freellmcredits  home

Free Groq credits & free tier

Groq runs open models extremely fast on custom hardware, and its $0 Free plan lets you start through an OpenAI-compatible API within published per-model limits.

The short answer

The GroqCloud Free plan serves OpenAI’s gpt-oss models and Qwen at 30 requests a minute and 1,000 a day. Its model list has turned over twice in 2026 — the Llama chat models had left the free table by September, and Groq’s Compound systems were shut down on 21 September — so check the offer page’s date before you build on a specific model.

  • Groq Free LLM router
    Open

    GroqCloud free tier

    Start building on GroqCloud's $0 Free plan — very fast inference on open models including OpenAI's gpt-oss and Qwen. Groq asks for a payment method when you upgrade to the paid Developer tier.

    Free value
    No fixed amount
    Card required
    Card: not stated
    Application deadline
    No deadline stated
    Allowance resets
    See provider terms · reset time not stated · timezone not stated — Per-model per-minute and daily limits apply; use account Limits for exact organisation quotas.
    Who
    Anyone who signs up for a GroqCloud account.

    Available subject to the provider eligibility and usage conditions.

    Check eligibility & claim GroqCloud free tier (opens in a new tab) Details

Frequently asked questions

Does Groq have a free tier?
Yes. GroqCloud has a $0 Free plan with an OpenAI-compatible API. You sign up at console.groq.com, create an API key and call open models such as OpenAI’s gpt-oss and Qwen within published per-model rate limits.
Do I need a credit card for Groq?
Groq does not say either way for the Free plan. What it does say is that “To upgrade from the Free tier to the Developer tier, you’ll need to provide a valid payment method”. We mark the Free plan’s card requirement as not stated rather than guess.
What are the Groq free-tier rate limits?
Groq publishes them per model in its rate-limits documentation. On the Free plan, openai/gpt-oss-120b allows 30 requests a minute, 1,000 a day, 8,000 tokens a minute and 200,000 tokens a day; openai/gpt-oss-20b and qwen/qwen3.8-27b share those limits.
Can I still use Llama or Groq Compound on the free plan?
No Llama chat model is on Groq’s Free plan table any more, and Groq decommissioned its groq/compound and compound-mini systems on 21 September 2026. The free chat models are now OpenAI’s gpt-oss and Qwen.