Updated September 12, 2026
API Rate Limits Compared: Every Major LLM Provider (September 2026)
API rate limits for every major LLM provider — September 10, 2026. Side-by-side tables for OpenAI, Anthropic, Google, Groq, xAI, DeepSeek, Mistral, Cerebras, SambaNova, and more.
Tenth edition. The ninth (August 10, 2026) covered 17 providers. This update reflects the four-week window since August 10. Google’s current flagship is now Gemini 3.7 Flash (GA, August 2026), replacing Gemini 3.6 Flash. Anthropic confirmed Sonnet 5 stays at $2/$10 per M — the increase to $3/$15 scheduled for September 1 was cancelled. The OpenAI Assistants API deprecation window (August 2026) has passed. GPT-5.6 Sol rate limit tier tables remain unpublished; GPT-5.5 figures are carried forward as a provisional floor.
Table of Contents
- What changed since August 10
- Free Tier — RPM & RPD
- Free Tier — TPM & TPD
- Entry Paid Tier — RPM & RPD
- Entry Paid Tier — TPM
- Scaled Tier — RPM & TPM
- Cerebras & SambaNova
- Audio Models — ASH & ASD
- Cloud Aggregators — Azure AI & AWS Bedrock
- More Providers — Perplexity, Alibaba, Moonshot
- Provider Notes
- Tips for managing rate limits
- Further Reading
Last updated: September 10, 2026.
What changed since August 10
Google’s current flagship is now Gemini 3.7 Flash. GA as of August 2026. Model ID gemini-3.7-flash; 1M context, 64K max output, tunable thinking levels. Built for complex coding, agentic workflows, and reliable multi-step execution. Now the default for Antigravity. Introductory pricing runs through December 31, 2026 at $0.75/$3.75 per M input/output, rising to $1.50/$7.50 on January 1, 2027. Gemini 3.6 Flash is the previous flagship, still GA on the same introductory pricing. Rate limit figures for Gemini 3.7 Flash are not yet published in public documentation — check AI Studio (aistudio.google.com/rate-limit) for your project’s actual limits. Gemini 3.6 Flash rate limit figures also remain unpublished. Gemini 3.1 Flash-Lite and 3.1 Pro remain the last confirmed published numbers and are carried forward in tables.
Anthropic Sonnet 5 stays at $2/$10 per M. Anthropic confirmed on August 10 that the September 1 increase to $3/$15 would not happen and that $2/$10 is now the permanent standard rate. Cost models tied to $2/$10 remain correct. Rate limits were never affected either way.
OpenAI Assistants API deprecation window has passed. The Assistants API deprecated August 2026. The Responses API is the replacement. Any pipeline still on the Assistants API is running on deprecated infrastructure. Migrate to the Responses API.
GPT-5.6 Sol rate limit tier tables remain unpublished. No change from the August 10 edition. GPT-5.5 figures carried forward as a provisional floor in all tables. GPT-5.6 Sol pricing: Sol $5/$30, Terra $2/$12, Luna $0.20/$1.20 per M input/output. These are the verified figures — do not derive per-token costs from OpenAI’s “up to 80% cut” marketing language, which describes movement across the Sol/Terra/Luna tier spectrum, not a cut to any single variant.
Claude Opus 5 rate limit tier tables remain unpublished. No change. Claude Opus 4.8 figures carried forward as a provisional floor. Opus 5 pricing: $5/$25 per M input/output.
Claude Fable 5 rate limit tier tables remain unpublished. No change. Confirmed as a permanent production tier since July 18; figures still absent from Anthropic’s public rate limit documentation.
Gemini 3.7 Flash introductory pricing window: note the expiry. Introductory pricing ($0.75/$3.75) runs through December 31, 2026. Gemini 3.6 Flash carries the same introductory pricing on the same schedule. Cost models that carry these rates forward past January 1, 2027 will be wrong — the standard rate is $1.50/$7.50 for both.
Qwen3.8-Max rate limit figures remain unpublished. No change from August 10. Announced August 3; 2.4T parameters, 1M context. Qwen3.7-Max figures carried forward as a provisional floor. Qwen3.8-Max pricing is not published — do not use any per-token figure for it.
Kimi K3 rate limit figures remain unconfirmed. No change. K2.5 figures carried forward as a provisional floor.
Still pending formal documentation:
- OpenAI: GPT-5.6 Sol rate limit tier tables not yet published. GPT-5.5 figures carried forward as a provisional floor.
- Anthropic: Claude Opus 5 rate limit figures not yet published. Claude Fable 5 rate limit figures not yet published.
- Google: Gemini 3.7 Flash rate limit figures not yet published. Gemini 3.6 Flash, 3.5 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber rate limit figures also unpublished. Gemini 3.5 Pro remains in limited preview.
- Alibaba: Qwen3.8-Max rate limit figures not yet published. Qwen3.7-Max limits carried forward as a provisional floor.
- Moonshot: Kimi K3 rate limit figures not yet confirmed. K2.5 figures carried forward.
Deprecation calendar:
- OpenAI Assistants API deprecated August 2026 — passed. Migrate to the Responses API now if you haven’t.
