Machine Learning — entry 006 of 24
Groq
Groq runs open models such as Llama, Mixtral, and Gemma on its own LPU inference hardware, exposing them through a REST API that's a near drop-in replacement for the OpenAI chat completions endpoint. Pricing is pay-per-token and varies by model (roughly $0.05-$3 per million tokens), with a free API key available to get started and no separate infrastructure fee. The console also offers batch processing at a 50% discount and prompt-caching discounts for repeated context.
GreatAPIs Score
Embed this badge
<a href="https://greatapis.com/api/groq/"><img src="https://greatapis.com/badge/groq.svg" alt="Scored 75 on greatapis.com"></a>Auth quickstart
- Sign up with the provider to get an API key.
- Send it on every request as a header
Authorization: <key> - The exact header isn't documented — Authorization is a common default; confirm in the provider's docs.
Your key is stored only in this browser (localStorage) and sent directly to the API — never to greatapis.
Developer reference
Free tier varies by model, e.g. llama-3.1-8b-instant: 30 requests/min, 14,400 requests/day, 6,000 tokens/min, 500,000 tokens/day; llama-3.3-70b-versatile: 30 requests/min, 1,000 requests/day; whisper-large-v3: 20 requests/min, 2,000 requests/day