LLM enrollment rate limits

The table below contains enrollment limits for tokens per minute (TPM) and requests per minute (RPM) for each enrollment tier. For enrollments with both Azure and OpenAI enabled, enrollment limits will be double what is shown below for Azure and OpenAI. Additionally, for enrollments geo-restricted to a single region, TPM and RPM may be lower than the table indicates in the Large and X-large tiers.

If multiple backends are enabled, the rate limits are summed across all backends.

Model NameModel BackendPer-user LimitsSmall TierMedium TierLarge TierXLarge Tier
Claude 3 HaikuAmazon Bedrock270K TPM
770 RPM
100K TPM
100 RPM
600K TPM
1K RPM
1.5M TPM
1.5K RPM
2M TPM
2K RPM
Claude 3.5 HaikuGoogle Vertex500K TPM
1K RPM
100K TPM
100 RPM
500K TPM
250 RPM
750K TPM
375 RPM
1M TPM
500 RPM
Claude 3.5 HaikuAmazon Bedrock500K TPM
1K RPM
100K TPM
100 RPM
1M TPM
1K RPM
1.5M TPM
1.5K RPM
2M TPM
2K RPM
Claude 3.7 SonnetDirect Anthropic400K TPM
100 RPM
100K TPM
25 RPM
500K TPM
500 RPM
750K TPM
750 RPM
1M TPM
1K RPM
Claude 3.7 SonnetGoogle Vertex400K TPM
100 RPM
100K TPM
25 RPM
400K TPM
50 RPM
600K TPM
75 RPM
800K TPM
100 RPM
Claude 3.7 SonnetAmazon Bedrock400K TPM
100 RPM
100K TPM
25 RPM
2M TPM
500 RPM
3M TPM
750 RPM
4M TPM
1K RPM
Claude Sonnet 4Direct Anthropic400K TPM
25 RPM
100K TPM
25 RPM
500K TPM
500 RPM
750K TPM
750 RPM
1M TPM
1K RPM
Claude Sonnet 4Google Vertex400K TPM
25 RPM
100K TPM
25 RPM
500K TPM
50 RPM
750K TPM
75 RPM
1M TPM
100 RPM
Claude Sonnet 4Amazon Bedrock400K TPM
25 RPM
100K TPM
25 RPM
2M TPM
500 RPM
3M TPM
750 RPM
4M TPM
1K RPM
Claude Opus 4Direct Anthropic100K TPM
5 RPM
100K TPM
25 RPM
250K TPM
250 RPM
375K TPM
375 RPM
500K TPM
500 RPM
Claude Opus 4Google Vertex100K TPM
5 RPM
100K TPM
25 RPM
150K TPM
50 RPM
200K TPM
75 RPM
250K TPM
100 RPM
Claude Opus 4Amazon Bedrock100K TPM
5 RPM
100K TPM
25 RPM
125K TPM
50 RPM
150K TPM
75 RPM
200K TPM
100 RPM
Claude Opus 4.1Direct Anthropic400K TPM
5 RPM
100K TPM
25 RPM
500K TPM
200 RPM
750K TPM
300 RPM
1M TPM
400 RPM
Claude Opus 4.1Google Vertex400K TPM
5 RPM
100K TPM
25 RPM
400K TPM
100 RPM
600K TPM
150 RPM
800K TPM
200 RPM
Claude Opus 4.1Amazon Bedrock400K TPM
5 RPM
100K TPM
25 RPM
500K TPM
100 RPM
1M TPM
150 RPM
2M TPM
200 RPM
Claude Sonnet 4.5Direct Anthropic1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Claude Sonnet 4.5Google Vertex1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
500 RPM
1.5M TPM
750 RPM
2M TPM
1K RPM
Claude Sonnet 4.5Amazon Bedrock1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
200 RPM
4M TPM
500 RPM
8M TPM
1K RPM
Claude Opus 4.5Direct Anthropic1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Claude Opus 4.5Google Vertex1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
100 RPM
1.5M TPM
150 RPM
2M TPM
200 RPM
Claude Opus 4.5Amazon Bedrock1M TPM
100 RPM
100K TPM
25 RPM
1M TPM
100 RPM
2M TPM
200 RPM
4M TPM
400 RPM
Claude Haiku 4.5Direct Anthropic1M TPM
100 RPM
100K TPM
100 RPM
1M TPM
250 RPM
1.5M TPM
375 RPM
2M TPM
500 RPM
Claude Haiku 4.5Google Vertex1M TPM
100 RPM
100K TPM
50 RPM
1M TPM
100 RPM
1.5M TPM
150 RPM
2M TPM
200 RPM
Claude Haiku 4.5Amazon Bedrock1M TPM
100 RPM
