← Index

io.github.Swih/mistral-mcp

io.github.Swih/mistral-mcp·v0.7.0·Image, Video & Audio
Quality Score
80
/100

Mistral AI MCP server: chat, OCR, Voxtral audio, Codestral FIM, vision, agents, batch.

§01  Install
Claude Desktop (claude_desktop_config.json)
{
  "mcpServers": {
    "mistral-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mistral-mcp"
      ],
      "env": {
        "MISTRAL_API_KEY": "<your-mistral_api_key>"
      }
    }
  }
}
Cursor (.cursor/mcp.json)
{
  "mcpServers": {
    "mistral-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mistral-mcp"
      ],
      "env": {
        "MISTRAL_API_KEY": "<your-mistral_api_key>"
      }
    }
  }
}
Cline (cline_mcp_settings.json)
npx -y mistral-mcp
§02  Environment variables
MISTRAL_API_KEY
requiredsecret

Mistral API key from https://console.mistral.ai/. The free Experiment tier provides 1B tokens/month.

§03  MCP Quality Score  ·  methodology
freshness
25
completeness
10
installability
25
documentation
15
stability
5
§04  Alternatives in Image, Video & Audio