MiniMax MCP

MiniMax MCP is an MCP server that connects MCP-compatible clients to MiniMax APIs for speech, audio, image, and video generation. Designed for developers, content creators, and AI engineers using assistants like Claude Desktop, Cursor, or Windsurf, this integration enables direct access to generative media features within conversational workflows. Users can synthesize natural-sounding speech from text using built-in voices, inspect existing voice profiles, and clone custom voices from user-provided audio recordings. In addition to speech synthesis, the server supports generative visual workflows by creating videos and still images directly from natural language prompts. It supports both local stdio execution and networked SSE transport modes, handling outputs either as remote URLs or downloaded local files. By exposing MiniMax's creative suite via the Model Context Protocol, the tool allows automated generation of media assets, multi-modal application prototyping, and script vocalization without switching between separate external web interfaces or bespoke API scripts.

Category: AI & LLM Tooling

Tags: image generation, minimax, text to speech, video generation, voice-cloning

Visit MiniMax MCP

How to install and configure MiniMax MCP

Follow these steps to set up MiniMax MCP: 1. Obtain an API key from the MiniMax platform. 2. Ensure uv is installed on your machine (e.g., via curl -LsSf https://astral.sh/uv/install.sh | sh). 3. In Claude Desktop, open Settings > Developer > Edit Config to modify claude_desktop_config.json (or add to Cursor under MCP settings): json { "mcpServers": { "MiniMax": { "command": "uvx", "args": ["minimax-mcp"], "env": { "MINIMAX_API_KEY": "<insert-your-api-key-here>", "MINIMAX_MCP_BASE_PATH": "<local-output-dir-path>", "MINIMAX_API_HOST": "https://api.minimaxi.chat", "MINIMAX_API_RESOURCE_MODE": "url" } } } } 4. For users in Mainland China, set MINIMAX_API_HOST to https://api.minimax.chat. Restart your client to finalize setup.

What you can do with MiniMax MCP

• Convert written text into natural-sounding speech for narration or podcast segments using specific voices through text_to_audio. - Clone custom voices from provided audio samples using voice_clone to produce personalized voiceovers for applications. - Generate short video sequences directly from descriptive prompts using the generate_video tool within AI development environments. - Create still images from text prompts with text_to_image to support concept art and rapid visual prototyping. - Query and list available MiniMax system voices using list_voices to select appropriate voices for automated speech production.

Key facts

• https://github.com/ropon/MiniMax-MCP
• AI & LLM Tooling, Design, Media & Creative
• image generation, minimax, text to speech, video generation, voice-cloning

Part of MCP Servers

How do I install MiniMax MCP?

You install and run MiniMax MCP using the uv package manager. Configure your client configuration file, such as claude_desktop_config.json, to execute minimax-mcp via uvx. Provide your MINIMAX_API_KEY, local storage path, and correct API host endpoint, then launch or restart your MCP client.

What can MiniMax MCP do?

MiniMax MCP exposes tools to convert text to speech, list supported system voices, clone custom voices using audio samples, generate video from text descriptions, and generate images directly from prompts through MiniMax APIs.

Which MCP clients work with MiniMax MCP?

It functions with standard Model Context Protocol clients that support local stdio or SSE transport, including Claude Desktop, Cursor, Windsurf, and OpenAI Agents.

Why does MiniMax MCP report an invalid API key error?

An invalid API key error usually indicates a mismatch between your API key and the API host setting. The server requires using https://api.minimaxi.chat for global accounts or https://api.minimax.chat for Mainland China accounts.

Is MiniMax MCP open source?

Yes, MiniMax MCP is open source and licensed under the MIT license, with source code available on GitHub.