Groq

Groq provides ultra-fast inference on open-source models (Llama, Gemma, Mistral, and more) using custom LPU hardware. OpenClaw connects to Groq through its OpenAI-compatible API.

Getting started

Config file example

json5
Copy
{
  env: { GROQ_API_KEY: "gsk_..." },
  agents: {
    defaults: {
      model: { primary: "groq/llama-3.3-70b-versatile" },
    },
  },
}

Built-in catalog

Groq's model catalog changes frequently. Run

to see currently available models, or check console.groq.com/docs/models.

Reasoning models

OpenClaw maps its shared

levels to Groq's model-specific values. For , disabled thinking sends and enabled thinking sends . For Groq GPT-OSS reasoning models, OpenClaw sends , , or ; disabled thinking omits because those models do not support a disabled value.

Audio transcription

Groq also provides fast Whisper-based audio transcription. When configured as a media-understanding provider, OpenClaw uses Groq's

model to transcribe voice messages through the shared surface.

json5
Copy
{
  tools: {
    media: {
      audio: {
        models: [{ provider: "groq" }],
      },
    },
  },
}

Audio transcription details

Environment note

Related