- DeepSeek V3.2 aliases deprecated July 24, 2026 — passed. V4-Flash and V4-Pro are the current endpoints.
- Anthropic Sonnet 5’s scheduled September 1, 2026 increase to $3/$15 was cancelled — $2/$10 per M is now the permanent standard rate.
- Gemini 3.7 Flash and 3.6 Flash introductory pricing ($0.75/$3.75) expires December 31, 2026 — standard $1.50/$7.50 from January 1, 2027.
What are rate limits?
Rate limits cap how much you can use an API within a given time window. Providers enforce them to manage capacity and ensure fair access. Exceeding a limit returns a 429 Too Many Requests error.
Most providers use the token bucket algorithm: capacity refills continuously up to your maximum, rather than resetting at fixed intervals. A 60 RPM limit means roughly 1 request per second with steady refill, not 60 requests then a hard stop.
The metrics
- RPM (Requests per minute): API calls per minute, regardless of size. A 10-token request and a 100K-token request both count as one.
- RPD (Requests per day): Daily cap on total API calls. Some providers use this instead of (or alongside) RPM, especially on free tiers.
- TPM (Tokens per minute): Total tokens (input + output) processed per minute. Usually the binding constraint in production.
- TPD (Tokens per day): Daily token cap. More common on free tiers.
- ITPM/OTPM (Input/Output tokens per minute): Anthropic separates input and output token limits, giving finer control. Cached input tokens don’t count toward ITPM on current Claude 4.x and 5.x models.
- ASH/ASD (Audio seconds per hour/day): For speech models like Whisper.
RPM limits set your concurrency ceiling. TPM limits set your throughput. For batch workloads, RPD and TPD matter more. For real-time apps, RPM and TPM hit first.
Free Tier — Requests per Minute (RPM) & Requests per Day (RPD)
| Provider / Model | RPM | RPD |
|---|---|---|
| Google — Gemini 3.1 Flash-Lite | 15 | 1,000 |
| Groq — Llama 4 Scout | 30 | 1,000 |
| Cerebras — Llama 4 Scout | 30 | — |
| SambaNova — open models | 10–30 | — |
| Fireworks AI — OS models | 10 | — |
| Cohere — Command R+* | 20 | ~33† |
*Cohere’s free tier figures reference Command R+. Command A+ free tier limits not yet published. †Cohere trial keys get 1,000 calls/month across all endpoints, roughly ~33/day.
Note: Gemini 3.7 Flash, 3.6 Flash, 3.5 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber free tier rate limit figures are not yet published in public documentation. Gemini 3.1 Flash-Lite figures shown above are the last confirmed published numbers. OpenAI, Anthropic, xAI, DeepSeek, and Mistral have no free API tier for current flagship models. OpenAI’s free tier does not support any GPT-5.x model via the API. Anthropic’s lowest entry is $5 (Tier 1). xAI and Mistral gate numerical limits behind their respective consoles.
Migration note: Gemini 2.0 Flash and Flash-Lite shut down June 1, 2026. DeepSeek V3.2 aliases deprecated July 24, 2026. Both sets of endpoints are gone.
Free Tier — Tokens per Minute (TPM) & Tokens per Day (TPD)
| Provider / Model | TPM | TPD |
|---|---|---|
| Google — Gemini 3.1 Flash-Lite | 250,000 | — |
| Groq — Llama 4 Scout | 30,000 | 500,000 |
| Groq — other open models | 6,000–30,000 | 500,000 |
| Cerebras — Llama 4 Scout | 60,000 | 1,000,000 |
| SambaNova (free tier) | — | — |
Standout: Google leads with 250K TPM on the free tier for Gemini 3.1 Flash-Lite, but the 10–15 RPM ceiling means it’s best suited to fewer, larger requests. Gemini 3.7 Flash, 3.6 Flash, 3.5 Flash, and 3.5 Flash-Lite free tier TPM figures are not yet published. Cerebras offers 60K TPM with 1M TPD and approximately 2,100 tokens/second inference speed on Llama 4 Scout — the fastest free option by throughput. SambaNova publishes RPM but not TPM/TPD specifics.
Entry Paid Tier — Requests per Minute (RPM) & Requests per Day (RPD)
The tier most developers start with. OpenAI Tier 1 ($5), Anthropic Tier 1 ($5), Google Tier 1 (pay-as-you-go), and others at their lowest paid level.
