Groq API free tier
Groq publishes sample Free-plan limits per model, applied at the organization level, and says the current exact limits for an organization are on the limits page in account settings, with possible exceptions. Text models such as gpt-oss-120b and Qwen3.8 27B list 30 requests a minute, 1,000 requests a day, 8,000 tokens a minute and 200,000 tokens a day.
Free tier: GroqCloud API, Free plan
Published on the provider page| Limit | Value | Applies to | Note |
|---|---|---|---|
| Requests per minute | 30 | openai/gpt-oss-120b | Free plan sample limit |
| Requests per day | 1,000 | openai/gpt-oss-120b | Free plan sample limit |
| Tokens per minute | 8,000 | openai/gpt-oss-120b | Free plan sample limit |
| Tokens per day | 200,000 | openai/gpt-oss-120b | Free plan sample limit |
| Requests per minute | 30 | qwen/qwen3.8-27b | Free plan sample limit |
| Requests per day | 1,000 | qwen/qwen3.8-27b | Free plan sample limit |
| Tokens per minute | 8,000 | qwen/qwen3.8-27b | Free plan sample limit |
| Tokens per day | 200,000 | qwen/qwen3.8-27b | Free plan sample limit |
- What moves you past this tier
- Upgrading to the Developer plan gives higher limits plus Batch and Flex processing; enterprise limits are available for select workloads.
- Card required
- Not stated by the provider
- Source
- Groq rate limits Read 2026-10-16
Limits apply per organization, not per user; the docs table is a sample and the console limits page is authoritative. groq/compound no longer appears in the Free plan table.
What Groq charges after the free tier
Groq publishes per-token prices for 4 text models. The highest-scoring ones are below, per million tokens, read from its own pages on October 20, 2026.
- Qwen 3.8 27B$0.80 in · $4.00 out
- GPT OSS 120B$0.15 in · $0.60 out
- GPT OSS 20B$0.075 in · $0.30 out
- Safety GPT OSS 20B$0.075 in · $0.30 out
Questions
Is the Groq API free?
GroqCloud has a Free plan. Groq publishes sample limits per model for it: 30 requests a minute, 1,000 requests a day, 8,000 tokens a minute and 200,000 tokens a day on text models such as gpt-oss-120b and Qwen3.8 27B, with the Compound systems at 30 requests a minute and 250 a day.
Are Groq rate limits per user or per organisation?
Per organisation. Groq states that rate limits apply at the organisation level, not to individual users, so extra keys or members do not add capacity. The documentation table is a sample; the exact current limits for an organisation are on the limits page in account settings, and Groq notes there may be exceptions.
What does the Groq Developer plan change?
Groq says upgrading to the Developer plan gives access to higher limits plus Batch and Flex processing, and that higher limits are available for select workloads and enterprise use cases. The plan’s price and whether a card is required are not stated on the rate-limits page we read, so we do not record them.