Python 88.3%
TypeScript 7.6%
Shell 4.1%
1[2 {3 "record_type": "rate_limits",4 "provider": "anthropic",5 "generated_at": "2026-09-18",6 "status": [7 "DOCUMENTED",8 "LIVE_DISCOVERED"9 ],10 "sources": [11 {12 "url": "https://platform.claude.com/docs/en/api/rate-limits",13 "retrieved_at": "2026-09-18"14 },15 {16 "url": "https://platform.claude.com/docs/en/api/service-tiers",17 "retrieved_at": "2026-09-18"18 },19 {20 "url": "https://platform.claude.com/docs/en/build-with-claude/fast-mode",21 "retrieved_at": "2026-09-18"22 },23 {24 "url": "https://platform.claude.com/docs/en/build-with-claude/token-counting",25 "retrieved_at": "2026-09-18"26 },27 {28 "url": "https://platform.claude.com/docs/en/build-with-claude/files",29 "retrieved_at": "2026-09-18"30 },31 {32 "url": "https://platform.claude.com/docs/en/api/overview",33 "retrieved_at": "2026-09-18"34 },35 {36 "url": "https://platform.claude.com/docs/en/release-notes/overview",37 "retrieved_at": "2026-09-18"38 }39 ],40 "documented": {41 "principles": [42 "Two limit kinds: spend limits (monthly USD cap per tier) and rate limits (RPM, ITPM, OTPM per model class).",43 "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.",44 "Token-bucket algorithm: capacity replenishes continuously (60 RPM may be enforced as 1 request/second; bursts can 429).",45 "Rate limits are applied separately per model class; different models can be used up to their own limits simultaneously.",46 "Rate limits are shared across inference_geo values (us and global draw from the same pool).",47 "Acceleration limits: a sharp usage increase can trigger 429 rate_limit_error (since 2025-08-11; previously 529 overloaded_error). Ramp traffic gradually.",48 "All limits are maxima, not guaranteed minimums.",49 "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."50 ],51 "tiers": {52 "Start": {53 "monthly_spend_cap_usd": 50054 },55 "Build": {56 "monthly_spend_cap_usd": 100057 },58 "Scale": {59 "monthly_spend_cap_usd": 20000060 },61 "Custom": {62 "monthly_spend_cap_usd": null,63 "note": "no cap; arranged with account team; contact sales via Console Rate limits page"64 },65 "Evaluation": {66 "note": "starting tier for new/low-history orgs, limits below Start; increases automatically"67 },68 "Claude Platform on AWS": {69 "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"70 }71 },72 "messages_api_per_model_class": {73 "unit": {74 "rpm": "requests/min",75 "itpm": "uncached input tokens/min",76 "otpm": "output tokens/min"77 },78 "classes": {79 "Claude Fable 5.x (Fable 5.1 + Fable 5 combined; Mythos 5.1 + Mythos 5 share a separate combined limit on the same terms)": {80 "Start": [81 1000,82 500000,83 10000084 ],85 "Build": [86 2000,87 1500000,88 30000089 ],90 "Scale": [91 4000,92 4000000,93 80000094 ]95 },96 "Claude Opus 5 (own bucket)": {97 "Start": [98 1000,99 2000000,100 400000101 ],102 "Build": [103 5000,104 5000000,105 1000000106 ],107 "Scale": [108 10000,109 10000000,110 2000000111 ]112 },113 "Claude Opus 4.x (Opus 4.8 + 4.7 + 4.6 + 4.5 combined)": {114 "Start": [115 1000,116 2000000,117 400000118 ],119 "Build": [120 5000,121 5000000,122 1000000123 ],124 "Scale": [125 10000,126 10000000,127 2000000128 ]129 },130 "Claude Sonnet 5 (own bucket)": {131 "Start": [132 1000,133 2000000,134 400000135 ],136 "Build": [137 5000,138 5000000,139 1000000140 ],141 "Scale": [142 10000,143 10000000,144 2000000145 ]146 },147 "Claude Sonnet 4.x (Sonnet 4.6 + 4.5 combined)": {148 "Start": [149 1000,150 2000000,151 400000152 ],153 "Build": [154 5000,155 5000000,156 1000000157 ],158 "Scale": [159 10000,160 10000000,161 2000000162 ]163 },164 "Claude Haiku 4.5": {165 "Start": [166 1000,167 2000000,168 400000169 ],170 "Build": [171 5000,172 5000000,173 1000000174 ],175 "Scale": [176 10000,177 10000000,178 2000000179 ]180 },181 "Claude Haiku 3.5 (retired on Claude API; counts cache_read_input_tokens toward ITPM)": {182 "Start": [183 1000,184 100000,185 20000186 ],187 "Build": [188 2000,189 200000,190 40000191 ],192 "Scale": [193 4000,194 400000,195 80000196 ]197 }198 },199 "values_order": [200 "rpm",201 "itpm",202 "otpm"203 ]204 },205 "cache_aware_itpm": {206 "counts_toward_itpm": [207 "input_tokens (tokens after the last cache breakpoint)",208 "cache_creation_input_tokens"209 ],210 "does_not_count": [211 "cache_read_input_tokens (all models except Claude Haiku 3.5)"212 ],213 "estimation": "ITPM estimated at request start, adjusted to actual input tokens during the request",214 "otpm": "evaluated in real time on actual generated tokens; max_tokens does not count",215 "total_input_formula": "total_input_tokens = cache_read_input_tokens + cache_creation_input_tokens + input_tokens",216 "example": "2,000,000 ITPM with 80% cache hit rate ~= 10,000,000 total input tokens/min"217 },218 "message_batches_api": {219 "note": "shared across all models; RPM applies to all Batches endpoints; queue limit counts individual batch requests not yet processed",220 "Start": {221 "rpm": 1000,222 "max_requests_in_processing_queue": 200000,223 "max_requests_per_batch": 100000224 },225 "Build": {226 "rpm": 2000,227 "max_requests_in_processing_queue": 300000,228 "max_requests_per_batch": 100000229 },230 "Scale": {231 "rpm": 4000,232 "max_requests_in_processing_queue": 500000,233 "max_requests_per_batch": 100000234 },235 "Custom": "contact sales"236 },237 "token_counting_api": {238 "Start": {239 "rpm": 5000240 },241 "Build": {242 "rpm": 10000243 },244 "Scale": {245 "rpm": 20000246 },247 "note": "free; separate from Messages limits"248 },249 "files_api": {250 "rpm_approx": 500,251 "note": "per organization, shared across upload/list/retrieve/download/delete; contact sales to raise"252 },253 "managed_agents": {254 "create_endpoints_rpm": 300,255 "read_endpoints_rpm": 1200,256 "note": "per organization, separate from Messages API"257 },258 "fast_mode": {259 "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",260 "headers": [261 "anthropic-fast-input-tokens-limit",262 "anthropic-fast-input-tokens-remaining",263 "anthropic-fast-input-tokens-reset",264 "anthropic-fast-output-tokens-limit",265 "anthropic-fast-output-tokens-remaining",266 "anthropic-fast-output-tokens-reset"267 ]268 },269 "programmatic_tool_calling": "each tool call from code execution counts as a separate invocation against the same limits as regular tool calls",270 "priority_tier": {271 "status": "no longer sold; existing commitments honoured to contract end",272 "burndown": {273 "cache_read": 0.1,274 "cache_write_5m": 1.25,275 "cache_write_1h": 2.0,276 "inference_geo_us_4_6_plus": 1.1,277 "other_tokens": 1.0278 },279 "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",280 "headers": [281 "anthropic-priority-input-tokens-limit",282 "anthropic-priority-input-tokens-remaining",283 "anthropic-priority-input-tokens-reset",284 "anthropic-priority-output-tokens-limit",285 "anthropic-priority-output-tokens-remaining",286 "anthropic-priority-output-tokens-reset"287 ],288 "unsupported_models": [289 "claude-fable-5-1",290 "claude-mythos-5-1",291 "claude-mythos-5",292 "claude-mythos-preview",293 "claude-opus-5",294 "claude-sonnet-5"295 ]296 },297 "workspaces": {298 "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",299 "rate_limits_api": "GET org/workspace configured limits programmatically (manage-claude/rate-limits-api; Admin API rate_limits list endpoints)"300 },301 "response_headers": {302 "retry-after": "seconds to wait; earlier retries fail; NOT sent on the spend-cap 429",303 "anthropic-ratelimit-requests-limit|remaining|reset": "requests per period; reset in RFC 3339",304 "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",305 "anthropic-ratelimit-input-tokens-limit|remaining|reset": "input tokens per period",306 "anthropic-ratelimit-output-tokens-limit|remaining|reset": "output tokens per period",307 "anthropic-priority-*": "Priority Tier only",308 "anthropic-fast-*": "fast mode only",309 "request-id": "unique request id (also in error bodies since 2025-08-19)",310 "anthropic-organization-id": "org id of the credential (since 2025-02-10)",311 "anthropic-workspace-id": "wrkspc_ id the credential resolved to (since 2026-08-11); absent on Admin API requests"312 },313 "http_429_semantics": {314 "rate_limit_exceeded": {315 "status": 429,316 "error_type": "rate_limit_error",317 "retry_after": true,318 "message": "describes which limit was exceeded"319 },320 "acceleration_limit": {321 "status": 429,322 "error_type": "rate_limit_error",323 "note": "sharp usage increase"324 },325 "tier_spend_cap_reached": {326 "status": 429,327 "error_type": "rate_limit_error",328 "details.error_code": "enforced_spend_limit_reached",329 "retry_after": false,330 "resumes": "00:00 UTC first day of next month or on tier increase"331 },332 "self_set_spend_limit_reached": {333 "status": 400,334 "error_type": "invalid_request_error",335 "message_prefix": "You have reached your specified API usage limits (or ...workspace API usage limits)"336 },337 "claude_code_workspace_limit": {338 "status": 429,339 "retry_after": true,340 "note": "Claude Code workspace limits are checked separately"341 },342 "overloaded": {343 "status": 529,344 "error_type": "overloaded_error",345 "note": "capacity; Priority Tier reduces these (see api/errors)"346 }347 },348 "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"349 },350 "observed_for_this_key": {351 "note": "Values returned to OUR key on 2026-09-18/19 UTC; account-specific, not general limits. Do not treat as documentation.",352 "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)",353 "headers": {354 "anthropic-ratelimit-input-tokens-limit": "10000000",355 "anthropic-ratelimit-output-tokens-limit": "2000000",356 "anthropic-ratelimit-requests-limit": "10000",357 "anthropic-ratelimit-tokens-limit": "12000000",358 "anthropic-ratelimit-*-remaining": "present (e.g. requests-remaining 9999 after one call; tokens-remaining 12000000)",359 "anthropic-ratelimit-*-reset": "present, RFC 3339 (e.g. 2026-09-19T01:44:39Z)",360 "request-id": "present",361 "anthropic-organization-id": "present",362 "CF-RAY": "present (Cloudflare edge, e.g. ...-YUL)"363 },364 "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",365 "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",366 "GET /v1/models and GET /v1/models/{id}": "no anthropic-ratelimit-* headers returned",367 "POST /v1/messages/count_tokens": "no anthropic-ratelimit-* headers returned",368 "GET /v1/organizations/me (Admin API, observed earlier in this run by another agent)": {369 "anthropic-ratelimit-requests-limit": "100",370 "note": "only the requests-limit family observed on the Admin API"371 },372 "retry-after": "not observed (no 429 triggered)"373 },374 "_fragment": "generated/fragments/rate-limits/anthropic-rate-limits.json"375 },376 {377 "record_type": "rate_limits",378 "provider": "gemini",379 "generated_at": "2026-09-18",380 "status": [381 "DOCUMENTED",382 "LIVE_DISCOVERED"383 ],384 "sources": [385 {386 "url": "https://ai.google.dev/gemini-api/docs/rate-limits",387 "retrieved_at": "2026-09-18"388 },389 {390 "url": "https://ai.google.dev/gemini-api/docs/billing",391 "retrieved_at": "2026-09-18"392 },393 {394 "url": "https://ai.google.dev/gemini-api/docs/priority-inference",395 "retrieved_at": "2026-09-18"396 },397 {398 "url": "https://ai.google.dev/gemini-api/docs/flex-inference",399 "retrieved_at": "2026-09-18"400 },401 {402 "url": "https://ai.google.dev/gemini-api/docs/batch-api",403 "retrieved_at": "2026-09-18"404 },405 {406 "url": "https://ai.google.dev/gemini-api/docs/live-api/session",407 "retrieved_at": "2026-09-18"408 },409 {410 "url": "https://ai.google.dev/gemini-api/docs/troubleshooting",411 "retrieved_at": "2026-09-18"412 },413 {414 "url": "https://aistudio.google.com/rate-limit",415 "retrieved_at": null,416 "note": "per-model RPM/TPM/RPD tables live here (account-specific UI), not in the docs page"417 }418 ],419 "documented": {420 "principles": [421 "Three dimensions: requests per minute (RPM), input tokens per minute (TPM), requests per day (RPD); exceeding any one of them returns 429 RESOURCE_EXHAUSTED.",422 "Limits are applied per Google Cloud project, not per API key. RPD quotas reset at midnight Pacific time.",423 "Model-specific extra dimensions exist: images per minute (IPM) for image-generation models, tokens per day (TPD) for some models.",424 "Experimental and preview models have more restrictive limits.",425 "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.'",426 "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'."427 ],428 "usage_tiers": [429 {430 "tier": "Free",431 "qualification": "Active project or free trial",432 "billing_tier_spend_cap": null,433 "spend_rate_limit_per_10_min": null,434 "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)."435 },436 {437 "tier": "Tier 1",438 "qualification": "Set up and link an active Cloud Billing account",439 "billing_tier_spend_cap": "$250",440 "spend_rate_limit_per_10_min": "$10"441 },442 {443 "tier": "Tier 2",444 "qualification": "Paid $100 cumulative on Google Cloud + 3 days since first successful payment",445 "billing_tier_spend_cap": "$2,000",446 "spend_rate_limit_per_10_min": "$50"447 },448 {449 "tier": "Tier 3",450 "qualification": "Paid $1,000 cumulative + 30 days since first successful payment",451 "billing_tier_spend_cap": "$20,000 - $100,000+",452 "spend_rate_limit_per_10_min": "$200"453 }454 ],455 "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.",456 "priority_inference": {457 "rate_limit": "0.3x the standard rate limit for each model and tier",458 "overflow": "graceful server-side downgrade to Standard processing instead of failing",459 "pricing": "75-100% more than Standard (tables show 1.8x)"460 },461 "flex_inference": {462 "rate_limit": "own limits; best-effort, sheddable capacity",463 "latency": "minutes (1-15 min target)",464 "errors": "429 RESOURCE_EXHAUSTED when capacity is shed; clients must retry (docs: 'Implement retries', per-request timeouts)",465 "pricing": "50% discount",466 "supported_models": [467 "gemini-3.8-flash",468 "gemini-3.7-flash",469 "gemini-3.6-flash",470 "gemini-3.5-flash-lite",471 "gemini-3.5-flash",472 "gemini-3.1-flash-lite",473 "gemini-3.1-pro-preview",474 "gemini-3-flash-preview",475 "gemini-2.5-pro",476 "gemini-2.5-flash"477 ]478 },479 "batch_api": {480 "concurrent_batch_requests": 100,481 "input_file_size_limit": "2 GB",482 "file_storage_limit": "20 GB",483 "turnaround": "target 24 hours (usually much faster)",484 "pricing": "50% of standard",485 "enqueued_tokens_per_model": {486 "note": "Maximum tokens enqueued across all active batch jobs for a given model.",487 "tier_1": {488 "gemini-3.1-pro-preview": 5000000,489 "gemini-3.5-flash-lite": 10000000,490 "gemini-3.8-flash": 3000000,491 "gemini-3.7-flash": 3000000,492 "gemini-3.1-flash-lite": 10000000,493 "gemini-3.1-flash-lite-preview": 10000000,494 "gemini-3.6-flash": 3000000,495 "gemini-3.5-flash": 3000000,496 "gemini-2.5-pro": 5000000,497 "gemini-2.5-pro-preview-tts": 25000,498 "gemini-2.5-flash": 3000000,499 "gemini-2.5-flash-preview": 3000000,500 "gemini-2.5-flash-image-preview": 3000000,501 "gemini-2.5-flash-preview-tts": 100000,502 "gemini-2.5-flash-lite": 10000000,503 "gemini-2.5-flash-lite-preview": 10000000,504 "gemini-2.0-flash": 10000000,505 "gemini-2.0-flash-image": 3000000,506 "gemini-2.0-flash-lite": 10000000,507 "gemini-3.1-flash-image-preview": 1000000,508 "gemini-3.1-flash-lite-image": 2000000,509 "gemini-3-pro-image-preview": 2000000,510 "gemini-embedding": 500000511 },512 "tier_2": {513 "gemini-3.1-pro-preview": 500000000,514 "gemini-3.5-flash-lite": 500000000,515 "gemini-3.1-flash-lite": 500000000,516 "gemini-3.1-flash-lite-preview": 500000000,517 "gemini-3.8-flash": 400000000,518 "gemini-3.7-flash": 400000000,519 "gemini-3.6-flash": 400000000,520 "gemini-3.5-flash": 400000000,521 "gemini-2.5-pro": 500000000,522 "gemini-2.5-pro-preview-tts": 100000,523 "gemini-2.5-flash": 400000000,524 "gemini-2.5-flash-preview": 400000000,525 "gemini-2.5-flash-image-preview": 400000000,526 "gemini-2.5-flash-preview-tts": 100000,527 "gemini-2.5-flash-lite": 500000000,528 "gemini-2.5-flash-lite-preview": 500000000,529 "gemini-2.0-flash": 1000000000,530 "gemini-2.0-flash-image": 400000000,531 "gemini-2.0-flash-lite": 1000000000,532 "gemini-3.1-flash-image-preview": 250000000,533 "gemini-3.1-flash-lite-image": 270000000,534 "gemini-3-pro-image-preview": 270000000,535 "gemini-embedding": 5000000536 },537 "tier_3": {538 "gemini-3.1-pro-preview": 1000000000,539 "gemini-3.5-flash-lite": 1000000000,540 "gemini-3.1-flash-lite": 1000000000,541 "gemini-3.1-flash-lite-preview": 1000000000,542 "gemini-3.8-flash": 1000000000,543 "gemini-3.7-flash": 1000000000,544 "gemini-3.6-flash": 1000000000,545 "gemini-3.5-flash": 1000000000,546 "gemini-2.5-pro": 1000000000,547 "gemini-2.5-pro-preview-tts": 1000000,548 "gemini-2.5-flash": 1000000000,549 "gemini-2.5-flash-preview": 1000000000,550 "gemini-2.5-flash-image-preview": 1000000000,551 "gemini-2.5-flash-preview-tts": 4000000,552 "gemini-2.5-flash-lite": 1000000000,553 "gemini-2.5-flash-lite-preview": 1000000000,554 "gemini-2.0-flash": 5000000000,555 "gemini-2.0-flash-image": 1000000000,556 "gemini-2.0-flash-lite": 5000000000,557 "gemini-3.1-flash-image-preview": 750000000,558 "gemini-3.1-flash-lite-image": 1000000000,559 "gemini-3-pro-image-preview": 1000000000,560 "gemini-embedding": 10000000561 },562 "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."563 }564 },565 "live_api_sessions": {566 "session_lifetime": "audio-only 15 minutes, audio+video 2 minutes without context-window compression; with contextWindowCompression sessions can run longer",567 "connection_lifetime": "~10 minutes per WebSocket connection; use sessionResumption handles (valid 2 h after the last session termination) to continue",568 "ephemeral_tokens": "short-lived tokens for client-to-server WebSocket connections (POST /v1beta/auth_tokens); Live API only",569 "note": "Live API concurrency/session-count limits are not published on the rate-limits page."570 },571 "grounding_free_quotas": {572 "google_search_gemini_3": "5,000 free search requests per month shared across all Gemini 3.x models, then $14 per 1,000 requests",573 "google_search_gemini_2_5": "1,500 RPD free (shared Flash/Flash-Lite; free tier 500 RPD), then $35 per 1,000 grounded prompts",574 "google_maps_gemini_3": "5,000 prompts per month free shared, then $14 per 1,000 search queries",575 "google_maps_gemini_2_5": "1,500 RPD free (10,000 RPD for Pro), then $25 per 1,000 grounded prompts"576 },577 "model_parameter_ranges_troubleshooting_page": {578 "candidateCount": "1-8",579 "temperature": "0.0-1.0 (troubleshooting page; live Model.maxTemperature=2 for most Gemini models, 1 for image models)",580 "topP": "0.0-1.0"581 },582 "increase_requests": "Paid-tier increase form https://forms.gle/ETzX94k8jf7iSotH9 ('no guarantees')."583 },584 "error_semantics_429": {585 "http_status": 429,586 "status": "RESOURCE_EXHAUSTED",587 "message_pattern": "You exceeded your current quota, please check your plan and billing details. ... * Quota exceeded for metric: generativelanguage.googleapis.com/<metric>, limit: <n>, model: <model> ... Please retry in <seconds>s.",588 "details": [589 {590 "@type": "type.googleapis.com/google.rpc.Help",591 "links": [592 {593 "description": "Learn more about Gemini API quotas",594 "url": "https://ai.google.dev/gemini-api/docs/rate-limits"595 }596 ]597 },598 {599 "@type": "type.googleapis.com/google.rpc.QuotaFailure",600 "violations": [601 {602 "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_input_token_count",603 "quotaId": "GenerateContentInputTokensPerModelPerDay-FreeTier",604 "quotaDimensions": {605 "location": "global",606 "model": "gemini-3.1-pro"607 }608 },609 {610 "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_requests",611 "quotaId": "GenerateRequestsPerDayPerProjectPerModel-FreeTier"612 },613 {614 "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_requests",615 "quotaId": "GenerateRequestsPerMinutePerProjectPerModel-FreeTier"616 }617 ]618 }619 ],620 "quota_metrics_observed": [621 "generativelanguage.googleapis.com/generate_content_free_tier_input_token_count",622 "generativelanguage.googleapis.com/generate_content_free_tier_requests"623 ],624 "quota_ids_observed": [625 "GenerateContentInputTokensPerModelPerDay-FreeTier",626 "GenerateRequestsPerDayPerProjectPerModel-FreeTier",627 "GenerateRequestsPerMinutePerProjectPerModel-FreeTier"628 ],629 "retry_hint": "The retry delay is only in the message text ('Please retry in 54.22098241s.'); no Retry-After header was returned.",630 "retryable": "yes with exponential backoff + jitter (docs); Python SDK retries 429/5xx up to 4 attempts, initial ~1 s, max 60 s"631 },632 "headers": {633 "documented_rate_limit_headers": "none - the Gemini docs do not document any x-ratelimit-* response headers",634 "observed_on_generateContent_2026_09_19": [635 "Server-Timing: gfet4t7; dur=<ms>",636 "X-Gemini-Service-Tier: standard (also present on 429 responses)",637 "Alt-Svc",638 "Vary",639 "X-Content-Type-Options",640 "X-Frame-Options",641 "X-XSS-Protection",642 "Accept-Ranges",643 "Transfer-Encoding",644 "Content-Type: application/json; charset=UTF-8",645 "Date",646 "Server: scaffolding on HTTPServer2"647 ],648 "observed_on_openai_compat": [649 "Server-Timing",650 "Set-Cookie",651 "Cache-Control",652 "Expires",653 "P3P",654 "Vary",655 "X-Content-Type-Options",656 "X-Frame-Options",657 "X-XSS-Protection"658 ],659 "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)."660 },661 "observed_for_this_key": {662 "note": "Account-specific observations from 2026-09-19 UTC probes; not documentation.",663 "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).",664 "successful_calls": [665 "gemini-3.5-flash-lite",666 "gemini-3.5-flash",667 "gemini-3.8-flash",668 "gemma-4-26b-a4b-it",669 "gemini-flash-latest (-> gemini-3.8-flash)",670 "OpenAI-compat chat.completions gemini-3.5-flash-lite"671 ],672 "restricted_calls": [673 "gemini-2.5-flash-lite -> 404 NOT_FOUND 'no longer available to new users'",674 "gemini-3.1-pro-preview / gemini-pro-latest -> 429 RESOURCE_EXHAUSTED free-tier limit 0"675 ],676 "usage_metadata_shape": {677 "promptTokenCount": 5,678 "candidatesTokenCount": 2,679 "totalTokenCount": 7,680 "promptTokensDetails": [681 {682 "modality": "TEXT",683 "tokenCount": 5684 }685 ],686 "thoughtsTokenCount": "present when the model thought (5 on 3.5/3.8 Flash and Gemma 4 with maxOutputTokens=8 -> finishReason MAX_TOKENS, empty text)",687 "serviceTier": "standard"688 }689 },690 "_fragment": "generated/fragments/rate-limits/gemini-rate-limits.json"691 },692 {693 "provider": "openai",694 "retrieved_at": "2026-09-18",695 "sources": [696 "https://developers.openai.com/api/docs/guides/rate-limits",697 "https://developers.openai.com/api/docs/models/<id> (Rate limits section)",698 "https://developers.openai.com/api/docs/guides/batch",699 "https://developers.openai.com/api/docs/guides/fast-mode"700 ],701 "concepts": [702 {703 "metric": "RPM",704 "meaning": "requests per minute",705 "scope": "organization and project, per model (or shared-limit group)"706 },707 {708 "metric": "RPD",709 "meaning": "requests per day",710 "scope": "some models / free tier"711 },712 {713 "metric": "TPM",714 "meaning": "tokens per minute (max(max_tokens, estimated prompt tokens) counted per request)",715 "scope": "per model"716 },717 {718 "metric": "TPD",719 "meaning": "tokens per day",720 "scope": "some models"721 },722 {723 "metric": "IPM",724 "meaning": "images per minute",725 "scope": "image models (gpt-image-*)"726 },727 {728 "metric": "minutes-of-audio per minute",729 "meaning": "audio duration admitted per minute",730 "scope": "streaming audio models (gpt-realtime-whisper, gpt-realtime-translate)"731 },732 {733 "metric": "concurrent sessions",734 "meaning": "simultaneous live sessions",735 "scope": "gpt-live-1 (v1/live/sessions)"736 },737 {738 "metric": "batch queue limit",739 "meaning": "total input tokens queued across pending batch jobs for a model; released when the batch completes",740 "scope": "Batch API, per model"741 },742 {743 "metric": "long-context limit",744 "meaning": "separate RPM/TPM/batch-queue table for requests >272K input tokens (GPT-5.4/5.5/5.6/6 Astra)",745 "scope": "long-context models; visible in the console"746 },747 {748 "metric": "shared limits",749 "meaning": "some model families share one limit pool (listed under 'shared limit' in the console)",750 "scope": "organization"751 },752 {753 "metric": "project-scoped token limit",754 "meaning": "optional per-project token limit exposed via x-ratelimit-*-project-tokens headers",755 "scope": "project"756 },757 {758 "metric": "monthly usage limit",759 "meaning": "approved monthly spend ceiling per organization, separate from configurable spend limits",760 "scope": "organization"761 },762 {763 "metric": "vector store ingestion",764 "meaning": "/vector_stores/{id}/files and /file_batches share 300 requests per minute per vector store",765 "scope": "per vector store"766 },767 {768 "metric": "ramp-rate (slow_down)",769 "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",770 "scope": "per model"771 }772 ],773 "usage_tiers": [774 {775 "tier": "Free",776 "qualification": "allowed geography",777 "usage_limit_usd_per_month": 100778 },779 {780 "tier": "Tier 1",781 "qualification": "$5 paid",782 "usage_limit_usd_per_month": 100783 },784 {785 "tier": "Tier 2",786 "qualification": "$50 paid",787 "usage_limit_usd_per_month": 500788 },789 {790 "tier": "Tier 3",791 "qualification": "$100 paid",792 "usage_limit_usd_per_month": 1000793 },794 {795 "tier": "Tier 4",796 "qualification": "$250 paid",797 "usage_limit_usd_per_month": 5000798 },799 {800 "tier": "Tier 5",801 "qualification": "$1,000 paid",802 "usage_limit_usd_per_month": 200000803 }804 ],805 "headers": [806 {807 "name": "Retry-After",808 "sample": "56",809 "description": "minimum seconds to wait before retrying a temporary 429 (slow_down / rate limit) or 503 (server_is_overloaded)"810 },811 {812 "name": "x-ratelimit-limit-requests",813 "sample": "60",814 "description": "max requests before exhausting the rate limit"815 },816 {817 "name": "x-ratelimit-limit-tokens",818 "sample": "150000",819 "description": "max tokens before exhausting the rate limit"820 },821 {822 "name": "x-ratelimit-remaining-requests",823 "sample": "59",824 "description": "remaining requests"825 },826 {827 "name": "x-ratelimit-remaining-tokens",828 "sample": "149984",829 "description": "remaining tokens"830 },831 {832 "name": "x-ratelimit-reset-requests",833 "sample": "1s",834 "description": "time until request limit resets"835 },836 {837 "name": "x-ratelimit-reset-tokens",838 "sample": "6m0s",839 "description": "time until token limit resets"840 },841 {842 "name": "x-ratelimit-limit-project-tokens",843 "sample": "60000",844 "description": "project token limit (present when a project-scoped limit applies)"845 },846 {847 "name": "x-ratelimit-remaining-project-tokens",848 "sample": "57000",849 "description": "remaining project tokens"850 },851 {852 "name": "x-ratelimit-reset-project-tokens",853 "sample": "3s",854 "description": "time until project token limit resets"855 }856 ],857 "errors": [858 {859 "http_status": 429,860 "type": "rate_limit_error",861 "code": "slow_down",862 "meaning": "request rate increased too quickly",863 "action": "honour Retry-After, reduce rate, ramp gradually"864 },865 {866 "http_status": 429,867 "type": "rate_limit_error",868 "code": "rate_limit_exceeded",869 "meaning": "RPM/TPM/RPD/TPD/IPM exhausted",870 "action": "exponential backoff with jitter; batch requests; reduce max_tokens"871 },872 {873 "http_status": 429,874 "type": "insufficient_quota",875 "code": "insufficient_quota",876 "meaning": "monthly usage / spend limit reached (429 also returned when a hard spend limit is hit)",877 "action": "do not retry; raise limits"878 },879 {880 "http_status": 503,881 "type": "service_unavailable_error",882 "code": "server_is_overloaded",883 "meaning": "model temporarily overloaded",884 "action": "honour Retry-After, retry with increasing delay"885 }886 ],887 "fine_tuning_limits_endpoint": "GET /v1/fine_tuning/model_limits",888 "capacity_products": [889 "Scale Tier (pay-as-you-go traffic routinely hitting ramp limits)",890 "Reserved Tier (GPT-5.6 and later)",891 "Ultrafast mode (limited preview, GPT-5.6 Sol, announced 2026-08-13)"892 ],893 "per_model_documented_tiers": {894 "babbage-002": {895 "default": {896 "metrics": [897 "RPM",898 "RPD",899 "TPM",900 "Batch queue limit"901 ],902 "note": null,903 "tiers": {904 "Tier 1": {905 "RPM": 500,906 "RPD": 10000,907 "TPM": 10000,908 "Batch queue limit": 100000909 },910 "Tier 2": {911 "RPM": 5000,912 "RPD": null,913 "TPM": 40000,914 "Batch queue limit": 200000915 },916 "Tier 3": {917 "RPM": 5000,918 "RPD": null,919 "TPM": 80000,920 "Batch queue limit": 5000000921 },922 "Tier 4": {923 "RPM": 10000,924 "RPD": null,925 "TPM": 300000,926 "Batch queue limit": 30000000927 },928 "Tier 5": {929 "RPM": 10000,930 "RPD": null,931 "TPM": 1000000,932 "Batch queue limit": 150000000933 }934 }935 }936 },937 "chat-latest": {938 "default": {939 "metrics": [940 "RPM",941 "TPM",942 "Batch queue limit"943 ],944 "note": null,945 "tiers": {946 "Tier 1": {947 "RPM": 500,948 "TPM": 30000,949 "Batch queue limit": 50000950 },951 "Tier 2": {952 "RPM": 5000,953 "TPM": 450000,954 "Batch queue limit": 1350000955 },956 "Tier 3": {957 "RPM": 5000,958 "TPM": 800000,959 "Batch queue limit": 100000000960 },961 "Tier 4": {962 "RPM": 10000,963 "TPM": 2000000,964 "Batch queue limit": 200000000965 },966 "Tier 5": {967 "RPM": 15000,968 "TPM": 40000000,969 "Batch queue limit": 15000000000970 }971 }972 }973 },974 "chatgpt-4o-latest": {975 "default": {976 "metrics": [977 "RPM",978 "TPM",979 "Batch queue limit"980 ],981 "note": null,982 "tiers": {983 "Tier 1": {984 "RPM": 500,985 "TPM": 30000,986 "Batch queue limit": 90000987 },988 "Tier 2": {989 "RPM": 5000,990 "TPM": 450000,991 "Batch queue limit": 1350000992 },993 "Tier 3": {994 "RPM": 5000,995 "TPM": 800000,996 "Batch queue limit": 50000000997 },998 "Tier 4": {999 "RPM": 10000,1000 "TPM": 2000000,1001 "Batch queue limit": 2000000001002 },1003 "Tier 5": {1004 "RPM": 10000,1005 "TPM": 30000000,1006 "Batch queue limit": 50000000001007 }1008 }1009 }1010 },1011 "chatgpt-image-latest": {1012 "default": {1013 "metrics": [1014 "TPM",1015 "IPM"1016 ],1017 "note": null,1018 "tiers": {1019 "Tier 1": {1020 "TPM": 100000,1021 "IPM": 51022 },1023 "Tier 2": {1024 "TPM": 250000,1025 "IPM": 201026 },1027 "Tier 3": {1028 "TPM": 800000,1029 "IPM": 501030 },1031 "Tier 4": {1032 "TPM": 3000000,1033 "IPM": 1501034 },1035 "Tier 5": {1036 "TPM": 8000000,1037 "IPM": 2501038 }1039 }1040 }1041 },1042 "codex-mini-latest": {1043 "default": {1044 "metrics": [1045 "RPM",1046 "TPM",1047 "Batch queue limit"1048 ],1049 "note": null,1050 "tiers": {1051 "Tier 1": {1052 "RPM": 1000,1053 "TPM": 100000,1054 "Batch queue limit": 10000001055 },1056 "Tier 2": {1057 "RPM": 2000,1058 "TPM": 200000,1059 "Batch queue limit": 20000001060 },1061 "Tier 3": {1062 "RPM": 5000,1063 "TPM": 4000000,1064 "Batch queue limit": 400000001065 },1066 "Tier 4": {1067 "RPM": 10000,1068 "TPM": 10000000,1069 "Batch queue limit": 10000000001070 },1071 "Tier 5": {1072 "RPM": 30000,1073 "TPM": 150000000,1074 "Batch queue limit": 150000000001075 }1076 }1077 }1078 },1079 "computer-use-preview": {1080 "default": {1081 "metrics": [1082 "RPM",1083 "TPM",1084 "Batch queue limit"1085 ],1086 "note": null,1087 "tiers": {1088 "Tier 3": {1089 "RPM": 3000,1090 "TPM": 20000000,1091 "Batch queue limit": 4500000001092 },1093 "Tier 4": {1094 "RPM": 3000,1095 "TPM": 20000000,1096 "Batch queue limit": 4500000001097 },1098 "Tier 5": {1099 "RPM": 3000,1100 "TPM": 20000000,1101 "Batch queue limit": 4500000001102 }1103 }1104 }1105 },1106 "dall-e-3": {1107 "default": {1108 "metrics": [1109 "RPM"1110 ],1111 "note": null,1112 "tiers": {1113 "Tier free": {1114 "RPM": "1 img/min"1115 },1116 "Tier 1": {1117 "RPM": "500 img/min"1118 },1119 "Tier 2": {1120 "RPM": "2500 img/min"1121 },1122 "Tier 3": {1123 "RPM": "5000 img/min"1124 },1125 "Tier 4": {1126 "RPM": "7500 img/min"1127 },1128 "Tier 5": {1129 "RPM": "10000 img/min"1130 }1131 }1132 }1133 },1134 "davinci-002": {1135 "default": {1136 "metrics": [1137 "RPM",1138 "RPD",1139 "TPM",1140 "Batch queue limit"1141 ],1142 "note": null,1143 "tiers": {1144 "Tier 1": {1145 "RPM": 500,1146 "RPD": 10000,1147 "TPM": 10000,1148 "Batch queue limit": 1000001149 },1150 "Tier 2": {1151 "RPM": 5000,1152 "RPD": null,1153 "TPM": 40000,1154 "Batch queue limit": 2000001155 },1156 "Tier 3": {1157 "RPM": 5000,1158 "RPD": null,1159 "TPM": 80000,1160 "Batch queue limit": 50000001161 },1162 "Tier 4": {1163 "RPM": 10000,1164 "RPD": null,1165 "TPM": 300000,1166 "Batch queue limit": 300000001167 },1168 "Tier 5": {1169 "RPM": 10000,1170 "RPD": null,1171 "TPM": 1000000,1172 "Batch queue limit": 1500000001173 }1174 }1175 }1176 },1177 "gpt-3.5-turbo": {1178 "default": {1179 "metrics": [1180 "RPM",1181 "RPD",1182 "TPM",1183 "Batch queue limit"1184 ],1185 "note": null,1186 "tiers": {1187 "Tier 1": {1188 "RPM": 500,1189 "RPD": 10000,1190 "TPM": 200000,1191 "Batch queue limit": 20000001192 },1193 "Tier 2": {1194 "RPM": 5000,1195 "RPD": null,1196 "TPM": 2000000,1197 "Batch queue limit": 50000001198 },1199 "Tier 3": {1200 "RPM": 5000,1201 "RPD": null,1202 "TPM": 4000000,1203 "Batch queue limit": 500000001204 },1205 "Tier 4": {1206 "RPM": 10000,1207 "RPD": null,1208 "TPM": 10000000,1209 "Batch queue limit": 10000000001210 },1211 "Tier 5": {1212 "RPM": 10000,1213 "RPD": null,1214 "TPM": 50000000,1215 "Batch queue limit": 100000000001216 }1217 }1218 }1219 },1220 "gpt-4": {1221 "default": {1222 "metrics": [1223 "RPM",1224 "RPD",1225 "TPM",1226 "Batch queue limit"1227 ],1228 "note": null,1229 "tiers": {1230 "Tier 1": {1231 "RPM": 500,1232 "RPD": 10000,1233 "TPM": 10000,1234 "Batch queue limit": 1000001235 },1236 "Tier 2": {1237 "RPM": 5000,1238 "RPD": null,1239 "TPM": 40000,1240 "Batch queue limit": 2000001241 },1242 "Tier 3": {1243 "RPM": 5000,1244 "RPD": null,1245 "TPM": 80000,1246 "Batch queue limit": 50000001247 },1248 "Tier 4": {1249 "RPM": 10000,1250 "RPD": null,1251 "TPM": 300000,1252 "Batch queue limit": 300000001253 },1254 "Tier 5": {1255 "RPM": 10000,1256 "RPD": null,1257 "TPM": 1000000,1258 "Batch queue limit": 1500000001259 }1260 }1261 }1262 },1263 "gpt-4-turbo": {1264 "default": {1265 "metrics": [1266 "RPM",1267 "TPM",1268 "Batch queue limit"1269 ],1270 "note": null,1271 "tiers": {1272 "Tier 1": {1273 "RPM": 500,1274 "TPM": 30000,1275 "Batch queue limit": 900001276 },1277 "Tier 2": {1278 "RPM": 5000,1279 "TPM": 450000,1280 "Batch queue limit": 13500001281 },1282 "Tier 3": {1283 "RPM": 5000,1284 "TPM": 600000,1285 "Batch queue limit": 400000001286 },1287 "Tier 4": {1288 "RPM": 10000,1289 "TPM": 800000,1290 "Batch queue limit": 800000001291 },1292 "Tier 5": {1293 "RPM": 10000,1294 "TPM": 2000000,1295 "Batch queue limit": 3000000001296 }1297 }1298 }1299 },1300 "gpt-4-turbo-preview": {1301 "default": {1302 "metrics": [1303 "RPM",1304 "TPM",1305 "Batch queue limit"1306 ],1307 "note": null,1308 "tiers": {1309 "Tier 1": {1310 "RPM": 500,1311 "TPM": 30000,1312 "Batch queue limit": 900001313 },1314 "Tier 2": {1315 "RPM": 5000,1316 "TPM": 450000,1317 "Batch queue limit": 13500001318 },1319 "Tier 3": {1320 "RPM": 5000,1321 "TPM": 600000,1322 "Batch queue limit": 400000001323 },1324 "Tier 4": {1325 "RPM": 10000,1326 "TPM": 800000,1327 "Batch queue limit": 800000001328 },1329 "Tier 5": {1330 "RPM": 10000,1331 "TPM": 2000000,1332 "Batch queue limit": 3000000001333 }1334 }1335 }1336 },1337 "gpt-4.1": {1338 "default": {1339 "metrics": [1340 "RPM",1341 "TPM",1342 "Batch queue limit"1343 ],1344 "note": null,1345 "tiers": {1346 "Tier 1": {1347 "RPM": 500,1348 "TPM": 30000,1349 "Batch queue limit": 900001350 },1351 "Tier 2": {1352 "RPM": 5000,1353 "TPM": 450000,1354 "Batch queue limit": 13500001355 },1356 "Tier 3": {1357 "RPM": 5000,1358 "TPM": 800000,1359 "Batch queue limit": 500000001360 },1361 "Tier 4": {1362 "RPM": 10000,1363 "TPM": 2000000,1364 "Batch queue limit": 2000000001365 },1366 "Tier 5": {1367 "RPM": 10000,1368 "TPM": 30000000,1369 "Batch queue limit": 50000000001370 }1371 }1372 },1373 "Long Context": {1374 "metrics": [1375 "RPM",1376 "TPM",1377 "Batch queue limit"1378 ],1379 "note": "128k input tokens",1380 "tiers": {1381 "Tier 1": {1382 "RPM": 100,1383 "TPM": 200000,1384 "Batch queue limit": 20000001385 },1386 "Tier 2": {1387 "RPM": 250,1388 "TPM": 500000,1389 "Batch queue limit": 200000001390 },1391 "Tier 3": {1392 "RPM": 500,1393 "TPM": 1000000,1394 "Batch queue limit": 400000001395 },1396 "Tier 4": {1397 "RPM": 1000,1398 "TPM": 5000000,1399 "Batch queue limit": 1000000001400 },1401 "Tier 5": {1402 "RPM": 4000,1403 "TPM": 10000000,1404 "Batch queue limit": 10000000001405 }1406 }1407 }1408 },1409 "gpt-4.1-mini": {1410 "Standard": {1411 "metrics": [1412 "RPM",1413 "RPD",1414 "TPM",1415 "Batch queue limit"1416 ],1417 "note": null,1418 "tiers": {1419 "Tier 1": {1420 "RPM": 500,1421 "RPD": 10000,1422 "TPM": 200000,1423 "Batch queue limit": 20000001424 },1425 "Tier 2": {1426 "RPM": 5000,1427 "RPD": null,1428 "TPM": 2000000,1429 "Batch queue limit": 200000001430 },1431 "Tier 3": {1432 "RPM": 5000,1433 "RPD": null,1434 "TPM": 4000000,1435 "Batch queue limit": 400000001436 },1437 "Tier 4": {1438 "RPM": 10000,1439 "RPD": null,1440 "TPM": 10000000,1441 "Batch queue limit": 10000000001442 },1443 "Tier 5": {1444 "RPM": 30000,1445 "RPD": null,1446 "TPM": 150000000,1447 "Batch queue limit": 150000000001448 }1449 }1450 },1451 "Long Context": {1452 "metrics": [1453 "RPM",1454 "TPM",1455 "Batch queue limit"1456 ],1457 "note": "128k input tokens",1458 "tiers": {1459 "Tier 1": {1460 "RPM": 200,1461 "TPM": 400000,1462 "Batch queue limit": 50000001463 },1464 "Tier 2": {1465 "RPM": 500,1466 "TPM": 1000000,1467 "Batch queue limit": 400000001468 },1469 "Tier 3": {1470 "RPM": 1000,1471 "TPM": 2000000,1472 "Batch queue limit": 800000001473 },1474 "Tier 4": {1475 "RPM": 2000,1476 "TPM": 