| Provider / Model | RPM | RPD |
|---|---|---|
| OpenAI — GPT-5.6 Sol* | 500* | — |
| OpenAI — GPT-4.1 Nano | 500 | — |
| Anthropic — Opus 5† | 50† | — |
| Anthropic — Sonnet 5 | 50 | — |
| Anthropic — Haiku 4.5 | 50 | — |
| Google — Gemini 3.1 Pro | 150 | 1,000 |
| Google — Gemini 3.1 Flash-Lite | 300 | 1,500 |
| xAI — Grok 4.5 | Console‡ | — |
| DeepSeek — V4-Flash | Dynamic§ | — |
| DeepSeek — V4-Flash 0731 | Dynamic§ | — |
| Fireworks AI — OS models | ≤6,000 | — |
| Cohere — Command A+** | 500 | — |
*GPT-5.6 Sol rate limit tier tables not yet published. 500 RPM carried forward from GPT-5.5 as a provisional floor. †Opus 5 rate limit tier tables not yet published. 50 RPM carried forward from Opus 4.8 as a provisional floor. ‡xAI publishes tier thresholds ($0/$50/$250/$1K/$5K) but numerical RPM/TPM are only visible in the xAI Console after login. §DeepSeek uses fully dynamic concurrency limits based on server load. No fixed RPM/TPM published. **Command A+ entry paid tier limits not yet published. 500 RPM is the Command R+ figure, carried forward as a reference.
Gemini 3.7 Flash note: Gemini 3.7 Flash is Google’s current flagship but has no published rate limit figures in public documentation. Gemini 3.6 Flash (previous flagship) and Gemini 3.5 Flash (GA) also have no published rate limit figures. Gemini 3.1 Pro and 3.1 Flash-Lite remain available as the previous generation, with the last confirmed published figures shown above. Check AI Studio for your project’s actual limits.
Mistral note: Mistral moved to console-only limits. Actual numbers require login at admin.mistral.ai/plateforme/limits. Mistral is excluded from RPM columns where no public figure exists.
Standout: Fireworks can spike to 6,000 RPM but it’s a dynamic ceiling, not guaranteed (soft limit starts at approximately 1 RPS and doubles hourly). Anthropic’s 50 RPM at Tier 1 is the lowest here, but jumps 20x to 1,000 RPM at Tier 2 ($40). Sonnet 5’s tokenizer emits approximately 30% more tokens per request than Sonnet 4.6 for identical prompts — TPM headroom shrinks faster than it did on 4.6 at the same RPM. Sonnet 5 is $2/$10 per M permanently; the September 1 increase was cancelled.
Entry Paid Tier — Tokens per Minute (TPM)
| Provider / Model | TPM |
|---|---|
| OpenAI — GPT-5.6 Sol* | 500K* |
| OpenAI — GPT-4.1 Nano | 200K |
| Anthropic — Opus 5† | 30K in / 8K out† |
| Anthropic — Sonnet 5 | 30K in / 8K out |
| Anthropic — Haiku 4.5 | 50K in / 10K out |
| Google — Gemini 3.1 Pro | 1M |
| Google — Gemini 3.1 Flash-Lite | 2M |
| DeepSeek — V4-Flash | Dynamic |
*GPT-5.6 Sol TPM figures not yet published. 500K carried forward from GPT-5.5 as a provisional floor. †Opus 5 TPM figures not yet published. Opus 4.8 figures carried forward as a provisional floor.
Standout: Google’s 3.1 Flash-Lite leads confirmed figures at 2M TPM. Anthropic’s Sonnet 5 shows 30K ITPM — the tokenizer change means identical prompts now consume more tokens against that limit than they did on Sonnet 4.6. Cached input tokens don’t count toward ITPM on Claude 4.x and 5.x models — with an 80% cache hit rate, effective throughput remains substantially higher than the raw ITPM figure.
Scaled Tier — RPM & TPM at Higher Spend
For teams past entry-level. OpenAI Tier 3 ($100+), Anthropic Tier 3 ($200+), Google Tier 2 ($250+).
| Provider (Tier) / Model | RPM | TPM |
|---|---|---|
| OpenAI (Tier 3) — GPT-5.6 Sol* | 5,000* | 2M* |
| OpenAI (Tier 3) — GPT-4.1 Nano | 5,000 | 4M |
| Anthropic (Tier 3) — Opus 5† | 2,000† | 800K in / 160K out† |
| Anthropic (Tier 3) — Sonnet 5 | 2,000 | 800K in / 160K out |
| Anthropic (Tier 3) — Haiku 4.5 | 2,000 | 1M in / 200K out |
| Google (Tier 2) — Gemini 3.1 Pro | 1,000 | 2M |
| Google (Tier 2) — Gemini 3.1 Flash-Lite | 2,000 | 4M |
*GPT-5.6 Sol Tier 3 figures not yet published. GPT-5.5 figures carried forward as a provisional floor. †Opus 5 Tier 3 figures not yet published. Opus 4.8 figures carried forward as a provisional floor.
At the highest standard tiers (OpenAI Tier 5 at $1,000+, Anthropic Tier 4 at $400+):
| Provider (Tier) / Model | RPM | TPM |
|---|---|---|
| OpenAI (Tier 5) — GPT-5.6 Sol* | 15,000* | 40M* |
| OpenAI (Tier 5) — GPT-4.1 Nano | 30,000 | 180M |
| Anthropic (Tier 4) — Opus 5† | 4,000† | 2M in / 400K out† |
| Anthropic (Tier 4) — Sonnet 5 | 4,000 | 2M in / 400K out |
| Anthropic (Tier 4) — Haiku 4.5 | 4,000 | 4M in / 800K out |
*GPT-5.6 Sol Tier 5 figures not yet published. GPT-5.5 figures carried forward as a provisional floor. †Opus 5 Tier 4 figures not yet published. Opus 4.8 figures carried forward as a provisional floor. Confirm current limits in the Anthropic console.
