MiniMax MCP Server

Example Use Cases

• 1

  Synthesize high-quality speech or clone voices from sample audio for broadcasting, narration, or personalized assistants.

• 2

  Generate images and videos from text prompts, manage asynchronous video jobs, and retrieve results via URL or local download.

• 3

  Create music tracks or design custom voices programmatically via MCP-compatible clients (e.g., Claude Desktop, Cursor, OpenAI Agents).

Description

This repository provides the official MiniMax Model Context Protocol (MCP) server to bridge MCP clients (e.g., Claude Desktop, Cursor, Windsurf, OpenAI Agents) with MiniMax's text-to-audio, voice cloning, image/video generation and music APIs. It supports two transport modes (stdio and SSE), local file handling or URL-based resources, and can download generated assets locally or return URLs. The server includes tools for listing voices, cloning voices, text-to-audio, voice design, image/video generation, music generation, and querying async video jobs. It is intended to make it easy to integrate MiniMax multimodal capabilities into agent workflows and IDEs.