10000000,1477 "Batch queue limit": 2000000001478 },1479 "Tier 5": {1480 "RPM": 8000,1481 "TPM": 20000000,1482 "Batch queue limit": 20000000001483 }1484 }1485 }1486 },1487 "gpt-4.1-nano": {1488 "Standard": {1489 "metrics": [1490 "RPM",1491 "RPD",1492 "TPM",1493 "Batch queue limit"1494 ],1495 "note": null,1496 "tiers": {1497 "Tier 1": {1498 "RPM": 500,1499 "RPD": 10000,1500 "TPM": 200000,1501 "Batch queue limit": 20000001502 },1503 "Tier 2": {1504 "RPM": 5000,1505 "RPD": null,1506 "TPM": 2000000,1507 "Batch queue limit": 200000001508 },1509 "Tier 3": {1510 "RPM": 5000,1511 "RPD": null,1512 "TPM": 4000000,1513 "Batch queue limit": 400000001514 },1515 "Tier 4": {1516 "RPM": 10000,1517 "RPD": null,1518 "TPM": 10000000,1519 "Batch queue limit": 10000000001520 },1521 "Tier 5": {1522 "RPM": 30000,1523 "RPD": null,1524 "TPM": 150000000,1525 "Batch queue limit": 150000000001526 }1527 }1528 },1529 "Long Context": {1530 "metrics": [1531 "RPM",1532 "TPM",1533 "Batch queue limit"1534 ],1535 "note": "128k input tokens",1536 "tiers": {1537 "Tier 1": {1538 "RPM": 200,1539 "TPM": 400000,1540 "Batch queue limit": 50000001541 },1542 "Tier 2": {1543 "RPM": 500,1544 "TPM": 1000000,1545 "Batch queue limit": 400000001546 },1547 "Tier 3": {1548 "RPM": 1000,1549 "TPM": 2000000,1550 "Batch queue limit": 800000001551 },1552 "Tier 4": {1553 "RPM": 2000,1554 "TPM": 10000000,1555 "Batch queue limit": 2000000001556 },1557 "Tier 5": {1558 "RPM": 8000,1559 "TPM": 20000000,1560 "Batch queue limit": 20000000001561 }1562 }1563 }1564 },1565 "gpt-4.5-preview": {1566 "default": {1567 "metrics": [1568 "RPM",1569 "TPM",1570 "Batch queue limit"1571 ],1572 "note": null,1573 "tiers": {1574 "Tier 1": {1575 "RPM": 1000,1576 "TPM": 125000,1577 "Batch queue limit": 500001578 },1579 "Tier 2": {1580 "RPM": 5000,1581 "TPM": 250000,1582 "Batch queue limit": 5000001583 },1584 "Tier 3": {1585 "RPM": 5000,1586 "TPM": 500000,1587 "Batch queue limit": 500000001588 },1589 "Tier 4": {1590 "RPM": 10000,1591 "TPM": 1000000,1592 "Batch queue limit": 1000000001593 },1594 "Tier 5": {1595 "RPM": 10000,1596 "TPM": 2000000,1597 "Batch queue limit": 50000000001598 }1599 }1600 }1601 },1602 "gpt-4o": {1603 "default": {1604 "metrics": [1605 "RPM",1606 "TPM",1607 "Batch queue limit"1608 ],1609 "note": null,1610 "tiers": {1611 "Tier 1": {1612 "RPM": 500,1613 "TPM": 30000,1614 "Batch queue limit": 900001615 },1616 "Tier 2": {1617 "RPM": 5000,1618 "TPM": 450000,1619 "Batch queue limit": 13500001620 },1621 "Tier 3": {1622 "RPM": 5000,1623 "TPM": 800000,1624 "Batch queue limit": 500000001625 },1626 "Tier 4": {1627 "RPM": 10000,1628 "TPM": 2000000,1629 "Batch queue limit": 2000000001630 },1631 "Tier 5": {1632 "RPM": 10000,1633 "TPM": 30000000,1634 "Batch queue limit": 50000000001635 }1636 }1637 }1638 },1639 "gpt-4o-audio-preview": {1640 "default": {1641 "metrics": [1642 "RPM",1643 "TPM",1644 "Batch queue limit"1645 ],1646 "note": null,1647 "tiers": {1648 "Tier 1": {1649 "RPM": 500,1650 "TPM": 30000,1651 "Batch queue limit": 900001652 },1653 "Tier 2": {1654 "RPM": 5000,1655 "TPM": 450000,1656 "Batch queue limit": 13500001657 },1658 "Tier 3": {1659 "RPM": 5000,1660 "TPM": 800000,1661 "Batch queue limit": 500000001662 },1663 "Tier 4": {1664 "RPM": 10000,1665 "TPM": 2000000,1666 "Batch queue limit": 20000001667 },1668 "Tier 5": {1669 "RPM": 10000,1670 "TPM": 30000000,1671 "Batch queue limit": 50000000001672 }1673 }1674 }1675 },1676 "gpt-4o-mini": {1677 "default": {1678 "metrics": [1679 "RPM",1680 "RPD",1681 "TPM",1682 "Batch queue limit"1683 ],1684 "note": null,1685 "tiers": {1686 "Tier 1": {1687 "RPM": 500,1688 "RPD": 10000,1689 "TPM": 200000,1690 "Batch queue limit": 20000001691 },1692 "Tier 2": {1693 "RPM": 5000,1694 "RPD": null,1695 "TPM": 2000000,1696 "Batch queue limit": 200000001697 },1698 "Tier 3": {1699 "RPM": 5000,1700 "RPD": null,1701 "TPM": 4000000,1702 "Batch queue limit": 400000001703 },1704 "Tier 4": {1705 "RPM": 10000,1706 "RPD": null,1707 "TPM": 10000000,1708 "Batch queue limit": 10000000001709 },1710 "Tier 5": {1711 "RPM": 30000,1712 "RPD": null,1713 "TPM": 150000000,1714 "Batch queue limit": 150000000001715 }1716 }1717 }1718 },1719 "gpt-4o-mini-audio-preview": {1720 "default": {1721 "metrics": [1722 "RPM",1723 "RPD",1724 "TPM",1725 "Batch queue limit"1726 ],1727 "note": null,1728 "tiers": {1729 "free": {1730 "RPM": 3,1731 "RPD": 200,1732 "TPM": 40000,1733 "Batch queue limit": null1734 },1735 "Tier 1": {1736 "RPM": 500,1737 "RPD": 10000,1738 "TPM": 200000,1739 "Batch queue limit": 20000001740 },1741 "Tier 2": {1742 "RPM": 5000,1743 "RPD": null,1744 "TPM": 2000000,1745 "Batch queue limit": 200000001746 },1747 "Tier 3": {1748 "RPM": 5000,1749 "RPD": null,1750 "TPM": 4000000,1751 "Batch queue limit": 400000001752 },1753 "Tier 4": {1754 "RPM": 10000,1755 "RPD": null,1756 "TPM": 10000000,1757 "Batch queue limit": 10000000001758 },1759 "Tier 5": {1760 "RPM": 30000,1761 "RPD": null,1762 "TPM": 150000000,1763 "Batch queue limit": 150000000001764 }1765 }1766 }1767 },1768 "gpt-4o-mini-realtime-preview": {1769 "default": {1770 "metrics": [1771 "RPM",1772 "RPD",1773 "TPM"1774 ],1775 "note": null,1776 "tiers": {1777 "Tier 1": {1778 "RPM": 200,1779 "RPD": 1000,1780 "TPM": 400001781 },1782 "Tier 2": {1783 "RPM": 400,1784 "RPD": null,1785 "TPM": 2000001786 },1787 "Tier 3": {1788 "RPM": 5000,1789 "RPD": null,1790 "TPM": 8000001791 },1792 "Tier 4": {1793 "RPM": 10000,1794 "RPD": null,1795 "TPM": 40000001796 },1797 "Tier 5": {1798 "RPM": 20000,1799 "RPD": null,1800 "TPM": 150000001801 }1802 }1803 }1804 },1805 "gpt-4o-mini-search-preview": {1806 "default": {1807 "metrics": [1808 "RPM",1809 "RPD",1810 "TPM",1811 "Batch queue limit"1812 ],1813 "note": null,1814 "tiers": {1815 "free": {1816 "RPM": 3,1817 "RPD": 200,1818 "TPM": 40000,1819 "Batch queue limit": null1820 },1821 "Tier 1": {1822 "RPM": 500,1823 "RPD": 10000,1824 "TPM": 200000,1825 "Batch queue limit": 20000001826 },1827 "Tier 2": {1828 "RPM": 5000,1829 "RPD": null,1830 "TPM": 2000000,1831 "Batch queue limit": 200000001832 },1833 "Tier 3": {1834 "RPM": 5000,1835 "RPD": null,1836 "TPM": 4000000,1837 "Batch queue limit": 400000001838 },1839 "Tier 4": {1840 "RPM": 10000,1841 "RPD": null,1842 "TPM": 10000000,1843 "Batch queue limit": 10000000001844 },1845 "Tier 5": {1846 "RPM": 30000,1847 "RPD": null,1848 "TPM": 150000000,1849 "Batch queue limit": 150000000001850 }1851 }1852 }1853 },1854 "gpt-4o-mini-transcribe": {1855 "default": {1856 "metrics": [1857 "RPM",1858 "TPM"1859 ],1860 "note": null,1861 "tiers": {1862 "Tier 1": {1863 "RPM": 500,1864 "TPM": 500001865 },1866 "Tier 2": {1867 "RPM": 2000,1868 "TPM": 1500001869 },1870 "Tier 3": {1871 "RPM": 5000,1872 "TPM": 6000001873 },1874 "Tier 4": {1875 "RPM": 10000,1876 "TPM": 20000001877 },1878 "Tier 5": {1879 "RPM": 10000,1880 "TPM": 80000001881 }1882 }1883 }1884 },1885 "gpt-4o-mini-tts": {1886 "default": {1887 "metrics": [1888 "RPM",1889 "TPM"1890 ],1891 "note": null,1892 "tiers": {1893 "Tier 1": {1894 "RPM": 500,1895 "TPM": 500001896 },1897 "Tier 2": {1898 "RPM": 2000,1899 "TPM": 1500001900 },1901 "Tier 3": {1902 "RPM": 5000,1903 "TPM": 6000001904 },1905 "Tier 4": {1906 "RPM": 10000,1907 "TPM": 20000001908 },1909 "Tier 5": {1910 "RPM": 10000,1911 "TPM": 80000001912 }1913 }1914 }1915 },1916 "gpt-4o-realtime-preview": {1917 "default": {1918 "metrics": [1919 "RPM",1920 "RPD",1921 "TPM"1922 ],1923 "note": null,1924 "tiers": {1925 "Tier 1": {1926 "RPM": 200,1927 "RPD": 1000,1928 "TPM": 400001929 },1930 "Tier 2": {1931 "RPM": 400,1932 "RPD": null,1933 "TPM": 2000001934 },1935 "Tier 3": {1936 "RPM": 5000,1937 "RPD": null,1938 "TPM": 8000001939 },1940 "Tier 4": {1941 "RPM": 10000,1942 "RPD": null,1943 "TPM": 40000001944 },1945 "Tier 5": {1946 "RPM": 20000,1947 "RPD": null,1948 "TPM": 150000001949 }1950 }1951 }1952 },1953 "gpt-4o-search-preview": {1954 "default": {1955 "metrics": [1956 "RPM",1957 "TPM",1958 "Batch queue limit"1959 ],1960 "note": null,1961 "tiers": {1962 "Tier 1": {1963 "RPM": 100,1964 "TPM": 30000,1965 "Batch queue limit": 01966 },1967 "Tier 2": {1968 "RPM": 500,1969 "TPM": 45000,1970 "Batch queue limit": 01971 },1972 "Tier 3": {1973 "RPM": 500,1974 "TPM": 80000,1975 "Batch queue limit": 01976 },1977 "Tier 4": {1978 "RPM": 1000,1979 "TPM": 200000,1980 "Batch queue limit": 01981 },1982 "Tier 5": {1983 "RPM": 1000,1984 "TPM": 3000000,1985 "Batch queue limit": 01986 }1987 }1988 }1989 },1990 "gpt-4o-transcribe": {1991 "default": {1992 "metrics": [1993 "RPM",1994 "TPM"1995 ],1996 "note": null,1997 "tiers": {1998 "Tier 1": {1999 "RPM": 500,2000 "TPM": 100002001 },2002 "Tier 2": {2003 "RPM": 2000,2004 "TPM": 1000002005 },2006 "Tier 3": {2007 "RPM": 5000,2008 "TPM": 4000002009 },2010 "Tier 4": {2011 "RPM": 10000,2012 "TPM": 20000002013 },2014 "Tier 5": {2015 "RPM": 10000,2016 "TPM": 60000002017 }2018 }2019 }2020 },2021 "gpt-4o-transcribe-diarize": {2022 "default": {2023 "metrics": [2024 "RPM",2025 "TPM"2026 ],2027 "note": null,2028 "tiers": {2029 "Tier 1": {2030 "RPM": 500,2031 "TPM": 100002032 },2033 "Tier 2": {2034 "RPM": 5000,2035 "TPM": 1000002036 },2037 "Tier 3": {2038 "RPM": 5000,2039 "TPM": 4000002040 },2041 "Tier 4": {2042 "RPM": 10000,2043 "TPM": 20000002044 },2045 "Tier 5": {2046 "RPM": 10000,2047 "TPM": 60000002048 }2049 }2050 }2051 },2052 "gpt-5": {2053 "default": {2054 "metrics": [2055 "RPM",2056 "TPM",2057 "Batch queue limit"2058 ],2059 "note": null,2060 "tiers": {2061 "Tier 1": {2062 "RPM": 500,2063 "TPM": 500000,2064 "Batch queue limit": 15000002065 },2066 "Tier 2": {2067 "RPM": 5000,2068 "TPM": 1000000,2069 "Batch queue limit": 30000002070 },2071 "Tier 3": {2072 "RPM": 5000,2073 "TPM": 2000000,2074 "Batch queue limit": 1000000002075 },2076 "Tier 4": {2077 "RPM": 10000,2078 "TPM": 4000000,2079 "Batch queue limit": 2000000002080 },2081 "Tier 5": {2082 "RPM": 15000,2083 "TPM": 40000000,2084 "Batch queue limit": 150000000002085 }2086 }2087 }2088 },2089 "gpt-5-chat-latest": {2090 "default": {2091 "metrics": [2092 "RPM",2093 "TPM",2094 "Batch queue limit"2095 ],2096 "note": null,2097 "tiers": {2098 "Tier 1": {2099 "RPM": 500,2100 "TPM": 30000,2101 "Batch queue limit": 500002102 },2103 "Tier 2": {2104 "RPM": 5000,2105 "TPM": 450000,2106 "Batch queue limit": 13500002107 },2108 "Tier 3": {2109 "RPM": 5000,2110 "TPM": 800000,2111 "Batch queue limit": 1000000002112 },2113 "Tier 4": {2114 "RPM": 10000,2115 "TPM": 2000000,2116 "Batch queue limit": 2000000002117 },2118 "Tier 5": {2119 "RPM": 15000,2120 "TPM": 40000000,2121 "Batch queue limit": 150000000002122 }2123 }2124 }2125 },2126 "gpt-5-codex": {2127 "default": {2128 "metrics": [2129 "RPM",2130 "TPM",2131 "Batch queue limit"2132 ],2133 "note": null,2134 "tiers": {2135 "Tier 1": {2136 "RPM": 500,2137 "TPM": 500000,2138 "Batch queue limit": 15000002139 },2140 "Tier 2": {2141 "RPM": 5000,2142 "TPM": 1000000,2143 "Batch queue limit": 30000002144 },2145 "Tier 3": {2146 "RPM": 5000,2147 "TPM": 2000000,2148 "Batch queue limit": 1000000002149 },2150 "Tier 4": {2151 "RPM": 10000,2152 "TPM": 4000000,2153 "Batch queue limit": 2000000002154 },2155 "Tier 5": {2156 "RPM": 15000,2157 "TPM": 10000000,2158 "Batch queue limit": 150000000002159 }2160 }2161 }2162 },2163 "gpt-5-mini": {2164 "default": {2165 "metrics": [2166 "RPM",2167 "TPM",2168 "Batch queue limit"2169 ],2170 "note": null,2171 "tiers": {2172 "Tier 1": {2173 "RPM": 500,2174 "TPM": 500000,2175 "Batch queue limit": 50000002176 },2177 "Tier 2": {2178 "RPM": 5000,2179 "TPM": 2000000,2180 "Batch queue limit": 200000002181 },2182 "Tier 3": {2183 "RPM": 5000,2184 "TPM": 4000000,2185 "Batch queue limit": 400000002186 },2187 "Tier 4": {2188 "RPM": 10000,2189 "TPM": 10000000,2190 "Batch queue limit": 10000000002191 },2192 "Tier 5": {2193 "RPM": 30000,2194 "TPM": 180000000,2195 "Batch queue limit": 150000000002196 }2197 }2198 }2199 },2200 "gpt-5-nano": {2201 "default": {2202 "metrics": [2203 "RPM",2204 "TPM",2205 "Batch queue limit"2206 ],2207 "note": null,2208 "tiers": {2209 "Tier 1": {2210 "RPM": 500,2211 "TPM": 200000,2212 "Batch queue limit": 20000002213 },2214 "Tier 2": {2215 "RPM": 5000,2216 "TPM": 2000000,2217 "Batch queue limit": 200000002218 },2219 "Tier 3": {2220 "RPM": 5000,2221 "TPM": 4000000,2222 "Batch queue limit": 400000002223 },2224 "Tier 4": {2225 "RPM": 10000,2226 "TPM": 10000000,2227 "Batch queue limit": 10000000002228 },2229 "Tier 5": {2230 "RPM": 30000,2231 "TPM": 180000000,2232 "Batch queue limit": 150000000002233 }2234 }2235 }2236 },2237 "gpt-5-pro": {2238 "default": {2239 "metrics": [2240 "RPM",2241 "TPM",2242 "Batch queue limit"2243 ],2244 "note": null,2245 "tiers": {2246 "Tier 1": {2247 "RPM": 500,2248 "TPM": 30000,2249 "Batch queue limit": 900002250 },2251 "Tier 2": {2252 "RPM": 5000,2253 "TPM": 450000,2254 "Batch