Groq API pricing

• Hard capFree stops at its limits rather than billing more.
• Daily capFree allows 1,000 requests per day (ai requests).

At the limit

Stops at the limit

Limits

• AI requests: 1,000 requests a dayPer organization, for each listed model
• AI requests: 30 requests a minutePer organization, for each listed model

• Groq's rate limits page (https://console.groq.com/docs/rate-limits) shows Free plan limits by default: GPT OSS 20B, GPT OSS 120B and Qwen3.8 27B each allow 30 requests a minute and 1,000 a day. Over a limit the API returns a 429 error.
• Token limits also apply on Free (8,000 tokens a minute and 200,000 a day for these models) and you hit whichever limit comes first. Token limits are not modelled in this estimate, so long prompts can run out before the request limit.
• Limits apply to your whole organization, not each user. Your exact limits are on the Limits page in the GroqCloud console.
• We found no commercial-use ban on the pages we checked.

Source: console.groq.com · checked 19 October 2026

At the limit

Bills for extra usage

• Billed per million tokens. Upgrading from the Free tier needs a payment method; you are charged at month end or when usage crosses $1, $10, $100, $500 and $1,000.
• The rate limits page (https://console.groq.com/docs/rate-limits) shows Free plan limits by default and Developer plan limits on a second tab. Developer raises GPT OSS 20B, GPT OSS 120B and Qwen3.8 27B to 1,000 requests a minute and 500,000 a day.
• Rate limits count requests and tokens per minute and per day, per organization. Over a limit the API returns a 429 error. Token limits are not modelled in this estimate.
• Llama 3.1 8B and Llama 3.3 70B are listed as contact sales, so they have no price here. Cached tokens do not count toward rate limits.
• Spend limits and budget alerts come with the Developer tier.

Source: console.groq.com · checked 19 October 2026