Why do I get an "invalid api key" error?
This occurs when your API key and API host are not regionally aligned. Use `https://api.minimax.io` for Global keys and `https://api.minimaxi.com` for Mainland China keys.
@MiniMax-AI
基本信息
Official MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"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": "<optional, [url|local], url is default, audio/image/video are downloaded locally or provided in URL format>"
}
}
}
}Convert text to audio with a given voice and save the output audio file to a given directory. If no directory is provided, the file will be saved to desktop. If no voice ID is provided, the default voice will be used. Note: This tool calls MiniMax API and may incur costs. Use only when explicitly requested by the user.
List all available voices. Only supported when api_host is https://api.minimax.chat.
Play an audio file. Supports WAV and MP3 formats. Does not support video.
Clone a voice using the provided audio file. New voices will incur costs when first used. Note: This tool calls MiniMax API and may incur costs. Use only when explicitly requested by the user.
Generate images based on text prompts. Note: This tool calls MiniMax API and may incur costs. Use only when explicitly requested by the user.
Generate a video based on text prompts. Note: This tool calls MiniMax API and may incur costs. Use only when explicitly requested by the user.
Generate a video based on an image. Note: This tool calls MiniMax API and may incur costs. Use only when explicitly requested by the user.
Query the status of a video generation task.
MiniMax MCP is the official Model Context Protocol server from MiniMax. It enables MCP clients (such as Claude Desktop, Cursor, Windsurf, and OpenAI Agents) to interact with MiniMax's powerful Text-to-Speech, voice cloning, video generation, and image generation APIs. It is designed for developers who want to integrate MiniMax's generative AI capabilities into their tools.
Obtain an API key from MiniMax and install uv (Python package manager). Configure your MCP client (e.g., Claude Desktop or Cursor) with the required environment variables: MINIMAX_API_KEY, MINIMAX_MCP_BASE_PATH, MINIMAX_API_HOST (region-specific), and optionally MINIMAX_API_RESOURCE_MODE. After setup, the client can invoke any of the eight available tools.
This occurs when your API key and API host are not regionally aligned. Use `https://api.minimax.io` for Global keys and `https://api.minimaxi.com` for Mainland China keys.
It means `uvx` is not in your PATH. Run `which uvx` to find its absolute path (e.g., `/usr/local/bin/uvx`) and update your MCP client configuration to use that path.
You must define completion rules either in a dedicated cursor rule file or in your IDE settings (e.g., Cursor) before starting the video generation task.
It supports both stdio (local, communicates via stdout, can process local files or URLs) and SSE (can be deployed locally or in the cloud, communicates over network, recommended for cloud deployment with URL inputs).
You need Python and `uv` (Python package manager). The server is run via `uvx minimax-mcp -y`. No other explicit runtime requirements are stated.

Creates and manages quizzes, polls, surveys, personality tests, forms, predictors, minigames and leaderboards on Riddle.com – with branching logic, custom design, live stats and embed code.

YouTube download API and MCP server. Download YouTube videos as MP4 (360p–4K) or audio as MP3/AAC, cut clips by start and end time, and get video metadata. No yt-dlp, proxies or cookies to maintain; 99.99% of downloads s

Create and edit images, videos, music, speech, and sound effects with your getimg account. Use ready-made workflows for professional headshots, photo restoration, stickers, lighting changes, background cleanup, product v
Generate images, videos, music, and voice-overs directly from your AI agent. The full Artlist AI creative suite as an MCP server — works with Claude, ChatGPT, and VS Code, with more platforms coming soon.

Transcribe YouTube videos, podcasts, meetings and media links into speaker-labelled, timestamped transcripts, then search who said what, ask questions answered with timestamped citations, pull chapters and paste-ready ti
评论