Integrate Mistral with OpenClaw: Quick Setup Guide

Ever feel like setting up a new AI provider takes more time than actually building your app? It is often a pain to figure out which environment variables to set or which model names to use for different tasks.

OpenClaw supports Mistral for text and image model routing using the mistral/... prefix. You can also handle audio transcription through Voxtral in media understanding. If you need memory embeddings, just set memorySearch.provider = "mistral".

CLI setup

To get things moving, you can use the onboard command. It handles the authentication setup for you.

openclaw onboard --auth-choice mistral-api-key
# or non-interactive
openclaw onboard --mistral-api-key "$MISTRAL_API_KEY"

Config snippet (LLM provider)

If you prefer manual configuration, you can add Mistral to your config file. This sets your API key and defines the primary model for your agents.

{
  env: { MISTRAL_API_KEY: "sk-..." },
  agents: { defaults: { model: { primary: "mistral/mistral-large-latest" } } },
}

Config snippet (audio transcription with Voxtral)

Mistral also handles audio tasks. You can enable audio tools and specify the Voxtral model in your settings.

{
  tools: {
    media: {
      audio: {
        enabled: true,
        models: [{ provider: "mistral", model: "voxtral-mini-latest" }],
      },
    },
  },
}

Notes

Here are a few details to keep in mind while you configure your setup:

• Mistral auth uses MISTRAL_API_KEY.
• Provider base URL defaults to https://api.mistral.ai/v1.
• Onboarding default model is mistral/mistral-large-latest.
• Media-understanding default audio model for Mistral is voxtral-mini-latest.
• Media transcription path uses /v1/audio/transcriptions.
• Memory embeddings path uses /v1/embeddings (default model: mistral-embed).

AI Setup Assistant

Next steps

• Memory configuration
• Media understanding