queue limit": 13500002255 },2256 "Tier 3": {2257 "RPM": 5000,2258 "TPM": 800000,2259 "Batch queue limit": 500000002260 },2261 "Tier 4": {2262 "RPM": 10000,2263 "TPM": 2000000,2264 "Batch queue limit": 2000000002265 },2266 "Tier 5": {2267 "RPM": 10000,2268 "TPM": 30000000,2269 "Batch queue limit": 50000000002270 }2271 }2272 }2273 },2274 "gpt-5.1": {2275 "default": {2276 "metrics": [2277 "RPM",2278 "TPM",2279 "Batch queue limit"2280 ],2281 "note": null,2282 "tiers": {2283 "Tier 1": {2284 "RPM": 500,2285 "TPM": 500000,2286 "Batch queue limit": 15000002287 },2288 "Tier 2": {2289 "RPM": 5000,2290 "TPM": 1000000,2291 "Batch queue limit": 30000002292 },2293 "Tier 3": {2294 "RPM": 5000,2295 "TPM": 2000000,2296 "Batch queue limit": 1000000002297 },2298 "Tier 4": {2299 "RPM": 10000,2300 "TPM": 4000000,2301 "Batch queue limit": 2000000002302 },2303 "Tier 5": {2304 "RPM": 15000,2305 "TPM": 40000000,2306 "Batch queue limit": 150000000002307 }2308 }2309 }2310 },2311 "gpt-5.1-chat-latest": {2312 "default": {2313 "metrics": [2314 "RPM",2315 "TPM",2316 "Batch queue limit"2317 ],2318 "note": null,2319 "tiers": {2320 "Tier 1": {2321 "RPM": 500,2322 "TPM": 30000,2323 "Batch queue limit": 500002324 },2325 "Tier 2": {2326 "RPM": 5000,2327 "TPM": 450000,2328 "Batch queue limit": 13500002329 },2330 "Tier 3": {2331 "RPM": 5000,2332 "TPM": 800000,2333 "Batch queue limit": 1000000002334 },2335 "Tier 4": {2336 "RPM": 10000,2337 "TPM": 2000000,2338 "Batch queue limit": 2000000002339 },2340 "Tier 5": {2341 "RPM": 15000,2342 "TPM": 40000000,2343 "Batch queue limit": 150000000002344 }2345 }2346 }2347 },2348 "gpt-5.1-codex": {2349 "default": {2350 "metrics": [2351 "RPM",2352 "TPM",2353 "Batch queue limit"2354 ],2355 "note": null,2356 "tiers": {2357 "Tier 1": {2358 "RPM": 500,2359 "TPM": 500000,2360 "Batch queue limit": 15000002361 },2362 "Tier 2": {2363 "RPM": 5000,2364 "TPM": 1000000,2365 "Batch queue limit": 30000002366 },2367 "Tier 3": {2368 "RPM": 5000,2369 "TPM": 2000000,2370 "Batch queue limit": 1000000002371 },2372 "Tier 4": {2373 "RPM": 10000,2374 "TPM": 4000000,2375 "Batch queue limit": 2000000002376 },2377 "Tier 5": {2378 "RPM": 15000,2379 "TPM": 40000000,2380 "Batch queue limit": 150000000002381 }2382 }2383 }2384 },2385 "gpt-5.1-codex-max": {2386 "default": {2387 "metrics": [2388 "RPM",2389 "TPM",2390 "Batch queue limit"2391 ],2392 "note": null,2393 "tiers": {2394 "Tier 1": {2395 "RPM": 500,2396 "TPM": 500000,2397 "Batch queue limit": 15000002398 },2399 "Tier 2": {2400 "RPM": 5000,2401 "TPM": 1000000,2402 "Batch queue limit": 30000002403 },2404 "Tier 3": {2405 "RPM": 5000,2406 "TPM": 2000000,2407 "Batch queue limit": 1000000002408 },2409 "Tier 4": {2410 "RPM": 10000,2411 "TPM": 4000000,2412 "Batch queue limit": 2000000002413 },2414 "Tier 5": {2415 "RPM": 15000,2416 "TPM": 40000000,2417 "Batch queue limit": 150000000002418 }2419 }2420 }2421 },2422 "gpt-5.1-codex-mini": {2423 "default": {2424 "metrics": [2425 "RPM",2426 "TPM",2427 "Batch queue limit"2428 ],2429 "note": null,2430 "tiers": {2431 "Tier 1": {2432 "RPM": 500,2433 "TPM": 500000,2434 "Batch queue limit": 50000002435 },2436 "Tier 2": {2437 "RPM": 5000,2438 "TPM": 2000000,2439 "Batch queue limit": 200000002440 },2441 "Tier 3": {2442 "RPM": 5000,2443 "TPM": 4000000,2444 "Batch queue limit": 400000002445 },2446 "Tier 4": {2447 "RPM": 10000,2448 "TPM": 10000000,2449 "Batch queue limit": 10000000002450 },2451 "Tier 5": {2452 "RPM": 30000,2453 "TPM": 180000000,2454 "Batch queue limit": 150000000002455 }2456 }2457 }2458 },2459 "gpt-5.2": {2460 "default": {2461 "metrics": [2462 "RPM",2463 "TPM",2464 "Batch queue limit"2465 ],2466 "note": null,2467 "tiers": {2468 "Tier 1": {2469 "RPM": 500,2470 "TPM": 500000,2471 "Batch queue limit": 15000002472 },2473 "Tier 2": {2474 "RPM": 5000,2475 "TPM": 1000000,2476 "Batch queue limit": 30000002477 },2478 "Tier 3": {2479 "RPM": 5000,2480 "TPM": 2000000,2481 "Batch queue limit": 1000000002482 },2483 "Tier 4": {2484 "RPM": 10000,2485 "TPM": 4000000,2486 "Batch queue limit": 2000000002487 },2488 "Tier 5": {2489 "RPM": 15000,2490 "TPM": 40000000,2491 "Batch queue limit": 150000000002492 }2493 }2494 }2495 },2496 "gpt-5.2-chat-latest": {2497 "default": {2498 "metrics": [2499 "RPM",2500 "TPM",2501 "Batch queue limit"2502 ],2503 "note": null,2504 "tiers": {2505 "Tier 1": {2506 "RPM": 500,2507 "TPM": 30000,2508 "Batch queue limit": 500002509 },2510 "Tier 2": {2511 "RPM": 5000,2512 "TPM": 450000,2513 "Batch queue limit": 13500002514 },2515 "Tier 3": {2516 "RPM": 5000,2517 "TPM": 800000,2518 "Batch queue limit": 1000000002519 },2520 "Tier 4": {2521 "RPM": 10000,2522 "TPM": 2000000,2523 "Batch queue limit": 2000000002524 },2525 "Tier 5": {2526 "RPM": 15000,2527 "TPM": 40000000,2528 "Batch queue limit": 150000000002529 }2530 }2531 }2532 },2533 "gpt-5.2-codex": {2534 "default": {2535 "metrics": [2536 "RPM",2537 "TPM",2538 "Batch queue limit"2539 ],2540 "note": null,2541 "tiers": {2542 "Tier 1": {2543 "RPM": 500,2544 "TPM": 500000,2545 "Batch queue limit": 15000002546 },2547 "Tier 2": {2548 "RPM": 5000,2549 "TPM": 1000000,2550 "Batch queue limit": 30000002551 },2552 "Tier 3": {2553 "RPM": 5000,2554 "TPM": 2000000,2555 "Batch queue limit": 1000000002556 },2557 "Tier 4": {2558 "RPM": 10000,2559 "TPM": 4000000,2560 "Batch queue limit": 2000000002561 },2562 "Tier 5": {2563 "RPM": 15000,2564 "TPM": 40000000,2565 "Batch queue limit": 150000000002566 }2567 }2568 }2569 },2570 "gpt-5.2-pro": {2571 "Standard": {2572 "metrics": [2573 "RPM",2574 "TPM",2575 "Batch queue limit"2576 ],2577 "note": null,2578 "tiers": {2579 "Tier 1": {2580 "RPM": 500,2581 "TPM": 30000,2582 "Batch queue limit": 900002583 },2584 "Tier 2": {2585 "RPM": 5000,2586 "TPM": 450000,2587 "Batch queue limit": 13500002588 },2589 "Tier 3": {2590 "RPM": 5000,2591 "TPM": 800000,2592 "Batch queue limit": 500000002593 },2594 "Tier 4": {2595 "RPM": 10000,2596 "TPM": 2000000,2597 "Batch queue limit": 2000000002598 },2599 "Tier 5": {2600 "RPM": 10000,2601 "TPM": 30000000,2602 "Batch queue limit": 50000000002603 }2604 }2605 },2606 "Long Context": {2607 "metrics": [2608 "RPM",2609 "TPM",2610 "Batch queue limit"2611 ],2612 "note": "128k input tokens",2613 "tiers": {2614 "Tier 1": {2615 "RPM": 100,2616 "TPM": 200000,2617 "Batch queue limit": 20000002618 },2619 "Tier 2": {2620 "RPM": 250,2621 "TPM": 500000,2622 "Batch queue limit": 200000002623 },2624 "Tier 3": {2625 "RPM": 500,2626 "TPM": 1000000,2627 "Batch queue limit": 400000002628 },2629 "Tier 4": {2630 "RPM": 1000,2631 "TPM": 5000000,2632 "Batch queue limit": 1000000002633 },2634 "Tier 5": {2635 "RPM": 4000,2636 "TPM": 10000000,2637 "Batch queue limit": 10000000002638 }2639 }2640 }2641 },2642 "gpt-5.3-chat-latest": {2643 "default": {2644 "metrics": [2645 "RPM",2646 "TPM",2647 "Batch queue limit"2648 ],2649 "note": null,2650 "tiers": {2651 "Tier 1": {2652 "RPM": 500,2653 "TPM": 30000,2654 "Batch queue limit": 500002655 },2656 "Tier 2": {2657 "RPM": 5000,2658 "TPM": 450000,2659 "Batch queue limit": 13500002660 },2661 "Tier 3": {2662 "RPM": 5000,2663 "TPM": 800000,2664 "Batch queue limit": 1000000002665 },2666 "Tier 4": {2667 "RPM": 10000,2668 "TPM": 2000000,2669 "Batch queue limit": 2000000002670 },2671 "Tier 5": {2672 "RPM": 15000,2673 "TPM": 40000000,2674 "Batch queue limit": 150000000002675 }2676 }2677 }2678 },2679 "gpt-5.3-codex": {2680 "default": {2681 "metrics": [2682 "RPM",2683 "TPM",2684 "Batch queue limit"2685 ],2686 "note": null,2687 "tiers": {2688 "Tier 1": {2689 "RPM": 500,2690 "TPM": 500000,2691 "Batch queue limit": 15000002692 },2693 "Tier 2": {2694 "RPM": 5000,2695 "TPM": 1000000,2696 "Batch queue limit": 30000002697 },2698 "Tier 3": {2699 "RPM": 5000,2700 "TPM": 2000000,2701 "Batch queue limit": 1000000002702 },2703 "Tier 4": {2704 "RPM": 10000,2705 "TPM": 4000000,2706 "Batch queue limit": 2000000002707 },2708 "Tier 5": {2709 "RPM": 15000,2710 "TPM": 40000000,2711 "Batch queue limit": 150000000002712 }2713 }2714 }2715 },2716 "gpt-5.4": {2717 "Standard": {2718 "metrics": [2719 "RPM",2720 "TPM",2721 "Batch queue limit"2722 ],2723 "note": null,2724 "tiers": {2725 "Tier 1": {2726 "RPM": 500,2727 "TPM": 500000,2728 "Batch queue limit": 15000002729 },2730 "Tier 2": {2731 "RPM": 5000,2732 "TPM": 1000000,2733 "Batch queue limit": 30000002734 },2735 "Tier 3": {2736 "RPM": 5000,2737 "TPM": 2000000,2738 "Batch queue limit": 1000000002739 },2740 "Tier 4": {2741 "RPM": 10000,2742 "TPM": 4000000,2743 "Batch queue limit": 2000000002744 },2745 "Tier 5": {2746 "RPM": 15000,2747 "TPM": 40000000,2748 "Batch queue limit": 150000000002749 }2750 }2751 },2752 "Long Context": {2753 "metrics": [2754 "RPM",2755 "TPM",2756 "Batch queue limit"2757 ],2758 "note": "272K input tokens",2759 "tiers": {2760 "Tier 1": {2761 "RPM": 200,2762 "TPM": 400000,2763 "Batch queue limit": 50000002764 },2765 "Tier 2": {2766 "RPM": 500,2767 "TPM": 1000000,2768 "Batch queue limit": 400000002769 },2770 "Tier 3": {2771 "RPM": 1000,2772 "TPM": 2000000,2773 "Batch queue limit": 800000002774 },2775 "Tier 4": {2776 "RPM": 2000,2777 "TPM": 10000000,2778 "Batch queue limit": 2000000002779 },2780 "Tier 5": {2781 "RPM": 8000,2782 "TPM": 20000000,2783 "Batch queue limit": 20000000002784 }2785 }2786 }2787 },2788 "gpt-5.4-mini": {2789 "default": {2790 "metrics": [2791 "RPM",2792 "TPM",2793 "Batch queue limit"2794 ],2795 "note": null,2796 "tiers": {2797 "Tier 1": {2798 "RPM": 500,2799 "TPM": 500000,2800 "Batch queue limit": 50000002801 },2802 "Tier 2": {2803 "RPM": 5000,2804 "TPM": 2000000,2805 "Batch queue limit": 200000002806 },2807 "Tier 3": {2808 "RPM": 5000,2809 "TPM": 4000000,2810 "Batch queue limit": 400000002811 },2812 "Tier 4": {2813 "RPM": 10000,2814 "TPM": 10000000,2815 "Batch queue limit": 10000000002816 },2817 "Tier 5": {2818 "RPM": 30000,2819 "TPM": 180000000,2820 "Batch queue limit": 150000000002821 }2822 }2823 }2824 },2825 "gpt-5.4-nano": {2826 "default": {2827 "metrics": [2828 "RPM",2829 "TPM",2830 "Batch queue limit"2831 ],2832 "note": null,2833 "tiers": {2834 "Tier 1": {2835 "RPM": 500,2836 "TPM": 200000,2837 "Batch queue limit": 20000002838 },2839 "Tier 2": {2840 "RPM": 5000,2841 "TPM": 2000000,2842 "Batch queue limit": 200000002843 },2844 "Tier 3": {2845 "RPM": 5000,2846 "TPM": 4000000,2847 "Batch queue limit": 400000002848 },2849 "Tier 4": {2850 "RPM": 10000,2851 "TPM": 10000000,2852 "Batch queue limit": 10000000002853 },2854 "Tier 5": {2855 "RPM": 30000,2856 "TPM": 180000000,2857 "Batch queue limit": 150000000002858 }2859 }2860 }2861 },2862 "gpt-5.4-pro": {2863 "Standard": {2864 "metrics": [2865 "RPM",2866 "TPM",2867 "Batch queue limit"2868 ],2869 "note": null,2870 "tiers": {2871 "Tier 1": {2872 "RPM": 50,2873 "TPM": 50000,2874 "Batch queue limit": 9000002875 },2876 "Tier 2": {2877 "RPM": 500,2878 "TPM": 100000,2879 "Batch queue limit": 13500002880 },2881 "Tier 3": {2882 "RPM": 500,2883 "TPM": 200000,2884 "Batch queue limit": 1000000002885 },2886 "Tier 4": {2887 "RPM": 1000,2888 "TPM": 400000,2889 "Batch queue limit": 2000000002890 },2891 "Tier 5": {2892 "RPM": 1500,2893 "TPM": 4000000,2894 "Batch queue limit": 150000000002895 }2896 }2897 },2898 "Long Context": {2899 "metrics": [2900 "RPM",2901 "TPM",2902 "Batch queue limit"2903 ],2904 "note": "272K input tokens",2905 "tiers": {2906 "Tier 1": {2907 "RPM": 20,2908 "TPM": 40000,2909 "Batch queue limit": 20000002910 },2911 "Tier 2": {2912 "RPM": 50,2913 "TPM": 100000,2914 "Batch queue limit": 200000002915 },2916 "Tier 3": {2917 "RPM": 100,2918 "TPM": 200000,2919 "Batch queue limit": 400000002920 },2921 "Tier 4": {2922 "RPM": 200,2923 "TPM": 1000000,2924 "Batch queue limit": 1000000002925 },2926 "Tier 5": {2927 "RPM": 800,2928 "TPM": 2000000,2929 "Batch queue limit": 10000000002930 }2931 }2932 }2933 },2934 "gpt-5.5": {2935 "Standard": {2936 "metrics": [2937 "RPM",2938 "TPM",2939 "Batch queue limit"2940 ],2941 "note": null,2942 "tiers": {2943 "Tier 1": {2944 "RPM": 500,2945 "TPM": 500000,2946 "Batch queue limit": 15000002947 },2948 "Tier 2": {2949 "RPM": 5000,2950 "TPM": 1000000,2951 "Batch queue limit": 30000002952 },2953 "Tier 3": {2954 "RPM": 5000,2955 "TPM": 2000000,2956 "Batch queue limit": 1000000002957 },2958 "Tier 4": {2959 "RPM": 10000,2960 "TPM": 4000000,2961 "Batch queue limit": 2000000002962 },2963 "Tier 5": {2964 "RPM": 15000,2965 "TPM": 40000000,2966 "Batch queue limit": 150000000002967 }2968 }2969 },2970 "Long Context": {2971 "metrics": [2972 "RPM",2973 "TPM",2974 "Batch queue limit"2975 ],2976 "note": "272K input tokens",2977 "tiers": {2978 "Tier 1": {2979 "RPM": 200,2980 "TPM": 400000,2981 "Batch queue limit": 50000002982 },2983 "Tier 2": {2984 "RPM": 500,2985 "TPM": 1000000,2986 "Batch queue limit": 400000002987 },2988 "Tier 3": {2989 "RPM": 1000,2990 "TPM": 2000000,2991 "Batch queue limit": 800000002992 },2993 "Tier 4": {2994 "RPM": 2000,2995 "TPM": 10000000,2996 "Batch queue limit": 2000000002997 },2998 "Tier 5": {2999 "RPM": 8000,3000 "TPM": 20000000,3001 "Batch queue limit": 20000000003002 }3003 }3004 }3005 },3006 "gpt-5.5-pro": {3007 "default": {3008 "metrics": [3009 "RPM",3010 "TPM",3011 "Batch queue limit"3012 ],3013 "note": null,3014 "tiers": {3015 "Tier 1": {3016 "RPM": 50,3017 "TPM": 50000,3018 "Batch queue limit": 5000003019 },3020 "Tier 2": {3021 "RPM": 500,3022 "TPM": 200000,3023 "Batch queue limit": 10000003024 },3025 "Tier 3": {3026 "RPM": 500,3027 "TPM": 500000,3028 "Batch queue limit": 100000003029 },3030 "Tier 4": {3031 "RPM": 1000,3032 "TPM": 1000000,3033 "Batch queue limit": 200000003034 },3035 "Tier 5": {3036 "RPM": 2000,3037 "TPM": 4000000,3038 "Batch queue limit": 15000000003039 }3040 }3041 }3042 },3043 "gpt-5.6-cyber": {3044 "Standard": {3045 "metrics": [3046 "RPM",3047 "TPM",3048 "Batch queue limit"3049 ],3050 "note": null,3051 "tiers": {3052 "Tier 1": {3053 "RPM": 500,3054 "TPM": 500000,3055 "Batch queue limit": 15000003056 },3057 "Tier 2": {3058 "RPM": 5000,3059 "TPM": 1000000,3060 "Batch queue limit": 30000003061 },3062 "Tier 3": {3063 "RPM": 5000,3064 "TPM": 2000000,3065 "Batch queue limit": 1000000003066 },3067 "Tier 4": {3068 "RPM": 10000,3069 "TPM": 4000000,3070 "Batch queue limit": 2000000003071 },3072 "Tier 5": {3073 "RPM": 15000,3074 "TPM": 40000000,3075 "Batch queue limit": 150000000003076 }3077 }3078 }3079 },3080 "gpt-5.6-luna": {3081 "default": {3082 "metrics": [3083 "RPM",3084 "TPM",3085 "Batch queue limit"3086 ],3087 "note": null,3088 "tiers": {3089 "Tier 1": {3090 "RPM": 500,3091 "TPM": 500000,3092 "Batch queue limit": 50000003093 },3094 "Tier 2": {3095 "RPM": 5000,3096 "TPM": 2000000,3097 "Batch queue limit": 200000003098 },3099 "Tier 3": {3100 "RPM": 5000,3101 "TPM": 4000000,3102 "Batch queue limit": 400000003103 },3104 "Tier 4": {3105 "RPM": 10000,3106 "TPM": 10000000,3107 "Batch queue limit": 10000000003108 },3109 "Tier 5": {3110 "RPM": 30000,3111 "TPM": 180000000,3112 "Batch queue limit": 150000000003113 }3114 }3115 }3116 },3117 "gpt-5.6-sol": {3118 "Standard": {3119 "metrics": [3120 "RPM",3121 "TPM",3122 "Batch queue limit"3123 ],3124 "note": null,3125 "tiers": {3126 "Tier 1": {3127 "RPM": 500,3128 "TPM": 500000,3129 "Batch queue limit": 15000003130 },3131 "Tier 2": {3132 "RPM": 5000,3133 "TPM": 1000000,3134 "Batch queue limit": 30000003135 },3136 "Tier 3": {3137 "RPM": 5000,3138 "TPM": 2000000,3139 "Batch queue limit": 1000000003140 },3141 "Tier 4": {3142 "RPM": 10000,3143 "TPM": 4000000,3144 "Batch queue limit": 2000000003145 },3146 "Tier 5": {3147 "RPM": 15000,3148 "TPM": 40000000,3149 "Batch queue limit": 150000000003150 }3151 }3152 }3153 },3154 "gpt-5.6-terra": {3155 "Standard": {3156 "metrics": [3157 "RPM",3158 "TPM",3159 "Batch queue limit"3160 ],3161 "note": null,3162 "tiers": {3163 "Tier 1": {3164 "RPM": 500,3165 "TPM": 500000,3166 "Batch queue limit": 15000003167 },3168 "Tier 2": {3169 "RPM": 5000,3170 "TPM": 1000000,3171 "Batch queue limit": 30000003172 },3173 "Tier 3": {3174 "RPM": 5000,3175 "TPM": 2000000,3176 "Batch queue limit": 1000000003177 },3178 "Tier 4": {3179 "RPM": 10000,3180 "TPM": 4000000,3181 "Batch queue limit": 2000000003182 },3183 "Tier 5": {3184 "RPM": 15000,3185 "TPM": 40000000,3186 "Batch queue limit": 150000000003187 }3188 }3189 }3190 },3191 "gpt-6-astra": {3192 "Standard": {3193 "metrics": [3194 "RPM",3195 "TPM",3196 "Batch queue limit"3197 ],3198 "note": null,3199 "tiers": {3200 "Tier 1": {3201 "RPM": 500,3202 "TPM": 500000,3203 "Batch queue limit": 15000003204 },3205 "Tier 2": {3206 "RPM": 5000,3207 "TPM": 1000000,3208 "Batch queue limit": 30000003209 },3210 "Tier 3": {3211 "RPM": 5000,3212 "TPM": 2000000,3213 "Batch queue limit": 1000000003214 },3215 "Tier 4": {3216 "RPM": 10000,3217 "TPM": 4000000,3218 "Batch queue limit": 2000000003219 },3220 "Tier 5": {3221 "RPM": 15000,3222 "TPM": 40000000,3223 "Batch queue limit": 150000000003224 }3225 }3226 }3227 },3228 "gpt-audio": {3229 "default": {3230 "metrics": [3231 "RPM",3232 "TPM",3233 "Batch queue limit"3234 ],3235 "note": null,3236 "tiers": {3237 "Tier 1": {3238 "RPM": 500,3239 "TPM": 30000,3240 "Batch queue limit": 900003241 },3242 "Tier 2": {3243 "RPM": 5000,3244 "TPM": 450000,3245 "Batch queue limit": 13500003246 },3247 "Tier 3": {3248 "RPM": 5000,3249 "TPM": 800000,3250 "Batch queue limit": 500000003251 },3252 "Tier 4": {3253 "RPM": 10000,3254 "TPM": 2000000,3255 "Batch queue limit": 20000003256 },3257 "Tier 5": {3258 "RPM": 10000,3259 "TPM": 30000000,3260 "Batch queue limit": 50000000003261 }3262 }3263 }3264 },3265 "gpt-audio-1.5": {3266 "default": {3267 "metrics": [3268 "RPM",3269 "TPM",3270 "Batch queue limit"3271 ],3272 "note": null,3273 "tiers": {3274 "Tier 1": {3275 "RPM": 500,3276 "TPM": 30000,3277 "Batch queue limit": 900003278 },3279 "Tier 2": {3280 "RPM": 5000,3281 "TPM": 450000,3282 "Batch queue limit": 13500003283 },3284 "Tier 3": {3285 "RPM": 5000,3286 "TPM": 800000,3287 "Batch queue limit": 500000003288 },3289 "Tier 4": {3290 "RPM": 10000,3291 "TPM": 2000000,3292 "Batch queue limit": 20000003293 },3294 "Tier 5": {3295 "RPM": 10000,3296 "TPM": 30000000,3297 "Batch queue limit": 50000000003298 }3299 }3300 }3301 },3302 "gpt-audio-mini": {3303 "default": {3304 "metrics": [3305 "RPM",3306 "TPM"3307 ],3308 "note": null,3309 "tiers": {3310 "Tier 1": {3311 "RPM": 500,3312 "TPM": 300003313 },3314 "Tier 2": {3315 "RPM": 5000,3316 "TPM": 4500003317 },3318 "Tier 3": {3319 "RPM": 5000,3320 "TPM": 8000003321 },3322 "Tier 4": {3323 "RPM": 10000,3324 "TPM": 20000003325 },3326 "Tier 5": {3327 "RPM": 10000,3328 "TPM": 300000003329 }3330 }3331 }3332 },3333 "gpt-daybreak-blue-latest": {3334 "Standard": {3335 "metrics": [3336 "RPM",3337 "TPM",3338 "Batch queue limit"3339 ],3340 "note": null,3341 "tiers": {3342 "Tier 1": {3343 "RPM": 500,3344 "TPM": 500000,3345 "Batch queue limit": 15000003346 },3347 "Tier 2": {3348 "RPM": 5000,3349 "TPM": 1000000,3350 "Batch queue limit": 30000003351 },3352 "Tier 3": {3353 "RPM": 5000,3354 "TPM": 2000000,3355 "Batch queue limit": 1000000003356 },3357 "Tier 4": {3358 "RPM": 10000,3359 "TPM": 4000000,3360 "Batch queue limit": 2000000003361 },3362 "Tier 5": {3363 "RPM": 15000,3364 "TPM": 40000000,3365 "Batch queue limit": 150000000003366 }3367 }3368 }3369 },3370 "gpt-daybreak-red-latest": {3371 "Standard": {3372 "metrics": [3373 "RPM",3374 "TPM",3375 "Batch queue limit"3376 ],3377 "note": null,3378 "tiers": {3379 "Tier 1": {3380 "RPM": 500,3381 "TPM": 500000,3382 "Batch queue limit": 15000003383 },3384 "Tier 2": {3385 "RPM": 5000,3386 "TPM": 1000000,3387 "Batch queue limit": 30000003388 },3389 "Tier 3": {3390 "RPM": 5000,3391 "TPM": 2000000,3392 "Batch queue limit": 1000000003393 },3394 "Tier 4": {3395 "RPM": 10000,3396 "TPM": 4000000,3397 "Batch queue limit": 2000000003398 },3399 "Tier 5": {3400 "RPM": 15000,3401 "TPM": 40000000,3402 "Batch queue limit": 150000000003403 }3404 }3405 }3406 },3407 "gpt-image-1": {3408 "default": {3409 "metrics": [3410 "TPM",3411 "IPM"3412 ],3413 "note": null,3414 "tiers": {3415 "Tier 1": {3416 "TPM": 100000,3417 "IPM": 53418 },3419 "Tier 2": {3420 "TPM": 250000,3421 "IPM": 203422 },3423 "Tier 3": {3424 "TPM": 800000,3425 "IPM": 503426 },3427 "Tier 4": {3428 "TPM": 3000000,3429 "IPM": 1503430 },3431 "Tier 5": {3432 "TPM": 8000000,3433 "IPM": 2503434 }3435 }3436 }3437 },3438 "gpt-image-1-mini": {3439 "default": {3440 "metrics": [3441 "IPM",3442 "TPM"3443 ],3444 "note": null,3445 "tiers": {3446 "Tier 1": {3447 "IPM": 5,3448 "TPM": 1000003449 },3450 "Tier 2": {3451 "IPM": 20,3452 "TPM": 2500003453 },3454 "Tier 3": {3455 "IPM": 50,3456 "TPM": 8000003457 },3458 "Tier 4": {3459 "IPM": 150,3460 "TPM": 30000003461 },3462 "Tier 5": {3463 "IPM": 250,3464 "TPM": 80000003465 }3466 }3467 }3468 },3469 "gpt-image-1.5": {3470 "default": {3471 "metrics": [3472 "TPM",3473 "IPM"3474 ],3475 "note": null,3476 "tiers": {3477 "Tier 1": {3478 "TPM": 100000,3479 "IPM": 53480 },3481 "Tier 2": {3482 "TPM": 250000,3483 "IPM": 203484 },3485 "Tier 3": {3486 "TPM": 800000,3487 "IPM": 503488 },3489 "Tier 4": {3490 "TPM": 3000000,3491 "IPM": 1503492 },3493 "Tier 5": {3494 "TPM": 8000000,3495 "IPM": 2503496 }3497 }3498 }3499 },3500 "gpt-image-2": {3501 "default": {3502 "metrics": [3503 "TPM",3504 "IPM"3505 ],3506 "note": null,3507 "tiers": {3508 "Tier 1": {3509 "TPM": 100000,3510 "IPM": 53511 },3512 "Tier 2": {3513 "TPM": 250000,3514 "IPM": 203515 },3516 "Tier 3": {3517 "TPM": 800000,3518 "IPM": 503519 },3520 "Tier 4": {3521 "TPM": 3000000,3522 "IPM": 1503523 },3524 "Tier 5": {3525 "TPM": 8000000,3526 "IPM": 2503527 }3528 }3529 }3530 },3531 "gpt-image-2.5-flare": {3532 "default": {3533 "metrics": [3534 "TPM",3535 "IPM"3536 ],3537 "note": null,3538 "tiers": {3539 "Tier 1": {3540 "TPM": 100000,3541 "IPM": 53542 },3543 "Tier 2": {3544 "TPM": 250000,3545 "IPM": 203546 },3547 "Tier 3": {3548 "TPM": 800000,3549 "IPM": 503550 },3551 "Tier 4": {3552 "TPM": 3000000,3553 "IPM": 1503554 },3555 "Tier 5": {3556 "TPM": 8000000,3557 "IPM": 2503558 }3559 }3560 }3561 },3562 "gpt-image-2.5-sunburst": {3563 "default": {3564 "metrics": [3565 "TPM",3566 "IPM"3567 ],3568 "note": null,3569 "tiers": {3570 "Tier 1": {3571 "TPM": 100000,3572 "IPM": 53573 },3574 "Tier 2": {3575 "TPM": 250000,3576 "IPM": 203577 },3578 "Tier 3": {3579 "TPM": 800000,3580 "IPM": 503581 },3582 "Tier 4": {3583 "TPM": 3000000,3584 "IPM": 1503585 },3586 "Tier 5": {3587 "TPM": 8000000,3588 "IPM": 2503589 }3590 }3591 }3592 },3593 "gpt-live-1": {3594 "Concurrent sessions": {3595 "metrics": [3596 "Concurrent sessions"3597 ],3598 "note": null,3599 "tiers": {3600 "Tier 1": {3601 "Concurrent sessions": 253602 },3603 "Tier 2": {3604 "Concurrent sessions": 503605 },3606 "Tier 3": {3607 "Concurrent sessions": 2003608 },3609 "Tier 4": {3610 "Concurrent sessions": 3003611 },3612 "Tier 5": {3613 "Concurrent sessions": 5003614 }3615 }3616 }3617 },3618 "gpt-live-transcribe": {3619 "default": {3620 "metrics": [3621 "RPM",3622 "TPM"3623 ],3624 "note": null,3625 "tiers": {3626 "Tier 1": {3627 "RPM": 500,3628 "TPM": 600003629 },3630 "Tier 2": {3631 "RPM": 2000,3632 "TPM": 2100003633 },3634 "Tier 3": {3635 "RPM": 5000,3636 "TPM": 3900003637 },3638 "Tier 4": {3639 "RPM": 10000,3640 "TPM": 6000003641 },3642 "Tier 5": {3643 "RPM": 10000,3644 "TPM": 7800003645 }3646 }3647 }3648 },3649 "gpt-oss-120b": {3650 "default": {3651 "metrics": [3652 "RPM",3653 "TPM",3654 "Batch queue limit"3655 ],3656 "note": null,3657 "tiers": {3658 "Tier 1": {3659 "RPM": 0,3660 "TPM": 0,3661 "Batch queue limit": 03662 },3663 "Tier 2": {3664 "RPM": 0,3665 "TPM": 0,3666 "Batch queue limit": 03667 },3668 "Tier 3": {3669 "RPM": 0,3670 "TPM": 0,3671 "Batch queue limit": 03672 },3673 "Tier 4": {3674 "RPM": 0,3675 "TPM": 0,3676 "Batch queue limit": 03677 },3678 "Tier 5": {3679 "RPM": 0,3680 "TPM": 0,3681 "Batch queue limit": 03682 }3683 }3684 }3685 },3686 "gpt-oss-20b": {3687 "default": {3688 "metrics": [3689 "RPM",3690 "TPM",3691 "Batch queue limit"3692 ],3693 "note": null,3694 "tiers": {3695 "Tier 1": {3696 "RPM": 0,3697 "TPM": 0,3698 "Batch queue limit": 03699 },3700 "Tier 2": {3701 "RPM": 0,3702 "TPM": 0,3703 "Batch queue limit": 03704 },3705 "Tier 3": {3706 "RPM": 0,3707 "TPM": 0,3708 "Batch queue limit": 03709 },3710 "Tier 4": {3711 "RPM": 0,3712 "TPM": 0,3713 "Batch queue limit": 03714 },3715 "Tier 5": {3716 "RPM": 0,3717 "TPM": 0,3718 "Batch queue limit": 03719 }3720 }3721 }3722 },3723 "gpt-realtime": {3724 "default": {3725 "metrics": [3726 "RPM",3727 "RPD",3728 "TPM"3729 ],3730 "note": null,3731 "tiers": {3732 "Tier 1": {3733 "RPM": 200,3734 "RPD": 1000,3735 "TPM": 400003736 },3737 "Tier 2": {3738 "RPM": 400,3739 "RPD": null,3740 "TPM": 2000003741 },3742 "Tier 3": {3743 "RPM": 5000,3744 "RPD": null,3745 "TPM": 8000003746 },3747 "Tier 4": {3748 "RPM": 10000,3749 "RPD": null,3750 "TPM": 40000003751 },3752 "Tier 5": {3753 "RPM": 20000,3754 "RPD": null,3755 "TPM": 150000003756 }3757 }3758 }3759 },3760 "gpt-realtime-1.5": {3761 "default": {3762 "metrics": [3763 "RPM",3764 "RPD",3765 "TPM"3766 ],3767 "note": null,3768 "tiers": {3769 "Tier 1": {3770 "RPM": 200,3771 "RPD": 1000,3772 "TPM": 400003773 },3774 "Tier 2": {3775 "RPM": 400,3776 "RPD": null,3777 "TPM": 2000003778 },3779 "Tier 3": {3780 "RPM": 5000,3781 "RPD": null,3782 "TPM": 8000003783 },3784 "Tier 4": {3785 "RPM": 10000,3786 "RPD": null,3787 "TPM": 40000003788 },3789 "Tier 5": {3790 "RPM": 20000,3791 "RPD": null,3792 "TPM": 150000003793 }3794 }3795 }3796 },3797 "gpt-realtime-2": {3798 "default": {3799 "metrics": [3800 "RPM",3801 "RPD",3802 "TPM"3803 ],3804 "note": null,3805 "tiers": {3806 "Tier 1": {3807 "RPM": 200,3808 "RPD": 1000,3809 "TPM": 400003810 },3811 "Tier 2": {3812 "RPM": 400,3813 "RPD": null,3814 "TPM": 2000003815 },3816 "Tier 3": {3817 "RPM": 5000,3818 "RPD": null,3819 "TPM": 8000003820 },3821 "Tier 4": {3822 "RPM": 10000,3823 "RPD": null,3824 "TPM": 40000003825 },3826 "Tier 5": {3827 "RPM": 20000,3828 "RPD": null,3829 "TPM": 150000003830 }3831 }3832 }3833 },3834 "gpt-realtime-2.1": {3835 "default": {3836 "metrics": [3837 "RPM",3838 "RPD",3839 "TPM"3840 ],3841 "note": null,3842 "tiers": {3843 "Tier 1": {3844 "RPM": 200,3845 "RPD": 1000,3846 "TPM": 400003847 },3848 "Tier 2": {3849 "RPM": 400,3850 "RPD": null,3851 "TPM": 2000003852 },3853 "Tier 3": {3854 "RPM": 5000,3855 "RPD": null,3856 "TPM": 8000003857 },3858 "Tier 4": {3859 "RPM": 10000,3860 "RPD": null,3861 "TPM": 40000003862 },3863 "Tier 5": {3864 "RPM": 20000,3865 "RPD": null,3866 "TPM": 150000003867 }3868 }3869 }3870 },3871 "gpt-realtime-2.1-mini": {3872 "default": {3873 "metrics": [3874 "RPM",3875 "TPM"3876 ],3877 "note": null,3878 "tiers": {3879 "Tier 1": {3880 "RPM": 200,3881 "TPM": 400003882 },3883 "Tier 2": {3884 "RPM": 400,3885 "TPM": 2000003886 },3887 "Tier 3": {3888 "RPM": 5000,3889 "TPM": 8000003890 },3891 "Tier 4": {3892 "RPM": 10000,3893 "TPM": 40000003894 },3895 "Tier 5": {3896 "RPM": 20000,3897 "TPM": 150000003898 }3899 }3900 }3901 },3902 "gpt-realtime-mini": {3903 "default": {3904 "metrics": [3905 "RPM",3906 "TPM"3907 ],3908 "note": null,3909 "tiers": {3910 "Tier 1": {3911 "RPM": 200,3912 "TPM": 400003913 },3914 "Tier 2": {3915 "RPM": 400,3916 "TPM": 2000003917 },3918 "Tier 3": {3919 "RPM": 5000,3920 "TPM": 8000003921 },3922 "Tier 4": {3923 "RPM": 10000,3924 "TPM": 40000003925 },3926 "Tier 5": {3927 "RPM": 20000,3928 "TPM": 150000003929 }3930 }3931 }3932 },3933 "gpt-realtime-translate": {3934 "default": {3935 "metrics": [3936 "Minutes-of-audio per minute"3937 ],3938 "note": null,3939 "tiers": {3940 "Tier 1": {3941 "Minutes-of-audio per minute": 503942 },3943 "Tier 2": {3944 "Minutes-of-audio per minute": 2003945 },3946 "Tier 3": {3947 "Minutes-of-audio per minute": 4003948 },3949 "Tier 4": {3950 "Minutes-of-audio per minute": 6503951 },3952 "Tier 5": {3953 "Minutes-of-audio per minute": 8503954 }3955 }3956 }3957 },3958 "gpt-realtime-whisper": {3959 "default": {3960 "metrics": [3961 "Minutes-of-audio per minute"3962 ],3963 "note": null,3964 "tiers": {3965 "Tier 1": {3966 "Minutes-of-audio per minute": 1003967 },3968 "Tier 2": {3969 "Minutes-of-audio per minute": 3503970 },3971 "Tier 3": {3972 "Minutes-of-audio per minute": 6503973 },3974 "Tier 4": {3975 "Minutes-of-audio per minute": 10003976 },3977 "Tier 5": {3978 "Minutes-of-audio per minute": 13003979 }3980 }3981 }3982 },3983 "gpt-transcribe": {3984 "default": {3985 "metrics": [3986 "RPM",3987 "TPM"3988 ],3989 "note": null,3990 "tiers": {3991 "Tier 1": {3992 "RPM": 500,3993 "TPM": 2000003994 },3995 "Tier 2": {3996 "RPM": 5000,3997 "TPM": 20000003998 },3999 "Tier 3": {4000 "RPM": 5000,4001 "TPM": 40000004002 },4003 "Tier 4": {4004 "RPM": 10000,4005 "TPM": 100000004006 },4007 "Tier 5": {4008 "RPM": 30000,4009 "TPM": 1500000004010 }4011 }4012 }4013 },4014 "o1": {4015 "default": {4016 "metrics": [4017 "RPM",4018 "TPM",4019 "Batch queue limit"4020 ],4021 "note": null,4022 "tiers": {4023 "Tier 1": {4024 "RPM": 500,4025 "TPM": 30000,4026 "Batch queue limit": 900004027 },4028 "Tier 2": {4029 "RPM": 5000,4030 "TPM": 450000,4031 "Batch queue limit": 13500004032 },4033 "Tier 3": {4034 "RPM": 5000,4035 "TPM": 800000,4036 "Batch queue limit": 500000004037 },4038 "Tier 4": {4039 "RPM": 10000,4040 "TPM": 2000000,4041 "Batch queue limit": 2000000004042 },4043 "Tier 5": {4044 "RPM": 10000,4045 "TPM": 30000000,4046 "Batch queue limit": 50000000004047 }4048 }4049 }4050 },4051 "o1-mini": {4052 "default": {4053 "metrics": [4054 "RPM",4055 "TPM",4056 "Batch queue limit"4057 ],4058 "note": null,4059 "tiers": {4060 "Tier 1": {4061 "RPM": 500,4062 "TPM": 200000,4063 "Batch queue limit": null4064 },4065 "Tier 2": {4066 "RPM": 5000,4067 "TPM": 2000000,4068 "Batch queue limit": null4069 },4070 "Tier 3": {4071 "RPM": 5000,4072 "TPM": 4000000,4073 "Batch queue limit": 400000004074 },4075 "Tier 4": {4076 "RPM": 10000,4077 "TPM": 10000000,4078 "Batch queue limit": 10000000004079 },4080 "Tier 5": {4081 "RPM": 30000,4082 "TPM": 150000000,4083 "Batch queue limit": 150000000004084 }4085 }4086 }4087 },4088 "o1-preview": {4089 "default": {4090 "metrics": [4091 "RPM",4092 "TPM",4093 "Batch queue limit"4094 ],4095 "note": null,4096 "tiers": {4097 "Tier 1": {4098 "RPM": 500,4099 "TPM": 30000,4100 "Batch queue limit": null4101 },4102 "Tier 2": {4103 "RPM": 5000,4104 "TPM": 450000,4105 "Batch queue limit": null4106 },4107 "Tier 3": {4108 "RPM": 5000,4109 "TPM": 800000,4110 "Batch queue limit": 500000004111 },4112 "Tier 4": {4113 "RPM": 10000,4114 "TPM": 2000000,4115 "Batch queue limit": 2000000004116 },4117 "Tier 5": {4118 "RPM": 10000,4119 "TPM": 30000000,4120 "Batch queue limit": 50000000004121 }4122 }4123 }4124 },4125 "o1-pro": {4126 "default": {4127 "metrics": [4128 "RPM",4129 "TPM",4130 "Batch queue limit"4131 ],4132 "note": null,4133 "tiers": {4134 "Tier 1": {4135 "RPM": 500,4136 "TPM": 30000,4137 "Batch queue limit": 900004138 },4139 "Tier 2": {4140 "RPM": 5000,4141 "TPM": 450000,4142 "Batch queue limit": 13500004143 },4144 "Tier 3": {4145 "RPM": 5000,4146 "TPM": 800000,4147 "Batch queue limit": 500000004148 },4149 "Tier 4": {4150 "RPM": 10000,4151 "TPM": 2000000,4152 "Batch queue limit": 2000000004153 },4154 "Tier 5": {4155 "RPM": 10000,4156 "TPM": 30000000,4157 "Batch queue limit": 50000000004158 }4159 }4160 }4161 },4162 "o3": {4163 "default": {4164 "metrics": [4165 "RPM",4166 "TPM",4167 "Batch queue limit"4168 ],4169 "note": null,4170 "tiers": {4171 "Tier 1": {4172 "RPM": 500,4173 "TPM": 30000,4174 "Batch queue limit": 900004175 },4176 "Tier 2": {4177 "RPM": 5000,4178 "TPM": 450000,4179 "Batch queue limit": 13500004180 },4181 "Tier 3": {4182 "RPM": 5000,4183 "TPM": 800000,4184 "Batch queue limit": 500000004185 },4186 "Tier 4": {4187 "RPM": 10000,4188 "TPM": 2000000,4189 "Batch queue limit": 2000000004190 },4191 "Tier 5": {4192 "RPM": 10000,4193 "TPM": 30000000,4194 "Batch queue limit": 50000000004195 }4196 }4197 }4198 },4199 "o3-deep-research": {4200 "default": {4201 "metrics": [4202 "RPM",4203 "TPM",4204 "Batch queue limit"4205 ],4206 "note": null,4207 "tiers": {4208 "Tier 1": {4209 "RPM": 500,4210 "TPM": 200000,4211 "Batch queue limit": 2000004212 },4213 "Tier 2": {4214 "RPM": 5000,4215 "TPM": 450000,4216 "Batch queue limit": 3000004217 },4218 "Tier 3": {4219 "RPM": 5000,4220 "TPM": 800000,4221 "Batch queue limit": 5000004222 },4223 "Tier 4": {4224 "RPM": 10000,4225 "TPM": 2000000,4226 "Batch queue limit": 20000004227 },4228 "Tier 5": {4229 "RPM": 10000,4230 "TPM": 30000000,4231 "Batch queue limit": 100000004232 }4233 }4234 }4235 },4236 "o3-mini": {4237 "default": {4238 "metrics": [4239 "RPM",4240 "TPM",4241 "Batch queue limit"4242 ],4243 "note": null,4244 "tiers": {4245 "Tier 1": {4246 "RPM": 1000,4247 "TPM": 100000,4248 "Batch queue limit": 10000004249 },4250 "Tier 2": {4251 "RPM": 2000,4252 "TPM": 200000,4253 "Batch queue limit": 20000004254 },4255 "Tier 3": {4256 "RPM": 5000,4257 "TPM": 4000000,4258 "Batch queue limit": 400000004259 },4260 "Tier 4": {4261 "RPM": 10000,4262 "TPM": 10000000,4263 "Batch queue limit": 10000000004264 },4265 "Tier 5": {4266 "RPM": 30000,4267 "TPM": 150000000,4268 "Batch queue limit": 150000000004269 }4270 }4271 }4272 },4273 "o3-pro": {4274 "default": {4275 "metrics": [4276 "RPM",4277 "TPM",4278 "Batch queue limit"4279 ],4280 "note": null,4281 "tiers": {4282 "Tier 1": {4283 "RPM": 500,4284 "TPM": 30000,4285 "Batch queue limit": 900004286 },4287 "Tier 2": {4288 "RPM": 5000,4289 "TPM": 450000,4290 "Batch queue limit": 13500004291 },4292 "Tier 3": {4293 "RPM": 5000,4294 "TPM": 800000,4295 "Batch queue limit": 500000004296 },4297 "Tier 4": {4298 "RPM": 10000,4299 "TPM": 2000000,4300 "Batch queue limit": 2000000004301 },4302 "Tier 5": {4303 "RPM": 10000,4304 "TPM": 30000000,4305 "Batch queue limit": 50000000004306 }4307 }4308 }4309 },4310 "o4-mini": {4311 "default": {4312 "metrics": [4313 "RPM",4314 "TPM",4315 "Batch queue limit"4316 ],4317 "note": null,4318 "tiers": {4319 "Tier 1": {4320 "RPM": 1000,4321 "TPM": 100000,4322 "Batch queue limit": 10000004323 },4324 "Tier 2": {4325 "RPM": 2000,4326 "TPM": 2000000,4327 "Batch queue limit": 20000004328 },4329 "Tier 3": {4330 "RPM": 5000,4331 "TPM": 4000000,4332 "Batch queue limit": 400000004333 },4334 "Tier 4": {4335 "RPM": 10000,4336 "TPM": 10000000,4337 "Batch queue limit": 10000000004338 },4339 "Tier 5": {4340 "RPM": 30000,4341 "TPM": 150000000,4342 "Batch queue limit": 150000000004343 }4344 }4345 }4346 },4347 "o4-mini-deep-research": {4348 "default": {4349 "metrics": [4350 "RPM",4351 "TPM",4352 "Batch queue limit"4353 ],4354 "note": null,4355 "tiers": {4356 "Tier 1": {4357 "RPM": 1000,4358 "TPM": 200000,4359 "Batch queue limit": 2000004360 },4361 "Tier 2": {4362 "RPM": 2000,4363 "TPM": 2000000,4364 "Batch queue limit": 3000004365 },4366 "Tier 3": {4367 "RPM": 5000,4368 "TPM": 4000000,4369 "Batch queue limit": 5000004370 },4371 "Tier 4": {4372 "RPM": 10000,4373 "TPM": 10000000,4374 "Batch queue limit": 20000004375 },4376 "Tier 5": {4377 "RPM": 30000,4378 "TPM": 150000000,4379 "Batch queue limit": 100000004380 }4381 }4382 }4383 },4384 "omni-moderation-latest": {4385 "default": {4386 "metrics": [4387 "RPM",4388 "RPD",4389 "TPM"4390 ],4391 "note": null,4392 "tiers": {4393 "free": {4394 "RPM": 250,4395 "RPD": 5000,4396 "TPM": 100004397 },4398 "Tier 1": {4399 "RPM": 500,4400 "RPD": 10000,4401 "TPM": 100004402 },4403 "Tier 2": {4404 "RPM": 500,4405 "RPD": null,4406 "TPM": 200004407 },4408 "Tier 3": {4409 "RPM": 1000,4410 "RPD": null,4411 "TPM": 500004412 },4413 "Tier 4": {4414 "RPM": 2000,4415 "RPD": null,4416 "TPM": 2500004417 },4418 "Tier 5": {4419 "RPM": 5000,4420 "RPD": null,4421 "TPM": 5000004422 }4423 }4424 }4425 },4426 "sora-2": {4427 "Standard RPM": {4428 "metrics": [4429 "RPM"4430 ],4431 "note": null,4432 "tiers": {4433 "Tier 1": {4434 "RPM": 254435 },4436 "Tier 2": {4437 "RPM": 504438 },4439 "Tier 3": {4440 "RPM": 1254441 },4442 "Tier 4": {4443 "RPM": 2004444 },4445 "Tier 5": {4446 "RPM": 3754447 }4448 }4449 }4450 },4451 "sora-2-pro": {4452 "Standard RPM": {4453 "metrics": [4454 "RPM"4455 ],4456 "note": null,4457 "tiers": {4458 "Tier 1": {4459 "RPM": 104460 },4461 "Tier 2": {4462 "RPM": 254463 },4464 "Tier 3": {4465 "RPM": 504466 },4467 "Tier 4": {4468 "RPM": 754469 },4470 "Tier 5": {4471 "RPM": 1504472 }4473 }4474 }4475 },4476 "text-embedding-3-large": {4477 "default": {4478 "metrics": [4479 "RPM",4480 "RPD",4481 "TPM",4482 "Batch queue limit"4483 ],4484 "note": null,4485 "tiers": {4486 "free": {4487 "RPM": 100,4488 "RPD": 2000,4489 "TPM": 40000,4490 "Batch queue limit": null4491 },4492 "Tier 1": {4493 "RPM": 3000,4494 "RPD": null,4495 "TPM": 1000000,4496 "Batch queue limit": 30000004497 },4498 "Tier 2": {4499 "RPM": 5000,4500 "RPD": null,4501 "TPM": 1000000,4502 "Batch queue limit": 200000004503 },4504 "Tier 3": {4505 "RPM": 5000,4506 "RPD": null,4507 "TPM": 5000000,4508 "Batch queue limit": 1000000004509 },4510 "Tier 4": {4511 "RPM": 10000,4512 "RPD": null,4513 "TPM": 5000000,4514 "Batch queue limit": 5000000004515 },4516 "Tier 5": {4517 "RPM": 10000,4518 "RPD": null,4519 "TPM": 10000000,4520 "Batch queue limit": 40000000004521 }4522 }4523 }4524 },4525 "text-embedding-3-small": {4526 "default": {4527 "metrics": [4528 "RPM",4529 "RPD",4530 "TPM",4531 "Batch queue limit"4532 ],4533 "note": null,4534 "tiers": {4535 "free": {4536 "RPM": 100,4537 "RPD": 2000,4538 "TPM": 40000,4539 "Batch queue limit": null4540 },4541 "Tier 1": {4542 "RPM": 3000,4543 "RPD": null,4544 "TPM": 1000000,4545 "Batch queue limit": 30000004546 },4547 "Tier 2": {4548 "RPM": 5000,4549 "RPD": null,4550 "TPM": 1000000,4551 "Batch queue limit": 200000004552 },4553 "Tier 3": {4554 "RPM": 5000,4555 "RPD": null,4556 "TPM": 5000000,4557 "Batch queue limit": 1000000004558 },4559 "Tier 4": {4560 "RPM": 10000,4561 "RPD": null,4562 "TPM": 5000000,4563 "Batch queue limit": 5000000004564 },4565 "Tier 5": {4566 "RPM": 10000,4567 "RPD": null,4568 "TPM": 10000000,4569 "Batch queue limit": 40000000004570 }4571 }4572 }4573 },4574 "text-embedding-ada-002": {4575 "default": {4576 "metrics": [4577 "RPM",4578 "RPD",4579 "TPM",4580 "Batch queue limit"4581 ],4582 "note": null,4583 "tiers": {4584 "free": {4585 "RPM": 100,4586 "RPD": 2000,4587 "TPM": 40000,4588 "Batch queue limit": null4589 },4590 "Tier 1": {4591 "RPM": 3000,4592 "RPD": null,4593 "TPM": 1000000,4594 "Batch queue limit": 