# MiniMax-MCP

The official MiniMax Model Context Protocol server that lets Claude Desktop, Cursor, Windsurf and other MCP clients generate speech, clone voices, and generate images and video.

MiniMax-MCP is MiniMax's official Model Context Protocol server, giving MCP clients such as Claude Desktop, Cursor, Windsurf and OpenAI Agents access to MiniMax's text-to-speech, voice-cloning, image-generation and video-generation APIs through a standard set of tools. It requires an API key from MiniMax, and the API host must match the account's region, either the global or the mainland China endpoint, or requests fail with an invalid-key error.

It supports two transports: stdio, for running locally with communication over standard input and output, or SSE, which can run locally or in the cloud and takes URL-based input for cloud deployments. Its tools include converting text to audio with a chosen voice, listing available voices, cloning a voice from sample audio, and generating video and images. A separate official command-line tool adds text, vision and search capabilities as an agent skill for Claude Code, Cursor and OpenClaw. It is written in Python and licensed under MIT, with a parallel official JavaScript implementation.

Website: https://www.minimax.io/platform