Minimax Music

You use the MCP MiniMax Music Server to generate AI-powered music via the MiniMax Music API. It connects to MCP hosts so AI agents can create music and audio content from text prompts, with two-step generation, status checks, and optional reference audio for style matching.

How to use

Install and run the server on your machine, then connect it to your MCP client. Use the client’s generation workflow to submit prompts, monitor progress, and retrieve the finished audio file. You can supply a reference audio URL to guide the style, and you can reuse a generation ID to check the status of an ongoing task.

Key capabilities you will use include submitting a new generation with a text prompt, optionally choosing the minimax-music model, optionally providing a reference audio URL, and then polling for status using the returned generation ID until the audio is ready for download.

How to install

Prerequisites: Node.js v16 or higher, TypeScript v5.3.3 or higher, a supported MCP host, and an active AIML API key.

3-step setup flow to get started locally:

# 1) Clone the server repository
git clone https://github.com/yourusername/mcp-minimax-music-server.git
cd mcp-minimax-music-server

# 2) Install dependencies
npm install

# 3) Build the server
npm run build

Configuration and usage notes

Configure the MCP client (Claude Desktop or another MCP host) to reference the minimax-music server as an MCP endpoint. You specify the local server path and environment variable for authentication.

Environment variable for API key must be provided when you start the server. The client should pass an API key with the Bearer prefix.

Example of how the client references the server in a typical local setup is shown in the configuration snippet.

{
  "mcpServers": {
    "minimax-music-server": {
      "command": "node",
      "args": [
        "G:\\mcp-minimax-music-server\\build\\index.js"
      ],
      "env": {
        "AIML_API_KEY": "Bearer your-aiml-api-key-here"
      }
    }
  }
}

Available tools

generate_audio

Generates audio from a text prompt using the MiniMax Music model. Supports optional reference audio, model selection, and optional generation_id for status tracking.