Standout: OpenAI’s Tier 5 numbers remain the highest of any provider with published or provisionally carried figures. GPT-4.1 Nano at 180M TPM and 30K RPM is built for high-volume classification and routing. Anthropic’s Tier 4 caps at 4K RPM, but cached tokens don’t count toward ITPM — effective throughput can be substantially higher with good cache hit rates.
Model pool notes:
- Opus 5 pool assignment is not yet formally documented. Opus 4.8, 4.7, and 4.6 shared one rate limit bucket; check your Anthropic dashboard for how Opus 5 is bucketed.
- Sonnet 5 and Sonnet 4.6 likely share a pool — not formally confirmed; check your dashboard.
- Sending traffic to multiple model versions within the same pool draws from one shared bucket, not separate ones.
Gemini 3.7 Flash scaled tier note: Gemini 3.7 Flash is Google’s current flagship with no published rate limit figures. Gemini 3.6 Flash (previous flagship) and Gemini 3.5 Flash (GA) are also without published rate limit figures. The 3.1 Flash-Lite and 3.1 Pro figures shown above remain the last confirmed published numbers.
Cerebras & SambaNova
Both specialize in custom silicon for inference speed.
Cerebras
Hardware: Wafer-Scale Engine (WSE-3). Fastest published inference: approximately 2,100 tokens/second on Llama 4 Scout.
| Tier | RPM | TPM | TPD |
|---|---|---|---|
| Free | 30 | 60,000 | 1,000,000 |
| Paid | Higher (contact sales) | Higher | Higher |
Hosts Llama 4 Scout and other open models. The speed advantage is real: tasks that take 30 seconds on GPU-based providers finish in under 5 seconds on the WSE-3. Paid tier limits are not publicly documented.
SambaNova
Hardware: Custom RDU (Reconfigurable Dataflow Unit). Best time-to-first-token (TTFT): approximately 0.2 seconds.
| Tier | RPM | Notes |
|---|---|---|
| Free | 10–30 (varies by model) | Hosts up to Llama 4 Maverick for free |
SambaNova offers Llama 4 Maverick (1M context) on the free tier — most other free-tier providers cap out at smaller models. TPM and TPD limits are not publicly documented.
Audio Models — ASH & ASD
Groq and Fireworks both publish audio-specific limits.
| Provider / Model | RPM | RPD | ASH | ASD | Audio min/min |
|---|---|---|---|---|---|
| Groq (Free) — Whisper Large v3 | 20 | 2,000 | 7,200 | 28,800 | — |
| Groq (Free) — Whisper Large v3 Turbo | 20 | 2,000 | 7,200 | 28,800 | — |
| Fireworks AI — Whisper v3-large | — | — | — | — | 200 |
| Fireworks AI — Whisper v3-turbo | — | — | — | — | 400 |
Groq: 7,200 ASH = 2 hours of audio per hour of wall time. 28,800 ASD = 8 hours per day. Adequate for a podcast transcription pipeline on the free tier.
Fireworks: 200 min/min for Whisper v3-large, 400 min/min for v3-turbo. Concurrent streaming capped at 10 connections.
OpenAI’s GPT Transcribe and GPT Live Transcribe are available via API. Independent benchmarks show both lag ElevenLabs, Google, and Mistral on error rates. OpenAI does not yet publish audio-specific ASH/ASD figures for these endpoints in its public rate limit documentation.
Cloud Aggregators — Azure AI & AWS Bedrock
Both use configurable per-deployment limits, not fixed tiers. The shared ratio is 6 RPM per 1,000 TPM.
