MiniMax

You can connect to MiniMax through an MCP server to perform powerful speech, voice cloning, video, and image generation tasks from MCP clients. This server enables seamless interaction with MiniMax APIs via a lightweight local or networked setup, so you can automate creative workflows from your preferred MCP client.

How to use

Set up your MCP client to connect to the MiniMax MCP server using the stdio transport. This keeps you running locally and lets your client send requests to generate audio, images, or video, or to perform voice design and cloning tasks.

How to install

Prerequisites you need before starting:

• Python and the uv package manager must be installed (the uv install script is available at https://astral.sh/uv/install.sh)
• You will run an MCP command via the uvx launcher to start the server

Install and run the MCP server using the following steps.

{
  "mcpServers": {
    "minimax": {
      "command": "uvx",
      "args": [
        "minimax-mcp",
        "-y"
      ],
      "env": {
        "MINIMAX_API_KEY": "insert-your-api-key-here",
        "MINIMAX_MCP_BASE_PATH": "local-output-dir-path, such as /User/xxx/Desktop",
        "MINIMAX_API_HOST": "api host, https://api.minimax.io | https://api.minimaxi.com",
        "MINIMAX_API_RESOURCE_MODE": "optional, [url|local], url is default, audio/image/video are downloaded locally or provided in URL format"
      }
    }
  }
}

Additional setup notes

Important: The API host must match the region of your API key. If you see an invalid API key error, verify you are using the correct host for your region.

Configuration and troubleshooting

Transport options include stdio for local deployment and SSE for cloud deployment. When running locally, you can process local files or valid URLs. If you deploy in the cloud, using URLs for inputs is recommended.

Common issues and fixes:

• Ensure your API key and host are aligned to your region
• If you encounter ENOENT when launching the MCP, locate the absolute path to the uvx binary and update the command path in your config

Examples of typical usage

Once configured, your MCP client can perform tasks such as converting text to audio with a chosen voice, cloning voices, generating videos from prompts, creating images from prompts, and designing new voices.

Available tools

text_to_audio

Convert text to audio using a specified voice.

list_voices

List all available voices.

voice_clone

Clone a voice using provided audio samples.

generate_video

Create a video from a text prompt.

text_to_image

Generate an image from a text prompt.

query_video_generation

Query the result of a video generation task.

music_generation

Generate a music track from a prompt and lyrics.

voice_design

Generate a custom voice from a description with preview text.