-image", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-imagine-image-quality", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-imagine-image-2.0", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-imagine-video", "retrieved_at": "2026-09-19" }, { "url": "https://docs.x.ai/developers/models/grok-imagine-video-1.5", "retrieved_at": "2026-09-19" } ], "documented": { "principles": [ "Every team has per-model limits on two dimensions: requests per second (RPS) and tokens per minute (TPM). RPS is derived from the per-minute request budget (RPM/60): a full minute of requests cannot be spent in one second.", "Limits scale with the team's tier, determined by cumulative spend on the xAI API since 2026-01-01 (prepaid purchases or fulfilled invoices). Tiers unlock automatically and never downgrade.", "Tiers apply to text, embedding and voice models. Imagine (image/video) limits are RPS-only and are raised via sales@x.ai.", "TPM counts prompt tokens (text, image, audio), completion tokens, reasoning tokens and cached prompt tokens (cached tokens still count although billed less).", "Exceeding any limit -> HTTP 429 Too Many Requests (gRPC RESOURCE_EXHAUSTED). Use exponential backoff.", "Batch API requests do not count towards rate limits.", "Per-API-key caps (qps, qpm, tpm) can additionally be set via the Management API (POST/PUT api-keys); the tpm limiter engages when strictly exceeded and does not abort in-flight requests.", "Team-level personalized limits are shown on the console Models page (https://console.x.ai/team/default/models)." ], "tiers": { "Tier 0": { "spend_threshold_usd": 0, "note": "default" }, "Tier 1": { "spend_threshold_usd": 50 }, "Tier 2": { "spend_threshold_usd": 250 }, "Tier 3": { "spend_threshold_usd": 1000 }, "Tier 4": { "spend_threshold_usd": 5000 }, "Enterprise": { "note": "on request (sales@x.ai)" } }, "per_model": { "grok-4.6": { "rps_T0_T4": [ 150, 172, 208, 312, 500 ], "tpm_T0_T4": [ "50M", "53M", "60M", "74M", "100M" ] }, "grok-4.5": { "rps_T0_T4": [ 150, 172, 208, 312, 500 ], "tpm_T0_T4": [ "50M", "53M", "60M", "74M", "100M" ] }, "grok-4.3": { "rps_T0_T4": [ 37, 50, 75, 125, 208 ], "tpm_T0_T4": [ "10M", "15M", "25M", "45M", "85M" ] }, "grok-4.20-0309-reasoning": { "rps_T0_T4": [ 37, 50, 75, 125, 208 ], "tpm_T0_T4": [ "10M", "15M", "25M", "45M", "85M" ] }, "grok-4.20-0309-non-reasoning": { "rps_T0_T4": [ 37, 50, 75, 125, 208 ], "tpm_T0_T4": [ "10M", "15M", "25M", "45M", "85M" ] }, "grok-build-0.1": { "rps_T0_T4": [ 37, 50, 75, 125, 208 ], "tpm_T0_T4": [ "10M", "15M", "25M", "45M", "85M" ] }, "grok-4.20-multi-agent-0309": { "rps_T0_T4": [ 9, 12, 18, 31, 56 ], "tpm_T0_T4": [ "2.5M", "3.7M", "6.2M", "11M", "21M" ] }, "grok-imagine-image": { "rps_T0_T4": [ 6, 12, 25, 50, 100 ], "tpm_T0_T4": null }, "grok-imagine-image-quality": { "rps_T0_T4": [ 6, 12, 25, 50, 100 ], "tpm_T0_T4": null }, "grok-imagine-image-2.0": { "rps_T0_T4": [ 6, 12, 25, 50, 100 ], "tpm_T0_T4": null }, "grok-imagine-video": { "rps_T0_T4": [ 10, 20, 39, 79, 158 ], "tpm_T0_T4": null }, "grok-imagine-video-1.5": { "rps_T0_T4": [ 10, 20, 39, 79, 158 ], "tpm_T0_T4": null } }, "voice_and_audio": { "grok-voice-think-fast-2.0": { "rps": null, "concurrent_sessions_T0_T4": [ 10, 20, 50, 100, 200 ], "max_session_minutes": 120 }, "text-to-speech": { "rps_T0_T4": [ 50, 50, 100, 250, 500 ], "concurrent_sessions_T0_T4": [ 100, 200, 200, 300, 500 ] }, "speech-to-text": { "rps_T0_T4": [ 10, 10, 20, 30, 40 ], "concurrent_sessions_T0_T4": [ 100, 200, 200, 300, 500 ], "note": "concurrent sessions apply to streaming" } }, "increasing_limits": [ "spend more (automatic)", "request an increase in the console Models page", "sales@x.ai for enterprise capacity" ], "http_429_semantics": "429 on inference endpoints when RPS or TPM is exceeded; no documented Retry-After header (none observed in this run since no 429 was triggered)" }, "observed_for_this_key": { "note": "Headers returned to OUR key on 2026-09-19 (Tier 0 team, spend < $50). Account-specific; not documentation.", "headers_seen": [ "x-ratelimit-limit-requests", "x-ratelimit-remaining-requests", "x-ratelimit-limit-tokens", "x-ratelimit-remaining-tokens", "x-request-id", "Server-Timing (cfEdge/cfOrigin)", "CF-RAY (Cloudflare edge, e.g. ...-YUL)" ], "headers_not_seen": [ "x-ratelimit-reset-*", "retry-after (no 429 triggered)" ], "per_model": { "grok-4.6": { "x-ratelimit-limit-requests": "7200", "x-ratelimit-limit-tokens": "50000000" }, "grok-4.5": { "x-ratelimit-limit-requests": "7200", "x-ratelimit-limit-tokens": "50000000" }, "grok-4.3": { "x-ratelimit-limit-requests": "1800", "x-ratelimit-limit-tokens": "10000000" }, "grok-4.20-0309-non-reasoning": { "x-ratelimit-limit-requests": "1800", "x-ratelimit-limit-tokens": "10000000" }, "grok-build-0.1": { "x-ratelimit-limit-requests": "1800", "x-ratelimit-limit-tokens": "10000000" } }, "tokenize_text": { "x-ratelimit-limit-requests": "1800", "x-ratelimit-limit-tokens": "absent" }, "list_endpoints": "GET /v1/models, /v1/language-models, /v1/api-key, /v1/me return no x-ratelimit-* headers", "interpretation": "x-ratelimit-limit-tokens matches the documented Tier 0 TPM (50M for grok-4.6/4.5, 10M for grok-4.3/4.20/build). x-ratelimit-limit-requests looks like a per-MINUTE request budget (7200 -> 120 RPS; 1800 -> 30 RPS), lower than the documented Tier 0 RPS (150 / 37) - the docs say RPS = RPM/60, so the header is probably the RPM budget and our team sits slightly below the published T0 figures. Not verified further.", "grok-4.20-multi-agent-0309": "POST /v1/responses returned no x-ratelimit-* headers in our probe" }, "_fragment": "generated/fragments/rate-limits/xai-rate-limits.json" } ]