Loading...
Loading...
Groq runs a custom LPU inference chip delivering sub-100ms latency on open-weight models — the fastest inference provider for Llama and Mistral.
8
indexed models
16
total catalog
$0.050
cheapest model
0
free / open source
| Model | Input $/1M | Output $/1M | Context | |
|---|---|---|---|---|
llama-3.1-8b-instant Meta's Llama 3.1 8B served on Groq's LPU for ultra-low latency — ideal for fast, lightweight text tasks. | $0.050 | $0.080 | 131k | |
llama-3.3-70b-versatile Meta's Llama 3.3 70B — latest iteration with improved instruction following, served on Groq LPU. | $0.590 | $0.790 | 131k | |
Compare Groq models side-by-side
See price, speed, and benchmark differences between any two models.
OpenAI's Whisper Large v3 served on Groq for fast, accurate speech-to-text transcription.
| — |
| — |
| 0k |
qwen/qwen3-32b Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for... | — | — | 131k |
whisper-large-v3-turbo Groq-optimised Whisper Large v3 Turbo — fastest ASR with accuracy comparable to Whisper v3. | — | — | 0k |
openai/gpt-oss-120b gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized... | — | — | 131k |
openai/gpt-oss-safeguard-20b gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust... | — | — | 131k |
openai/gpt-oss-20b gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for... | — | — | 131k |