Model APIs - 21 entries
OpenAI API rate limits
OpenAI's official docs, read on October 17, 2026, set rate limits for the OpenAI API across Build, Launch and Grow. For Astra, Sol, Terra (Standard rate limits, guide table), Build is listed at 5,000 RPM and 1,000,000 TPM, and Launch at 10,000 RPM and 4,000,000 TPM.
Last verified from developers.openai.com. What the docs leave out
- Tiers
- 3
- Tables
- 14
- Entries
- 21
- Verified
- Oct 17, 2026
Limits by model, endpoint and plan
Values are copied exactly as the docs show them, units included, in the docs' own order. A dash means the docs list no value for that row.
All 14 tables on this page
- Organization monthly usage limit (all models)
- Astra, Sol, Terra (Standard rate limits, guide table)
- Luna (Standard rate limits, guide table)
- gpt-6-astra
- gpt-6-astra (service_tier: ultrafast)
- gpt-6.1-sol
- gpt-6-luna, gpt-5.6-terra and gpt-5.6-cyber
- gpt-image-2.5-sunburst and gpt-image-2.5-flare
- gpt-realtime-2.1 and gpt-realtime-2.1-mini
- gpt-realtime-translate
- gpt-transcribe
- text-embedding-3-small
- Fast mode (service_tier: "fast" or "priority")
- Ramp-rate limit, vector store ingestion, rate-limit headers
Organization monthly usage limit (all models)
Free / Build / Launch / Grow
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free: qualification | User must be in an allowed geography |
| Free: usage limit | $100 / month |
| Build: qualification | $5 in total credit purchases |
| Build: usage limit | $500 / month |
| Launch: qualification | $100 in total credit purchases |
| Launch: usage limit | $5,000 / month |
| Grow: qualification | $500 in total credit purchases |
| Grow: usage limit | $200,000 / month |
- Monthly usage limits are separate from configurable spend limits (spend alert, or hard spend limit where "Affected API requests return a 429 error"). Rate limits are defined at the organization level and at the project level, not user level. Metrics used: RPM, RPD, TPM, TPD, IPM, and audio minutes per minute.
Astra, Sol, Terra (Standard rate limits, guide table)
2 tiers
| Tier | RPM |
|---|---|
| Build | 5,000 |
| |
| Launch | 10,000 |
| |
- Build: Guide table groups Astra, Sol, Terra. Standard rate limits differ from Ultrafast rate limits. Long context models like GPT-5.5 have a separate long-context limit shown only in the developer console.
Luna (Standard rate limits, guide table)
3 tiers
| Tier | RPM |
|---|---|
| Build | 5,000 |
| |
| Launch | 10,000 |
| |
| Grow | 30,000 |
| |
gpt-6-astra
Free / Build / Launch / GrowModel page: GPT-6 Astra
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free | Not supported |
| Build RPM | 5,000 |
| Build TPM | 1,000,000 |
| Build Batch queue limit | 3,000,000 |
| Launch RPM | 10,000 |
| Launch TPM | 4,000,000 |
| Launch Batch queue limit | 200,000,000 |
| Grow RPM | 15,000 |
| Grow TPM | 40,000,000 |
| Grow Batch queue limit | 15,000,000,000 |
- 1,050,000 context window; 128,000 max output tokens. Batch queue limits count total input tokens queued for a model.
gpt-6-astra (service_tier: ultrafast)
Build / Launch / GrowModel page: GPT-6 Astra
| Metric | Build / Launch / Grow |
|---|---|
| Build TPM | 500,000 |
| Launch TPM | 1,000,000 |
| Grow TPM | 5,000,000 |
- "Ultrafast has separate rate limits from Standard and Fast modes." Default Ultrafast token rate limits. Available to all API users.
gpt-6.1-sol
Free / Build / Launch / GrowModel page: GPT-6.1 Sol
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free | Not supported |
| Build RPM | 5,000 |
| Build TPM | 1,000,000 |
| Launch RPM | 10,000 |
| Launch TPM | 4,000,000 |
| Grow RPM | 15,000 |
| Grow TPM | 40,000,000 |
- Batch queue limit column not captured for this model page.
gpt-6-luna, gpt-5.6-terra and gpt-5.6-cyber
3 scopes - Free / Build / Launch / Grow
| Model or scope | Free |
|---|---|
| gpt-6-luna | Not supported |
| |
| gpt-5.6-terra | Not supported |
| |
| gpt-5.6-cyber | Not supported |
| |
- gpt-5.6-cyber: Model is for authorized vulnerability research and security testing.
