Rate limits

To maintain system stability and equitable access, our API enforces rate limiting—controlling how frequently an organization can send requests within a certain time window.

How do these rate limits work?

API rate limits are defined in two ways:

TPM (Tokens per Minute) for LLM models
RPH (Requests per Hour) for video models

These limits are enforced at the organization level.

Usage tiers

Rate limits vary by usage tier, with each tier offering different quotas for each model. By default, organizations are assigned to Tier 1. To request a higher rate limit, please contact support@gmicloud.ai for an upgrade.

Rate Limit Table

Model Name	Tier 1 TPM	Tier 2 TPM	Tier 3 TPM	Tier 4 TPM	Tier 5 TPM
deepseek-ai/DeepSeek-R1	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Llama-8B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Distill-Llama-70B	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R1-Zero	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-R2	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-V3	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-V3-Base	100,000	450,000	800,000	2,000,000	30,000,000
deepseek-ai/DeepSeek-V3-0324	100,000	450,000	800,000	2,000,000	150,000,000
deepseek-ai/DeepSeek-V3.1	100,000	2,000,000	4,000,000	10,000,000	150,000,000
deepseek-ai/DeepSeek-V3.1-Terminus	100,000	2,000,000	4,000,000	10,000,000	150,000,000
deepseek-ai/DeepSeek-Prover-V2-671B	100,000	450,000	800,000	2,000,000	150,000,000
deepseek-ai/DeepSeek-R1-0528	100,000	2,000,000	4,000,000	10,000,000	150,000,000
meta-llama/Llama-3.1-8B	100,000	2,000,000	4,000,000	10,000,000	150,000,000
meta-llama/Llama-3.3-70B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
meta-llama/Llama-4-Scout-17B-16E-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
meta-llama/Llama-4-Maverick-17B-128E-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/QwQ-32B	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/QwQ-32B-Preview	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/QwQ-32B-GGUF	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/QwQ-32B-AWQ	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-0.5B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-1.5B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct-1M	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct-AWQ	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct-GGUF	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct-GPTQ-Int4	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-7B-Instruct-GPTQ-Int8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-VL-7B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-Coder-7B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-Math-7B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-14B	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-14B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-14B-Instruct-AWQ	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-14B-Instruct-GGUF	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-14B-Instruct-GPTQ-Int4	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-Coder-14B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-32B	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-32B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen2.5-Coder-32B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-235B-A22B-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-32B-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-30B-A3B	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-Next-80B-A3B-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-Next-80B-A3B-Thinking	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-235B-A22B-Instruct-2507-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
Qwen/Qwen3-235B-A22B-Thinking-2507-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
moonshotai/Kimi-K2-Instruct	100,000	2,000,000	4,000,000	10,000,000	150,000,000
moonshotai/Kimi-K2-Instruct-0905	100,000	2,000,000	4,000,000	10,000,000	150,000,000
openai/gpt-oss-120b	100,000	2,000,000	4,000,000	10,000,000	150,000,000
zai-org/GLM-4.5-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
zai-org/GLM-4.5-Air-FP8	100,000	2,000,000	4,000,000	10,000,000	150,000,000
zai-org/GLM-4.6	100,000	2,000,000	4,000,000	10,000,000	150,000,000

Getting started

API Reference

Marketplace

Resources

Billing

How do these rate limits work?

Usage tiers

Rate Limit Table

Getting started

API Reference

Marketplace

Resources

Billing

​How do these rate limits work?

​Usage tiers

​Rate Limit Table

How do these rate limits work?

Usage tiers

Rate Limit Table