Azure AI (Microsoft)
| Deployment Type | How Limits Work |
|---|---|
| Pay-as-you-go (Standard) | TPM quota per model per region, auto-scales with usage |
| Provisioned (PTU) | Reserved throughput units, no per-request limits |
| Global/Data Zone | Higher default quotas, multi-region routing |
GPT-5.6 Sol is the default in Microsoft 365 Copilot. Azure AI hosts GPT-5.6 Sol, GPT-5.5, GPT-5.4, Claude Opus 5/Sonnet 5, and Gemini 3.1 Pro across regions. Gemini 3.7 Flash and 3.6 Flash availability on Azure is not yet confirmed — verify model and region availability before planning deployments. PTU deployments remove per-request limits. Azure AI Quotas & Limits
AWS Bedrock
Hosts Claude (Opus 5, Opus 4.8, Sonnet 5, Haiku 4.5), Llama 4, Mistral, and Nova models with per-model, per-region quotas. Claude Fable 5 availability on Bedrock — check the Bedrock model catalog for current status following the July 18 permanent-tier confirmation. Amazon is scaling back most Nova AI models (Premier, Omni, Reel, Canvas into “keep the lights on” mode) while a new Frontier research team works on a foundation model expected at re:Invent. Default quotas vary by model and can be increased via AWS Service Quotas console. Provisioned Throughput deployments remove rate limits entirely. AWS Bedrock Quotas
More Providers — Perplexity, Alibaba (Qwen), Moonshot (Kimi)
| Provider / Model | RPM (T0) | RPM (T1) | RPM (T3+) | TPM | TPD (T0) |
|---|---|---|---|---|---|
| Perplexity — Sonar Pro | 50 | 150 | 1,000 | — | — |
| Perplexity — Sonar | 50 | 150 | 1,000 | — | — |
| Perplexity — Deep Research | 5 | 10 | 40 | — | — |
| Alibaba (Qwen) — Qwen3.8-Max* | — | 600* | 600* | 1M* | — |
| Alibaba (Qwen) — Qwen3.7-Max† | — | 600† | 600† | 1M† | — |
| Alibaba (Qwen) — Qwen3.5 Plus | — | 15,000 | 30,000 | 5M | — |
| Alibaba (Qwen) — Qwen3.5 Flash | — | 15,000 | 30,000 | 10M | — |
| Moonshot — Kimi K3‡ | 3‡ | 200‡ | 5,000‡ | — | 1.5M‡ |
| Moonshot — Kimi K2.5 | 3 | 200 | 5,000 | — | 1.5M |
*Qwen3.8-Max (August 3) rate limit figures not yet published. Qwen3.7-Max figures carried forward as a provisional floor. †Qwen3.7-Max (May 20) rate limit figures not yet formally published. Qwen3 Max limits carried forward as a provisional floor. ‡Kimi K3 rate limit figures not yet confirmed in Moonshot’s public documentation. K2.5 figures carried forward as a provisional floor.
Perplexity tiers: T0 = new account, T1 = $50+, T3 = $500+. Alibaba limits shown for international (Singapore) deployment; Beijing deployment is more generous (Qwen3.5 Flash at 30K RPM — Qwen3.8-Max and 3.7-Max Beijing limits not yet confirmed). Moonshot T1 = $10+ cumulative recharge; T0 has 1.5M TPD cap, T1+ is unlimited.
Standout: Alibaba’s Qwen3.5 Flash at 30K RPM / 10M TPM (Beijing) remains the highest confirmed throughput of any provider in this post. Moonshot’s T5 tier (10K RPM, 1,000 concurrent connections, $3,000+ recharge) is competitive at the high end. Kimi K3’s 2.8T parameters and Opus-class performance at Sonnet-level pricing make it a meaningful option for high-volume open-weight workloads — confirmed rate limit figures are still absent from Moonshot’s public documentation.
Provider Notes
- OpenAI: GPT-5.6 Sol is the current flagship, broadly available since July 9–11, 2026. Variants Sol ($5/$30 per M), Terra ($2/$12), and Luna ($0.20/$1.20). GPT-5.6 Sol rate limit tier tables not yet published in public docs — GPT-5.5 figures carried forward as a provisional floor in all tables above. GPT-5.5 remains available (variants: GPT-5.5 Thinking, GPT-5.5 Pro). GPT-5.5 Instant remains available. GPT-5.4 remains available (variants: GPT-5.4 Thinking, GPT-5.4 Pro, GPT-5.4-Cyber for vetted security teams). GPT-5.3 Codex consolidated into ChatGPT — no longer a standalone API endpoint. GPT-4.1 Nano ($0.10/$0.40 per M, $0.025 cached, 1M context) is the budget/high-throughput model. The GPT-4.1 family (4.1, 4.1 Mini, 4.1 Nano) remains active in the catalogue and was not retired. GPT-5.2 being phased out. GPT-4 and GPT-4o retired February 2026. GPT Transcribe and GPT Live Transcribe available via API; audio-specific rate limit figures not yet published. Assistants API deprecated August 2026 — migrate to the Responses API. Batch API offers 50% discount with higher queue limits. Docs
