Groq
Category: Platform
| Detail | Value |
|---|---|
| Models | LPU inference for open models |
| Context | 128K |
| Pricing | Free tier, then pay-per-token |
Overview
Custom LPU (Language Processing Unit) hardware delivering the fastest inference speeds in the industry — 500+ tokens/sec. Free tier available. Supports Llama, Mistral, and Gemma models.