100K TPM
50 RPM
1M TPM
200 RPM
2.5M TPM
500 RPM
5M TPM
1K RPM
Claude Opus 4.6Direct Anthropic1.5M TPM
150 RPM
100K TPM
10 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Claude Opus 4.6Google Vertex1.5M TPM
150 RPM
200K TPM
20 RPM
1M TPM
100 RPM
1.5M TPM
150 RPM
2M TPM
200 RPM
Claude Opus 4.6Amazon Bedrock1.5M TPM
150 RPM
200K TPM
20 RPM
3M TPM
300 RPM
4M TPM
400 RPM
6M TPM
600 RPM
Claude Sonnet 4.6Direct Anthropic1M TPM
100 RPM
100K TPM
10 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Claude Sonnet 4.6Google Vertex1M TPM
100 RPM
200K TPM
20 RPM
1M TPM
500 RPM
1.5M TPM
750 RPM
2M TPM
1K RPM
Claude Sonnet 4.6Amazon Bedrock1M TPM
100 RPM
200K TPM
20 RPM
2M TPM
250 RPM
4M TPM
500 RPM
8M TPM
1K RPM
Claude Opus 4.7Direct Anthropic1.5M TPM
150 RPM
100K TPM
10 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Claude Opus 4.7Google Vertex1.5M TPM
150 RPM
200K TPM
20 RPM
1M TPM
100 RPM
1.5M TPM
150 RPM
2M TPM
200 RPM
Claude Opus 4.7Amazon Bedrock1.5M TPM
150 RPM
200K TPM
20 RPM
3M TPM
300 RPM
4M TPM
400 RPM
6M TPM
600 RPM
Llama 3.1 8b InstructPalantir Hub50K TPM
100 RPM
100K TPM
100 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.1 8b InstructAmazon Bedrock50K TPM
100 RPM
100K TPM
100 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.1 70b InstructPalantir Hub50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.1 70b InstructAmazon Bedrock50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.3 70b InstructPalantir Hub50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.3 70b InstructAmazon Bedrock50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 4 Scout 17b 16E InstructPalantir Hub100K TPM
100 RPM
100K TPM
100 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 4 Scout 17b 16E InstructAmazon Bedrock100K TPM
100 RPM
100K TPM
100 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 4 Maverick 17b 128E InstructAmazon Bedrock100K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.3 Nemotron Super 49b v1.5Palantir Hub50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Llama 3.2 NV EmbedQA 1B v2Palantir Hub50K TPM
100 RPM
60K TPM
150 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
NVIDIA Nemotron 3 Nano 30BAmazon Bedrock50K TPM
100 RPM
100K TPM
25 RPM
500K TPM
100 RPM
1M TPM
150 RPM
2M TPM
200 RPM
NVIDIA Nemotron 3 Super 120BAmazon Bedrock500K TPM
100 RPM
40K TPM
10 RPM
1M TPM
200 RPM
2M TPM
300 RPM
4M TPM
400 RPM
Grok 3xAI100K TPM
100 RPM
100K TPM
25 RPM
1M TPM
100 RPM
2M TPM
250 RPM
3M TPM
500 RPM
Grok 4xAI1M TPM
100 RPM
500K TPM
100 RPM
4M TPM
200 RPM
8M TPM
500 RPM
12M TPM
1K RPM
Grok 4 Fast (Reasoning)xAI1M TPM
100 RPM
100K TPM
25 RPM
4M TPM
200 RPM
8M TPM
400 RPM
12M TPM
1K RPM
Grok 4 Fast (Non-Reasoning)xAI1M TPM
100 RPM
100K TPM
100 RPM
4M TPM
200 RPM
8M TPM
400 RPM
12M TPM
1K RPM
Grok 4.1 Fast (Reasoning)xAI1M TPM
100 RPM
100K TPM
25 RPM
4M TPM
200 RPM
8M TPM
400 RPM
12M TPM
1K RPM
Grok 4.1 Fast (Non-Reasoning)xAI1M TPM
100 RPM
100K TPM
100 RPM
4M TPM
200 RPM
8M TPM
400 RPM
12M TPM
1K RPM
Grok Code Fast 1xAI400K TPM
100 RPM
100K TPM
100 RPM
2M TPM
200 RPM
4M TPM
400 RPM
6M TPM
1K RPM
Grok 3 Mini (with Thinking)xAI50K TPM
100 RPM
100K TPM