- Anthropic: Claude Opus 5 (July 2026) is the current flagship — scored 30.2% on ARC-AGI-3, nearly 4x the previous record; matches Fable 5-level performance on coding and knowledge tasks at 50% lower token cost; $5/$25 per M. Rate limit figures for Opus 5 not yet published; Opus 4.8 figures carried forward as a provisional floor. Opus 4.8 (May 2026) is the previous flagship, still available. Claude Sonnet 5 (June 30, 2026) is the current balanced tier — tokenizer emits approximately 30% more tokens per request than Sonnet 4.6 for identical text; $2/$10 per M permanently, the September 1 increase to $3/$15 having been cancelled. Claude Haiku 4.5 is the fast/budget tier at $1/$5 per M. Claude Fable 5 (confirmed permanent July 18) is a committed production tier; rate limit figures not yet published; $10/$50 per M. Claude Fable 5 available on the Claude API, Bedrock, Google Cloud, and Microsoft Foundry. Claude Mythos 5 (defensive-cybersecurity focus) accessible only to vetted US organizations via Project Glasswing — not generally available. Opus 5 pool assignment not yet formally documented; Opus 4.8, 4.7, and 4.6 shared a combined pool — check your Anthropic dashboard. Sonnet 5 and Sonnet 4.6 likely share a pool — not yet formally confirmed. Cached input tokens don’t count toward ITPM on current 4.x and 5.x models. No free tier; lowest entry is $5. Claude 3 Haiku retired April 2026. Docs
- Google: Static rate limit tables removed from public docs in Q1 2026. Actual limits only visible in AI Studio dashboard (
aistudio.google.com/rate-limit). Gemini 3.7 Flash (August 2026,gemini-3.7-flash) is the current flagship — 1M context, 64K max output, tunable thinking levels; default for Antigravity; introductory pricing $0.75/$3.75 per M through December 31, 2026, rising to $1.50/$7.50 on January 1, 2027. Rate limit figures not yet published. Gemini 3.6 Flash (July 2026, previous flagship) is still GA on the same introductory pricing; rate limit figures not yet published. Gemini 3.5 Flash (GA since May 19) approximately 4x faster output than other frontier models, leads on agentic and coding benchmarks; rate limit figures not yet published. Gemini 3.5 Flash-Lite and Gemini 3.5 Flash Cyber (cybersecurity domain) launched July 2026; rate limit figures not yet published. Gemini 3.5 Pro remains in limited preview (2M context, Deep Think). Gemini 3.1 Pro remains available as the previous-generation Pro model ($2/$12 per M ≤200K, $4/$18 >200K). Gemini Omni rate limit figures absent. Gemini 2.0 Flash/Flash-Lite shut down June 1, 2026. Google is planning Gemini 4. Docs - Groq: Free tier publishes exact numbers (30 RPM, 6K–30K TPM depending on model). Daily token budgets (TPD) alongside TPM. Cached tokens don’t count. Hosts Llama 4 Scout and other open models on free tier. Runs on custom LPU hardware. Docs
- xAI: Current flagship is Grok 4.5 (approximately July 9, 2026; $2/$6 per M; 500K context). Grok 4.3 (April 30, 2026; 1M context; $1.25/$2.50 per M) is the previous flagship, still available. grok-build-0.1 is the agentic coding model ($1/$2 per M). 5-tier structure ($0/$50/$250/$1K/$5K thresholds) but numerical RPM/TPM only visible in xAI Console. Docs
- Mistral: European data residency (GDPR). 5-tier structure (Free through Tier 4 at $500+). Limits enforced per RPS (not RPM), TPM, and tokens/month. Actual numbers require login at admin.mistral.ai. Mistral Large 3 (675B MoE, Apache 2.0, $0.50/$1.50 per M) and Mistral Small 4 (Apache 2.0, $0.15/$0.60 per M) are available. Mistral Medium 3.5 ($1.50/$7.50 per M) is the proprietary frontier tier. Devstral 2 (72.2% SWE-bench, $0.40/$2.00 per M) is available for coding workloads. Docs
- DeepSeek: Fully dynamic concurrency limits based on server load. No published RPM/TPM. V4-Flash 0731 is a fast/efficient variant that reached within one point of GPT-5.6 Luna on the Artificial Analysis Intelligence Index at approximately 60% lower cost per task. V4-Flash and V4-Pro are the current endpoints. V4-Flash pricing: $0.14/M input (cache miss), $0.0028/M (cache hit), $0.28/M output. V4-Pro pricing: $0.435/M input (cache miss), $0.003625/M (cache hit), $0.87/M output. V3.2 aliases deprecated July 24, 2026 — those endpoints are gone. Docs
- Cerebras: Fastest inference (approximately 2,100 tok/s on Llama 4 Scout). Free tier: 30 RPM, 60K TPM, 1M TPD. Paid tier limits not publicly documented. Custom WSE-3 silicon. Docs
- SambaNova: Best TTFT (approximately 0.2s). Free tier: 10–30 RPM depending on model size. Hosts up to Llama 4 Maverick for free. TPM and TPD limits not publicly documented. Custom RDU hardware. Docs
- Together AI: Dynamic rate limits since January 2026. No fixed tiers or published numbers. Limits grow with sustained usage and are returned in API response headers. Docs
- Fireworks AI: Dynamic ceiling up to 6,000 RPM (soft limit starts approximately 1 RPS, doubles hourly). On-demand GPU deployments remove limits. Spending tier caps: $50–$50K/month by tier. Docs
- Cohere: Command A+ (May 20, Apache 2.0, 218B MoE / 25B active; runs on 2x H100 or 1x B200; native citations; 256K context; 48 languages) is the flagship. Self-hosting is viable on 2x H100 or 1x B200 and removes API rate limits entirely. API rate limit figures for Command A+ not yet published — 500 RPM (production chat) shown in tables is the Command R+ figure. Trial keys: 1,000 calls/month. No per-token API price published for Command A+. Docs
- Perplexity: 6 tiers (T0–T5) based on cumulative spend. Leaky bucket algorithm. Deep Research model has very low limits (5–100 RPM). Agent API: 50–2,000 RPM by tier. Docs