gpt-image-2.5-sunburst and gpt-image-2.5-flare
2 scopes - Free / Build / Launch / Grow
| Model or scope | Free |
|---|---|
| gpt-image-2.5-sunburst | Not supported |
| |
| gpt-image-2.5-flare | Not supported |
| |
gpt-realtime-2.1 and gpt-realtime-2.1-mini
2 scopes - Free / Build / Launch / Grow
| Model or scope | Free |
|---|---|
| gpt-realtime-2.1 | Not supported |
| |
| gpt-realtime-2.1-mini | Not supported |
| |
gpt-realtime-translate
Free / Build / Launch / Grow
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free | Not supported |
| Build minutes-of-audio per minute | 200 |
| Launch minutes-of-audio per minute | 650 |
| Grow minutes-of-audio per minute | 850 |
gpt-transcribe
Free / Build / Launch / Grow
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free | Not supported |
| Build RPM | 5,000 |
| Build TPM | 2,000,000 |
| Launch RPM | 10,000 |
| Launch TPM | 10,000,000 |
| Grow RPM | 30,000 |
| Grow TPM | 150,000,000 |
text-embedding-3-small
Free / Build / Launch / Grow
| Metric | Free / Build / Launch / Grow |
|---|---|
| Free RPM | 100 |
| Free RPD | 2,000 |
| Free TPM | 40,000 |
| Free Batch queue limit | - |
| Build RPM | 5,000 |
| Build TPM | 1,000,000 |
| Build Batch queue limit | 20,000,000 |
| Launch RPM | 10,000 |
| Launch TPM | 5,000,000 |
| Launch Batch queue limit | 500,000,000 |
| Grow RPM | 10,000 |
| Grow TPM | 10,000,000 |
| Grow Batch queue limit | 4,000,000,000 |
- One of the few models with published Free tier limits. RPD listed as "-" for paid tiers.
Fast mode (service_tier: "fast" or "priority")
| Metric | Limit |
|---|---|
| Rate limit sharing | "For a given model, Standard processing and Fast mode share the same rate limit." |
- "Fast mode consumption counts toward rate limits the same way as Standard processing." If traffic ramps too fast, some Fast mode requests may be downgraded to standard speeds and charged standard rates; the response then contains service_tier: "default". Priority processing was renamed Fast mode on July 30, 2026.
Ramp-rate limit, vector store ingestion, rate-limit headers
| Metric | Limit |
|---|---|
| Ramp-rate rule of thumb | once your traffic reaches 1 million input tokens per minute (TPM), increase it by no more than 50% every 15 minutes |
| Vector store ingestion | 300 requests per minute for each vector store (/vector_stores/{vector_store_id}/files and /vector_stores/{vector_store_id}/file_batches share the limit) |
| Rapid ramp error | 429 rate_limit_error slow_down |
| Model overload error | 503 service_unavailable_error server_is_overloaded |
- "A slow_down error can occur even when your traffic is within its requests-per-minute and tokens-per-minute limits." Headers: Retry-After, x-ratelimit-limit-requests, x-ratelimit-limit-tokens, x-ratelimit-remaining-requests, x-ratelimit-remaining-tokens, x-ratelimit-reset-requests, x-ratelimit-reset-tokens, x-ratelimit-limit-project-tokens, x-ratelimit-remaining-project-tokens, x-ratelimit-reset-project-tokens. Fine-tuning limits via: curl https://api.openai.com/v1/fine_tuning/model_limits -H "Authorization: Bearer $OPENAI_API_KEY"
How to raise your limits
Organization monthly usage limit (all models), Astra, Sol, Terra (Standard rate limits, guide table) and Luna (Standard rate limits, guide table)
"The three paid usage tiers are Build, Launch, and Grow. Your organization's usage tier upgrades automatically as its total credit purchases reach each threshold. Higher tiers generally provide higher rate limits across models." To upgrade: "select Upgrade tier in the Usage Tiers section" of Settings > Organization > Limits.
gpt-6-astra, gpt-6.1-sol and 10 more
"Your usage tier determines how high these limits are set and automatically increases as you send more requests and spend more on the API."
gpt-6-astra (service_tier: ultrafast)
"If your organization works with an OpenAI account team, contact them to request higher rate limits."
Ramp-rate limit, vector store ingestion, rate-limit headers
"Enterprise customers whose pay-as-you-go traffic routinely hits ramp-rate limits can consider Scale Tier for more predictable capacity on eligible models. For GPT-5.6 and later models, see Reserved Tier."
What the docs leave out
- Long-context rate limits (for example GPT-5.5): guide says they are only viewable in the developer console at platform.openai.com/settings/organization/limits (login required)
- Shared-limit model family groupings and shared TPM values: only listed on the logged-in organization limits page (guide example: 'if the listed shared TPM is 3.5M')
- Per-organization approved monthly usage limit beyond the tier table and fine-tuning rate limits: console or API only.