25 RPM
600K TPM
50 RPM
1M TPM
100 RPM
1.2M TPM
150 RPM
Grok 420 0121 ReasoningxAI500K TPM
100 RPM
100K TPM
25 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Grok 420 0118 ReasoningxAI500K TPM
100 RPM
100K TPM
25 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Grok 420 Reasoning LatestxAI500K TPM
100 RPM
50K TPM
20 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Grok 420 Non-Reasoning LatestxAI500K TPM
100 RPM
50K TPM
20 RPM
1M TPM
200 RPM
1.5M TPM
300 RPM
2M TPM
400 RPM
Schematic 7BPalantir Hub50K TPM
100 RPM
60K TPM
150 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
Document Information ExtractionPalantir Hub1M TPM
40 RPM
1M TPM
40 RPM
1.5M TPM
300 RPM
2M TPM
450 RPM
3M TPM
600 RPM
Snowflake Arctic Embed MediumPalantir Hub500K TPM
500 RPM
60K TPM
150 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
GPT-4oDirect OpenAI400K TPM
800 RPM
100K TPM
25 RPM
1M TPM
1K RPM
1.5M TPM
2K RPM
3M TPM
4K RPM
GPT-4oAzure OpenAI400K TPM
800 RPM
100K TPM
25 RPM
1M TPM
1K RPM
1.5M TPM
2K RPM
3M TPM
4K RPM
GPT-4o miniDirect OpenAI300K TPM
800 RPM
100K TPM
100 RPM
1M TPM
1K RPM
1.5M TPM
2K RPM
3M TPM
4K RPM
GPT-4o miniAzure OpenAI300K TPM
800 RPM
100K TPM
100 RPM
1M TPM
1K RPM
1.5M TPM
2K RPM
3M TPM
4K RPM
GPT-4.1Direct OpenAI400K TPM
1K RPM
100K TPM
25 RPM
1.5M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-4.1Azure OpenAI400K TPM
1K RPM
100K TPM
25 RPM
1.5M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-4.1 miniDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
2M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-4.1 miniAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
10M TPM
2.5K RPM
30M TPM
7.5K RPM
50M TPM
12.5K RPM
GPT-4.1 nanoDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
2M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-4.1 nanoAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
1M TPM
2.5K RPM
30M TPM
7.5K RPM
50M TPM
12.5K RPM
GPT-5Direct OpenAI1M TPM
1K RPM
100K TPM
25 RPM
3M TPM
1.5K RPM
6M TPM
3K RPM
10M TPM
5K RPM
GPT-5Azure OpenAI1M TPM
1K RPM
100K TPM
25 RPM
3M TPM
1K RPM
5M TPM
2.5K RPM
10M TPM
5K RPM
GPT-5 miniDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
3M TPM
1K RPM
5M TPM
2K RPM
7M TPM
4K RPM
GPT-5 miniAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
10M TPM
5K RPM
20M TPM
10K RPM
30M TPM
15K RPM
GPT-5 nanoDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
5M TPM
2.5K RPM
10M TPM
5K RPM
20M TPM
10K RPM
GPT-5 nanoAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
10M TPM
5K RPM
30M TPM
15K RPM
50M TPM
25K RPM
GPT-5 CodexDirect OpenAI1M TPM
1K RPM
100K TPM
25 RPM
3M TPM
1K RPM
4M TPM
2K RPM
5M TPM
4K RPM
GPT-5 CodexAzure OpenAI1M TPM
1K RPM
100K TPM
25 RPM
2M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-5.1Direct OpenAI500K TPM
1K RPM
100K TPM
25 RPM
1.5M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-5.1Azure OpenAI500K TPM
1K RPM
100K TPM
25 RPM
2M TPM
500 RPM
4M TPM
1K RPM
6M TPM
2K RPM
GPT-5.1 CodexDirect OpenAI1M TPM
1K RPM
100K TPM
25 RPM
3M TPM
1K RPM
4M TPM
2K RPM
5M TPM
4K RPM
GPT-5.1 CodexAzure OpenAI1M TPM
1K RPM
100K TPM
25 RPM
2M TPM
1K RPM
3M TPM
2K RPM
4M TPM
4K RPM