- Alibaba (Qwen): Qwen3.8-Max (August 3, 2026; 2.4T parameters; 1M context; ranks 5th on Text Arena, 2nd on Vision Arena; closed-weight) is the current flagship — published rate limits and pricing not yet confirmed. Qwen3.7-Max (May 20; 1M context; $2.50/$7.50 per M) is the previous flagship. Open-weight family is Qwen3.6 (Apache 2.0 — e.g., Qwen3.6-35B-A3B, Qwen3.6-27B). Region-specific limits apply — Beijing deployment is more generous than Singapore/Global. Qwen3.5 Flash at 30K RPM / 10M TPM (Beijing) is the highest confirmed throughput of any provider listed. Docs
- Moonshot (Kimi): 6 tiers (T0–T5). T0 ($1 recharge): 3 RPM, 1 concurrent, 1.5M TPD. T5 ($3,000): 10K RPM, 1,000 concurrent, 5M TPM, unlimited TPD. Kimi K3 (2.8T parameters, 1M context, Opus-class performance at Sonnet-level pricing) is the current flagship — rate limit figures not yet confirmed in public documentation; K2.5 figures carried forward as a provisional floor. Automatic 75% caching discount applied without opt-in. Docs
- AWS Bedrock: Hosts Claude (Opus 5, Opus 4.8, Sonnet 5, Haiku 4.5), Llama 4, Mistral with per-model per-region quotas. Claude Fable 5 availability — check the Bedrock model catalog. Amazon scaling back most Nova AI models (Premier, Omni, Reel, Canvas) while a new Frontier team works on a foundation model expected at re:Invent. 6 RPM per 1K TPM ratio. Provisioned Throughput removes limits. Docs
- Azure AI: Hosts GPT-5.6 Sol (default in Microsoft 365 Copilot), GPT-5.5, GPT-5.4, Claude Opus 5/Sonnet 5, and Gemini 3.1 Pro across regions. Gemini 3.7 Flash and 3.6 Flash availability on Azure not yet confirmed — verify model and region availability before planning deployments. PTU deployments remove per-request limits. Docs
- NVIDIA NIM: Hosted API is for prototyping (40 RPM). Self-hosted NIM containers have no limits. Docs
Tips for managing rate limits
For a deep dive on token pricing, cost optimization strategies, caching mechanics, model routing architectures, and production cost modeling, see LLM Token Costs and Efficiency. This section covers rate limit management specifically.
Handling 429s
Use exponential backoff when you hit 429s. The Anthropic SDK, OpenAI SDK, and most third-party clients handle this automatically. Don’t build your own retry loop unless you need custom jitter or circuit-breaking behavior.
Check response headers before you hit the wall. Together AI and Fireworks return your current rate limit state in every API response. DeepSeek adjusts concurrency limits dynamically based on server load. Reading these headers in production enables proactive throttling rather than reactive retries.
GPT-5.6 Sol: treat published limits as provisional
GPT-5.6 Sol is broadly available, but its rate limit tier tables are not yet published. The figures in this post are carried forward from GPT-5.5. Sol ($5/$30), Terra ($2/$12), and Luna ($0.20/$1.20) are the verified per-tier prices — the cost reduction relative to GPT-5.5 comes from selecting Terra or Luna rather than from any price cut to Sol itself. Check platform.openai.com/account/limits for your account’s actual current limits.
Sonnet 5 tokenizer: revise your throughput math
Claude Sonnet 5’s tokenizer emits approximately 30% more tokens per request than Sonnet 4.6 for identical prompts. A workload that consumed 20K ITPM on Sonnet 4.6 will consume approximately 26K ITPM on Sonnet 5 with the same inputs. Recheck your limits against actual Sonnet 5 token counts, particularly at Tier 1 where ITPM is already constrained at 30K.
Sonnet 5 pricing: the increase was cancelled
The $2/$10 per M rate is permanent. Anthropic confirmed on August 10, 2026 that the $3/$15 standard rate scheduled for September 1 would not take effect. Cost models built on $2/$10 need no revision. Rate limits were never tied to the pricing change either way.
Gemini 3.7 Flash introductory pricing: calendar the expiry
Gemini 3.7 Flash and 3.6 Flash both run introductory pricing ($0.75/$3.75 per M) through December 31, 2026. Standard pricing ($1.50/$7.50) takes effect January 1, 2027. Cost models that project past that date should use the standard rate, not the introductory one.
Caching for throughput (not just cost)
Prompt caching has a rate limit benefit separate from its cost benefit. The cost savings are covered in the token costs post. The rate limit implications:
- Anthropic: Cached input tokens don’t count toward ITPM limits at all on Claude 4.x and 5.x models. With an 80% cache hit rate, a 2M ITPM limit effectively handles approximately 10M total input tokens per minute. Given Sonnet 5’s higher per-token counts, caching is more important for staying within ITPM limits than it was on Sonnet 4.6. Cache read cost is 10% of input price; cache write cost is 125% of input price — compute from the verified input price rather than recalling a figure.
- Groq: Cached tokens don’t count toward TPM/TPD limits on the free tier, effectively expanding your daily token budget.