30000004595 },4596 "Tier 2": {4597 "RPM": 5000,4598 "RPD": null,4599 "TPM": 1000000,4600 "Batch queue limit": 200000004601 },4602 "Tier 3": {4603 "RPM": 5000,4604 "RPD": null,4605 "TPM": 5000000,4606 "Batch queue limit": 1000000004607 },4608 "Tier 4": {4609 "RPM": 10000,4610 "RPD": null,4611 "TPM": 5000000,4612 "Batch queue limit": 5000000004613 },4614 "Tier 5": {4615 "RPM": 10000,4616 "RPD": null,4617 "TPM": 10000000,4618 "Batch queue limit": 40000000004619 }4620 }4621 }4622 },4623 "tts-1": {4624 "default": {4625 "metrics": [4626 "RPM",4627 "RPD"4628 ],4629 "note": null,4630 "tiers": {4631 "free": {4632 "RPM": 3,4633 "RPD": 2004634 },4635 "Tier 1": {4636 "RPM": 500,4637 "RPD": null4638 },4639 "Tier 2": {4640 "RPM": 2500,4641 "RPD": null4642 },4643 "Tier 3": {4644 "RPM": 5000,4645 "RPD": null4646 },4647 "Tier 4": {4648 "RPM": 7500,4649 "RPD": null4650 },4651 "Tier 5": {4652 "RPM": 10000,4653 "RPD": null4654 }4655 }4656 }4657 },4658 "tts-1-hd": {4659 "default": {4660 "metrics": [4661 "RPM"4662 ],4663 "note": null,4664 "tiers": {4665 "free": {4666 "RPM": null4667 },4668 "Tier 1": {4669 "RPM": 5004670 },4671 "Tier 2": {4672 "RPM": 25004673 },4674 "Tier 3": {4675 "RPM": 50004676 },4677 "Tier 4": {4678 "RPM": 75004679 },4680 "Tier 5": {4681 "RPM": 100004682 }4683 }4684 }4685 },4686 "whisper-1": {4687 "default": {4688 "metrics": [4689 "RPM",4690 "RPD"4691 ],4692 "note": null,4693 "tiers": {4694 "free": {4695 "RPM": 3,4696 "RPD": 2004697 },4698 "Tier 1": {4699 "RPM": 500,4700 "RPD": null4701 },4702 "Tier 2": {4703 "RPM": 2500,4704 "RPD": null4705 },4706 "Tier 3": {4707 "RPM": 5000,4708 "RPD": null4709 },4710 "Tier 4": {4711 "RPM": 7500,4712 "RPD": null4713 },4714 "Tier 5": {4715 "RPM": 10000,4716 "RPD": null4717 }4718 }4719 }4720 }4721 },4722 "per_model_notes": {4723 "gpt-live-1": [4724 "Rate limits are measured in concurrent sessions.",4725 "Unsupported usage tiers: Free."4726 ]4727 },4728 "observed": [4729 {4730 "date": "2026-09-18",4731 "request": "POST /v1/responses (this atlas key)",4732 "headers": {4733 "x-ratelimit-limit-requests": "30000",4734 "x-ratelimit-limit-tokens": "180000000"4735 },4736 "note": "observed for OUR key/org/project on 2026-09-18 — account-specific, do not generalize"4737 },4738 {4739 "date": "2026-09-19",4740 "request": "POST /v1/responses model=gpt-5.6-luna",4741 "headers": {4742 "x-ratelimit-limit-requests": "30000",4743 "x-ratelimit-limit-tokens": "180000000",4744 "x-ratelimit-remaining-requests": "29999",4745 "x-ratelimit-remaining-tokens": "180000000",4746 "x-ratelimit-reset-requests": "2ms",4747 "x-ratelimit-reset-tokens": "0s"4748 },4749 "note": "observed for OUR key — account-specific, do not generalize"4750 },4751 {4752 "date": "2026-09-19",4753 "request": "POST /v1/responses model=gpt-5.6-sol",4754 "headers": {4755 "x-ratelimit-limit-requests": "15000",4756 "x-ratelimit-limit-tokens": "40000000",4757 "x-ratelimit-remaining-requests": "14999",4758 "x-ratelimit-remaining-tokens": "40000000",4759 "x-ratelimit-reset-requests": "4ms",4760 "x-ratelimit-reset-tokens": "0s"4761 },4762 "note": "observed for OUR key — account-specific, do not generalize"4763 },4764 {4765 "date": "2026-09-19",4766 "request": "POST /v1/responses model=gpt-5.6-terra",4767 "headers": {4768 "x-ratelimit-limit-requests": "15000",4769 "x-ratelimit-limit-tokens": "40000000",4770 "x-ratelimit-remaining-requests": "14999",4771 "x-ratelimit-remaining-tokens": "40000000",4772 "x-ratelimit-reset-requests": "4ms",4773 "x-ratelimit-reset-tokens": "0s"4774 },4775 "note": "observed for OUR key — account-specific, do not generalize"4776 },4777 {4778 "date": "2026-09-19",4779 "request": "POST /v1/responses model=gpt-6-astra",4780 "headers": {4781 "x-ratelimit-limit-requests": "15000",4782 "x-ratelimit-limit-tokens": "40000000",4783 "x-ratelimit-remaining-requests": "14999",4784 "x-ratelimit-remaining-tokens": "40000000",4785 "x-ratelimit-reset-requests": "4ms",4786 "x-ratelimit-reset-tokens": "0s"4787 },4788 "note": "observed for OUR key — account-specific, do not generalize"4789 },4790 {4791 "date": "2026-09-19",4792 "request": "POST /v1/responses model=gpt-5.5",4793 "headers": {4794 "x-ratelimit-limit-requests": "15000",4795 "x-ratelimit-limit-tokens": "40000000",4796 "x-ratelimit-remaining-requests": "14999",4797 "x-ratelimit-remaining-tokens": "40000000",4798 "x-ratelimit-reset-requests": "4ms",4799 "x-ratelimit-reset-tokens": "0s"4800 },4801 "note": "observed for OUR key — account-specific, do not generalize"4802 },4803 {4804 "date": "2026-09-19",4805 "request": "POST /v1/responses model=gpt-5.4-mini",4806 "headers": {4807 "x-ratelimit-limit-requests": "30000",4808 "x-ratelimit-limit-tokens": "180000000",4809 "x-ratelimit-remaining-requests": "29999",4810 "x-ratelimit-remaining-tokens": "180000000",4811 "x-ratelimit-reset-requests": "2ms",4812 "x-ratelimit-reset-tokens": "0s"4813 },4814 "note": "observed for OUR key — account-specific, do not generalize"4815 }4816 ],4817 "_fragment": "generated/fragments/rate-limits/openai-rate-limits.json"4818 },4819 {4820 "record_type": "rate_limits",4821 "provider": "xai",4822 "generated_at": "2026-09-18",4823 "status": [4824 "DOCUMENTED",4825 "LIVE_DISCOVERED"4826 ],4827 "sources": [4828 {4829 "url": "https://docs.x.ai/developers/rate-limits",4830 "retrieved_at": "2026-09-19"4831 },4832 {4833 "url": "https://docs.x.ai/developers/management-api-guide",4834 "retrieved_at": "2026-09-19"4835 },4836 {4837 "url": "https://docs.x.ai/developers/debugging",4838 "retrieved_at": "2026-09-19"4839 },4840 {4841 "url": "https://docs.x.ai/developers/pricing",4842 "retrieved_at": "2026-09-19"4843 },4844 {4845 "url": "https://docs.x.ai/developers/advanced-api-usage/batch-api",4846 "retrieved_at": "2026-09-19"4847 },4848 {4849 "url": "https://docs.x.ai/developers/models/grok-4.6",4850 "retrieved_at": "2026-09-19"4851 },4852 {4853 "url": "https://docs.x.ai/developers/models/grok-4.5",4854 "retrieved_at": "2026-09-19"4855 },4856 {4857 "url": "https://docs.x.ai/developers/models/grok-4.3",4858 "retrieved_at": "2026-09-19"4859 },4860 {4861 "url": "https://docs.x.ai/developers/models/grok-4.20-0309-reasoning",4862 "retrieved_at": "2026-09-19"4863 },4864 {4865 "url": "https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning",4866 "retrieved_at": "2026-09-19"4867 },4868 {4869 "url": "https://docs.x.ai/developers/models/grok-build-0.1",4870 "retrieved_at": "2026-09-19"4871 },4872 {4873 "url": "https://docs.x.ai/developers/models/grok-4.20-multi-agent-0309",4874 "retrieved_at": "2026-09-19"4875 },4876 {4877 "url": "https://docs.x.ai/developers/models/grok-imagine-image",4878 "retrieved_at": "2026-09-19"4879 },4880 {4881 "url": "https://docs.x.ai/developers/models/grok-imagine-image-quality",4882 "retrieved_at": "2026-09-19"4883 },4884 {4885 "url": "https://docs.x.ai/developers/models/grok-imagine-image-2.0",4886 "retrieved_at": "2026-09-19"4887 },4888 {4889 "url": "https://docs.x.ai/developers/models/grok-imagine-video",4890 "retrieved_at": "2026-09-19"4891 },4892 {4893 "url": "https://docs.x.ai/developers/models/grok-imagine-video-1.5",4894 "retrieved_at": "2026-09-19"4895 }4896 ],4897 "documented": {4898 "principles": [4899 "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.",4900 "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.",4901 "Tiers apply to text, embedding and voice models. Imagine (image/video) limits are RPS-only and are raised via sales@x.ai.",4902 "TPM counts prompt tokens (text, image, audio), completion tokens, reasoning tokens and cached prompt tokens (cached tokens still count although billed less).",4903 "Exceeding any limit -> HTTP 429 Too Many Requests (gRPC RESOURCE_EXHAUSTED). Use exponential backoff.",4904 "Batch API requests do not count towards rate limits.",4905 "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.",4906 "Team-level personalized limits are shown on the console Models page (https://console.x.ai/team/default/models)."4907 ],4908 "tiers": {4909 "Tier 0": {4910 "spend_threshold_usd": 0,4911 "note": "default"4912 },4913 "Tier 1": {4914 "spend_threshold_usd": 504915 },4916 "Tier 2": {4917 "spend_threshold_usd": 2504918 },4919 "Tier 3": {4920 "spend_threshold_usd": 10004921 },4922 "Tier 4": {4923 "spend_threshold_usd": 50004924 },4925 "Enterprise": {4926 "note": "on request (sales@x.ai)"4927 }4928 },4929 "per_model": {4930 "grok-4.6": {4931 "rps_T0_T4": [4932 150,4933 172,4934 208,4935 312,4936 5004937 ],4938 "tpm_T0_T4": [4939 "50M",4940 "53M",4941 "60M",4942 "74M",4943 "100M"4944 ]4945 },4946 "grok-4.5": {4947 "rps_T0_T4": [4948 150,4949 172,4950 208,4951 312,4952 5004953 ],4954 "tpm_T0_T4": [4955 "50M",4956 "53M",4957 "60M",4958 "74M",4959 "100M"4960 ]4961 },4962 "grok-4.3": {4963 "rps_T0_T4": [4964 37,4965 50,4966 75,4967 125,4968 2084969 ],4970 "tpm_T0_T4": [4971 "10M",4972 "15M",4973 "25M",4974 "45M",4975 "85M"4976 ]4977 },4978 "grok-4.20-0309-reasoning": {4979 "rps_T0_T4": [4980 37,4981 50,4982 75,4983 125,4984 2084985 ],4986 "tpm_T0_T4": [4987 "10M",4988 "15M",4989 "25M",4990 "45M",4991 "85M"4992 ]4993 },4994 "grok-4.20-0309-non-reasoning": {4995 "rps_T0_T4": [4996 37,4997 50,4998 75,4999 125,5000 2085001 ],5002 "tpm_T0_T4": [5003 "10M",5004 "15M",5005 "25M",5006 "45M",5007 "85M"5008 ]5009 },5010 "grok-build-0.1": {5011 "rps_T0_T4": [5012 37,5013 50,5014 75,5015 125,5016 2085017 ],5018 "tpm_T0_T4": [5019 "10M",5020 "15M",5021 "25M",5022 "45M",5023 "85M"5024 ]5025 },5026 "grok-4.20-multi-agent-0309": {5027 "rps_T0_T4": [5028 9,5029 12,5030 18,5031 31,5032 565033 ],5034 "tpm_T0_T4": [5035 "2.5M",5036 "3.7M",5037 "6.2M",5038 "11M",5039 "21M"5040 ]5041 },5042 "grok-imagine-image": {5043 "rps_T0_T4": [5044 6,5045 12,5046 25,5047 50,5048 1005049 ],5050 "tpm_T0_T4": null5051 },5052 "grok-imagine-image-quality": {5053 "rps_T0_T4": [5054 6,5055 12,5056 25,5057 50,5058 1005059 ],5060 "tpm_T0_T4": null5061 },5062 "grok-imagine-image-2.0": {5063 "rps_T0_T4": [5064 6,5065 12,5066 25,5067 50,5068 1005069 ],5070 "tpm_T0_T4": null5071 },5072 "grok-imagine-video": {5073 "rps_T0_T4": [5074 10,5075 20,5076 39,5077 79,5078 1585079 ],5080 "tpm_T0_T4": null5081 },5082 "grok-imagine-video-1.5": {5083 "rps_T0_T4": [5084 10,5085 20,5086 39,5087 79,5088 1585089 ],5090 "tpm_T0_T4": null5091 }5092 },5093 "voice_and_audio": {5094 "grok-voice-think-fast-2.0": {5095 "rps": null,5096 "concurrent_sessions_T0_T4": [5097 10,5098 20,5099 50,5100 100,5101 2005102 ],5103 "max_session_minutes": 1205104 },5105 "text-to-speech": {5106 "rps_T0_T4": [5107 50,5108 50,5109 100,5110 250,5111 5005112 ],5113 "concurrent_sessions_T0_T4": [5114 100,5115 200,5116 200,5117 300,5118 5005119 ]5120 },5121 "speech-to-text": {5122 "rps_T0_T4": [5123 10,5124 10,5125 20,5126 30,5127 405128 ],5129 "concurrent_sessions_T0_T4": [5130 100,5131 200,5132 200,5133 300,5134 5005135 ],5136 "note": "concurrent sessions apply to streaming"5137 }5138 },5139 "increasing_limits": [5140 "spend more (automatic)",5141 "request an increase in the console Models page",5142 "sales@x.ai for enterprise capacity"5143 ],5144 "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)"5145 },5146 "observed_for_this_key": {5147 "note": "Headers returned to OUR key on 2026-09-19 (Tier 0 team, spend < $50). Account-specific; not documentation.",5148 "headers_seen": [5149 "x-ratelimit-limit-requests",5150 "x-ratelimit-remaining-requests",5151 "x-ratelimit-limit-tokens",5152 "x-ratelimit-remaining-tokens",5153 "x-request-id",5154 "Server-Timing (cfEdge/cfOrigin)",5155 "CF-RAY (Cloudflare edge, e.g. ...-YUL)"5156 ],5157 "headers_not_seen": [5158 "x-ratelimit-reset-*",5159 "retry-after (no 429 triggered)"5160 ],5161 "per_model": {5162 "grok-4.6": {5163 "x-ratelimit-limit-requests": "7200",5164 "x-ratelimit-limit-tokens": "50000000"5165 },5166 "grok-4.5": {5167 "x-ratelimit-limit-requests": "7200",5168 "x-ratelimit-limit-tokens": "50000000"5169 },5170 "grok-4.3": {5171 "x-ratelimit-limit-requests": "1800",5172 "x-ratelimit-limit-tokens": "10000000"5173 },5174 "grok-4.20-0309-non-reasoning": {5175 "x-ratelimit-limit-requests": "1800",5176 "x-ratelimit-limit-tokens": "10000000"5177 },5178 "grok-build-0.1": {5179 "x-ratelimit-limit-requests": "1800",5180 "x-ratelimit-limit-tokens": "10000000"5181 }5182 },5183 "tokenize_text": {5184 "x-ratelimit-limit-requests": "1800",5185 "x-ratelimit-limit-tokens": "absent"5186 },5187 "list_endpoints": "GET /v1/models, /v1/language-models, /v1/api-key, /v1/me return no x-ratelimit-* headers",5188 "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.",5189 "grok-4.20-multi-agent-0309": "POST /v1/responses returned no x-ratelimit-* headers in our probe"5190 },5191 "_fragment": "generated/fragments/rate-limits/xai-rate-limits.json"5192 }5193]