- TPD values per model: TPD is named as a metric but no public per-model TPD figures were found on the pages read.
- Gpt-6.1-sol Batch queue limit: not captured from its model page (only RPM/TPM table returned)
- Free tier limits for flagship text, image, realtime, transcription and TTS models: model pages list 'Not supported'; the guide still lists a Free tier with a $100 / month usage limit.
- Fast mode separate per-tier numbers: none published (shares Standard limits)
- Scale Tier and Reserved Tier capacity figures: not published on the rate-limits guide.
- Model pages not read: gpt-realtime-2, gpt-realtime-1.5, gpt-realtime-mini, gpt-live-transcribe, gpt-realtime-whisper, gpt-4o-transcribe, gpt-4o-mini-transcribe, other embedding and moderation models, video models.
FAQ
What are the OpenAI API rate limits?
OpenAI's official docs, read on October 17, 2026, set rate limits for the OpenAI API across Build, Launch and Grow. For Astra, Sol, Terra (Standard rate limits, guide table), Build is listed at 5,000 RPM and 1,000,000 TPM, and Launch at 10,000 RPM and 4,000,000 TPM.
How do I raise my OpenAI API rate limits?
"The three paid usage tiers are Build, Launch, and Grow. Your organization's usage tier upgrades automatically as its total credit purchases reach each threshold. Higher tiers generally provide higher rate limits across models." To upgrade: "select Upgrade tier in the Usage Tiers section" of Settings > Organization > Limits. "Your usage tier determines how high these limits are set and automatically increases as you send more requests and spend more on the API." "If your organization works with an OpenAI account team, contact them to request higher rate limits."
Where do these OpenAI API numbers come from?
Every value is copied verbatim from OpenAI's official documentation at developers.openai.com, read on October 17, 2026. This page lists all 21 entries in 14 tables, each linked to the page it came from.
Related reading
- Codex Usage Limits and Pricing in 2026: Pro 100, Pro 200, Pro 500, Resets, and 'Model at Capacity' ErrorsCodex usage limits and pricing per ChatGPT plan: what Plus, Pro $200 and Pro $500 buy, how resets work, and what to do about 'Selected model is at capacity'.
- OpenAI Cuts GPT-5.6 Luna 80% and Terra 20%: The Cost-Per-Task Math for Agent BuildersGPT-5.6 Luna drops 80% to $0.20/$1.20 per million tokens and Terra 20% to $2/$12. The cost-per-task math for agent builders, Codex quotas and rivals.
- Codex CLI 0.163: Instant Steering Is Now the DefaultCodex CLI 0.163.0 makes mid-response steering the default and adds searchable /config and a leader key. What changed and what to check before upgrading.
- Tool Calling Under the Hood: Claude Messages vs OpenAI ResponsesHow tool calling works on the wire: JSON schema tools, parallel calls, tool_choice, strict mode and streaming deltas in Claude Messages vs OpenAI Responses.
- Codex 'gpt-5.5 Model Is Not Supported' Error: The FixCodex says gpt-5.5 is not supported with a ChatGPT account because OpenAI retired it from ChatGPT sign-in on October 14. The fix and what still works.
- ChatGPT Free vs Go vs Plus After GPT-6: Which Model You GetFree and Go ($8) both get GPT-6 Luna in Chat. Plus ($20) is the cheapest plan with GPT-6 Sol and full Codex. What each ChatGPT plan buys now.
Sources
Last verified . Every value is copied verbatim from the official pages below, read on that date.
- https://developers.openai.com/api/docs/guides/rate-limits
- https://developers.openai.com/api/docs/models/gpt-6-astra
- https://developers.openai.com/api/docs/guides/ultrafast-mode
- https://developers.openai.com/api/docs/models/gpt-6.1-sol
- https://developers.openai.com/api/docs/models/gpt-6-luna
- https://developers.openai.com/api/docs/models/gpt-5.6-terra
- https://developers.openai.com/api/docs/models/gpt-5.6-cyber
- https://developers.openai.com/api/docs/models/gpt-image-2.5-sunburst
- https://developers.openai.com/api/docs/models/gpt-image-2.5-flare
- https://developers.openai.com/api/docs/models/gpt-realtime-2.1
- https://developers.openai.com/api/docs/models/gpt-realtime-2.1-mini
- https://developers.openai.com/api/docs/models/gpt-realtime-translate
- https://developers.openai.com/api/docs/models/gpt-transcribe
- https://developers.openai.com/api/docs/models/text-embedding-3-small
- https://developers.openai.com/api/docs/guides/fast-mode
Newsletter
Know when limits change
One email per week. New model releases, pricing and limit changes, and what they mean for what you build.