- Moonshot (Kimi): Automatic 75% caching discount applied without opt-in. No cache management required.
DeepSeek V3.2 is gone
The July 24 deprecation deadline passed. If anything in your pipeline is still referencing a V3.2 alias, it is receiving errors, not rate limit responses. V4-Flash and V4-Flash 0731 are the current fast endpoints; V4-Pro for higher capability.
OpenAI Assistants API is deprecated
The Assistants API deprecated August 2026. The Responses API is the replacement. If your pipeline is still on the Assistants API, it is running on deprecated infrastructure. Migration guidance is in OpenAI’s migration docs.
Batch APIs: higher queue limits
OpenAI, Anthropic, and Google all offer batch endpoints with queue limits that far exceed real-time TPM. OpenAI’s GPT-5.6 Sol (provisional) inherits GPT-5.5’s 1.5M batch queue tokens at Tier 1 vs. 500K real-time TPM. For workloads that don’t need sub-second responses, batch endpoints let you move more tokens through tighter rate limits. See batch API cost savings for pricing details.
Model routing for rate limit distribution
Routing requests to different models by complexity distributes rate limit pressure across multiple buckets instead of concentrating it on your most constrained tier. A three-tier architecture (budget for triage, mid-tier for generation, flagship for reasoning) also cuts costs substantially. Full routing architecture with cost modeling is in the token costs post.
GPT-4.1 Nano (30K RPM / 180M TPM at Tier 5, $0.10/$0.40 per M) and DeepSeek V4-Flash 0731 (dynamic, approximately 60% lower cost than GPT-5.6 Luna per task) are strong candidates for the triage and generation layers of a routing stack. Kimi K3 (Opus-class at Sonnet-level pricing) is worth evaluating as a reasoning layer once rate limit figures are confirmed.
Custom silicon for fewer concurrent connections
Cerebras (approximately 2,100 tok/s) and SambaNova (approximately 0.2s TTFT) are faster than GPU-based providers by a large margin. A task that requires 10 parallel GPU-based API calls to meet a latency SLA might need 2–3 calls on Cerebras. Fewer concurrent calls means less rate limit pressure per unit of output.
Self-hosting as an escape hatch
Cohere’s Command A+ (Apache 2.0, 218B MoE / 25B active) runs on 2x H100 or a single B200. Kimi K3 (2.8T parameters) and Thinky Inkling (975B, Apache 2.0) are also open-weight options at substantially higher hardware requirements. For teams already operating that hardware, self-hosting removes API rate limits entirely. The economics only work at scale — hardware cost needs to be weighed against API spend.
Watch for shared pools and hidden constraints
- Anthropic model pools: Opus 5 pool assignment is not yet formally documented — check your dashboard. Opus 4.8, 4.7, and 4.6 shared a combined pool. Sonnet 5 and Sonnet 4.6 likely share another pool — not yet formally confirmed. Traffic to multiple model versions within the same pool draws from one shared bucket.
- Google’s invisible limits: With static rate limit tables removed from public docs, check AI Studio (
aistudio.google.com/rate-limit) to see your actual per-project limits. Gemini 3.7 Flash, 3.6 Flash, 3.5 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber all have no published rate limit figures. The 3.1 Flash-Lite and 3.1 Pro figures in this post are the last confirmed published numbers and may be stale for current deployments. - DeepSeek’s dynamic ceiling: No published RPM/TPM means no guaranteed minimum. Under heavy load, effective concurrency can drop without warning. Build fallback routing to a second provider if DeepSeek availability is critical to your pipeline.
- New model lag: Qwen3.8-Max (August 3), Kimi K3, Claude Opus 5, GPT-5.6 Sol, and Gemini 3.7 Flash all launched without formal rate limit tables. Figures in this post for those models are either provisional floors based on predecessors or absent entirely. Plan around confirmed figures where possible, and monitor provider dashboards for when official limits publish.
Further Reading
- LLM Token Costs and Efficiency — per-token pricing, caching mechanics, model routing, batch discounts, production cost modeling
- OpenAI rate limits — tiers, usage tracking, batch API, Responses API migration
- Anthropic rate limits — build tier system, prompt caching behavior
- Google Gemini rate limits — free vs paid, AI Studio dashboard
- Groq rate limits — real-time dashboard, daily token caps
- xAI Grok API — rate limits and pricing
- DeepSeek API — dynamic limits, V4 pricing, V3.2 deprecation
- Mistral rate limits — tier structure, RPS enforcement
- Cerebras API — WSE-3 inference, free tier
- SambaNova API — RDU inference, free tier
- Perplexity rate limits — search-augmented API tiers
- Alibaba Qwen rate limits — region-specific limits, Qwen3.8-Max
- Moonshot Kimi rate limits — tier structure, caching, K3 availability
- AWS Bedrock quotas — per-model per-region, Fable 5 status
- Azure AI quotas — PTU capacity, global deployment, GPT-5.6 Sol availability
- Cohere rate limits — production vs trial, Command A+ self-hosting
- NVIDIA NIM — hosted prototyping limits, self-hosted containers