GPT-5.1 Codex miniDirect OpenAI1M TPM
500 RPM
100K TPM
100 RPM
3M TPM
1K RPM
4M TPM
2K RPM
5M TPM
4K RPM
GPT-5.1 Codex miniAzure OpenAI1M TPM
500 RPM
100K TPM
100 RPM
2M TPM
1K RPM
3M TPM
2K RPM
5M TPM
4K RPM
GPT-5.2Direct OpenAI500K TPM
1K RPM
250K TPM
50 RPM
3M TPM
1.5K RPM
6M TPM
3K RPM
10M TPM
5K RPM
GPT-5.2Azure OpenAI500K TPM
1K RPM
250K TPM
50 RPM
2M TPM
500 RPM
4M TPM
1K RPM
6M TPM
2K RPM
GPT-5.4Direct OpenAI1M TPM
1K RPM
250K TPM
50 RPM
3M TPM
1.5K RPM
6M TPM
3K RPM
10M TPM
5K RPM
GPT-5.4Azure OpenAI1M TPM
1K RPM
250K TPM
50 RPM
4M TPM
2K RPM
6M TPM
3K RPM
8M TPM
4K RPM
GPT-5.4 miniDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
3M TPM
1.5K RPM
6M TPM
3K RPM
10M TPM
5K RPM
GPT-5.4 miniAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
4.5M TPM
2.2K RPM
9M TPM
4.5K RPM
15M TPM
7.5K RPM
GPT-5.4 nanoDirect OpenAI1M TPM
1K RPM
100K TPM
100 RPM
3M TPM
1.5K RPM
6M TPM
3K RPM
10M TPM
5K RPM
GPT-5.4 nanoAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
4.5M TPM
2.2K RPM
9M TPM
4.5K RPM
15M TPM
7.5K RPM
GPT-5.3 CodexDirect OpenAI1M TPM
1K RPM
100K TPM
25 RPM
3M TPM
1K RPM
4M TPM
2K RPM
5M TPM
4K RPM
GPT-5.3 CodexAzure OpenAI1M TPM
1K RPM
100K TPM
100 RPM
4M TPM
2K RPM
6M TPM
4K RPM
8M TPM
8K RPM
GPT-OSS-20BPalantir Hub50K TPM
100 RPM
100K TPM
100 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
GPT-OSS-120BPalantir Hub50K TPM
100 RPM
100K TPM
25 RPM
300K TPM
450 RPM
450K TPM
675 RPM
600K TPM
900 RPM
o1Azure OpenAI600K TPM
5 RPM
100K TPM
25 RPM
250K TPM
50 RPM
400K TPM
60 RPM
750K TPM
75 RPM
o3Direct OpenAI400K TPM
100 RPM
100K TPM
25 RPM
1M TPM
1K RPM
2M TPM
2K RPM
4M TPM
4K RPM
o3Azure OpenAI400K TPM
100 RPM
100K TPM
25 RPM
1M TPM
1K RPM
2M TPM
2K RPM
4M TPM
4K RPM
o4-miniDirect OpenAI300K TPM
100 RPM
100K TPM
25 RPM
1M TPM
1K RPM
2M TPM
2K RPM
4M TPM
4K RPM
o4-miniAzure OpenAI300K TPM
100 RPM
100K TPM
25 RPM
1M TPM
1K RPM
2M TPM
2K RPM
4M TPM
4K RPM
text-embedding-ada-002Direct OpenAI1M TPM
1.5K RPM
450K TPM
450 RPM
2.1M TPM
2.1K RPM
3.1M TPM
3.1K RPM
4.2M TPM
4.2K RPM
text-embedding-ada-002Azure OpenAI1M TPM
1.5K RPM
450K TPM
450 RPM
2.1M TPM
2.1K RPM
3.1M TPM
3.1K RPM
4.2M TPM
4.2K RPM
Text Embedding 3 SmallDirect OpenAI1M TPM
1.5K RPM
60K TPM
400 RPM
500K TPM
2K RPM
1M TPM
3K RPM
1.5M TPM
6K RPM
Text Embedding 3 SmallAzure OpenAI1M TPM
1.5K RPM
60K TPM
400 RPM
500K TPM
2K RPM
1M TPM
3K RPM
1.5M TPM
6K RPM
Text Embedding 3 LargeDirect OpenAI1M TPM
1.5K RPM
60K TPM
400 RPM
1M TPM
2K RPM
2M TPM
3K RPM
3M TPM
6K RPM
Text Embedding 3 LargeAzure OpenAI1M TPM
1.5K RPM
60K TPM
400 RPM
1M TPM
2K RPM
2M TPM
3K RPM
3M TPM
6K RPM
Gemini 2.5 FlashGoogle Vertex1M TPM
200 RPM
100K TPM
25 RPM
2M TPM
1.2K RPM
3M TPM
2.4K RPM
4M TPM
4K RPM
Gemini 2.5 ProGoogle Vertex1M TPM
200 RPM
100K TPM
25 RPM
4M TPM
600 RPM
6M TPM
1.2K RPM
8M TPM
2K RPM
Gemini 2.5 Flash LiteGoogle Vertex1M TPM
200 RPM
100K TPM
100 RPM
2M TPM
1.2K RPM
3M TPM
2.4K RPM
4M TPM
4K RPM
Gemini 3 Flash (Preview)Google Vertex1M TPM
200 RPM
100K TPM
100 RPM
6M TPM
900 RPM
9M TPM
1.8K RPM
12M TPM
3K RPM
Gemini 3.1 Pro (Preview)Google Vertex1M TPM
200 RPM
500K TPM
100 RPM
6M TPM
900 RPM
9M TPM
1.8K RPM
12M TPM
3K RPM