MiniMax MCP Server
Official MiniMax MCP for speech synthesis, voice cloning, and image/video generation directly inside an AI client.
- Skill Road
- MiniMax MCP Server
Description
The MiniMax MCP Server is the official Model Context Protocol integration from MiniMax, an AI company founded in 2022 that positions itself as a provider of multimodal foundation models spanning text, audio, image, and video. The server exposes MiniMax's generative audio, image, and video capabilities to MCP-capable AI clients such as Claude Desktop, Cursor, Windsurf, or OpenAI Agents, so speech, images, and video clips can be created directly from a chat or coding assistant without opening the MiniMax web interface. It's aimed primarily at developers and content teams who want to fold generative media into existing AI workflows instead of switching between separate tools.
Speech synthesis, voice cloning, and voice design
The largest functional block covers audio generation. The text_to_audio tool converts text into spoken audio with a selectable voice, and list_voices lists the available voices along with their identifiers. voice_clone reportedly derives a new voice from an audio sample, while voice_design creates a voice purely from a text description of its characteristics, with no sample recording needed at all. This combination sets the server apart from plain text-to-speech offerings: it covers both classic speech output and custom voice branding, useful for product demos, explainer videos, or accessible content.
Image and video generation
Beyond audio, the server offers image and video generation: text_to_image generates images from text prompts, generate_video creates short video clips based on the MiniMax-Hailuo-02 model, and query_video_generation checks the processing status of asynchronous video jobs, since video generation typically takes longer than a single chat response. Combined with the audio tools, this creates an end-to-end, purely API-driven path from a text idea to a finished image, voice, or video result, which distinguishes the server from MCP servers specialized in a single medium.
Installation and regional configuration
The server runs locally via the Python package manager uv and starts with uvx minimax-mcp -y; an official JavaScript variant also exists as a separate repository. Configuration happens through environment variables: MINIMAX_API_KEY for the region-specific API key, MINIMAX_API_HOST to choose between the global endpoint api.minimax.io and the mainland China endpoint api.minimaxi.com, MINIMAX_MCP_BASE_PATH for the local output directory of generated files, and MINIMAX_API_RESOURCE_MODE to choose whether results are returned as a URL or a local file. According to the provider, the API key and host must come from the same region, otherwise authentication fails. As transport, the server supports both stdio for local execution and SSE for network-based, self-hosted cloud deployment.
Cost and access
Access requires a MiniMax account with an API key obtained through the platform's developer dashboard. The official documentation does not list a free tier for ongoing use; the provider explicitly warns in its README that using the tools may incur costs according to MiniMax's pricing structure. Current prices for speech, image, and video generation are listed on MiniMax's official pricing page.
Data handling and server location
MiniMax operates separate API endpoints for the global market and for mainland China, each with its own keys. Using the server means sending text inputs, audio samples for voice cloning, and generated media to the chosen MiniMax region for processing; results then enter the context of the connected AI model. Before production use, it's worth reviewing the privacy and terms of use for the chosen region, especially when processing voice samples of real people or confidential text content.
Who benefits from the MiniMax MCP Server?
The server suits teams that want to fold generative audio, image, or video content into existing AI-assisted development or content workflows, for example to produce prototype voiceovers, marketing images, or short video clips directly from a coding assistant. For pure text or research workflows without media generation, it's overkill — a plain research or document MCP server is sufficient there. Anyone who already holds MiniMax API access, or is specifically looking for an alternative to several separate specialized image, speech, or video services, benefits most from bundling multiple modalities into a single MCP server.
Requirements
A MiniMax account with an API key from the developer dashboard, the Python package manager uv, and an MCP-capable AI client. The JavaScript variant instead requires Node.js.
Installation instructions
The MiniMax MCP Server brings MiniMax's speech synthesis, voice cloning, and image/video generation into an MCP-capable AI client. This guide walks through obtaining an API key, installation, and the key environment variables.
Quick start
The server starts locally via uvx minimax-mcp -y and is configured through environment variables (MINIMAX_API_KEY, MINIMAX_API_HOST, MINIMAX_MCP_BASE_PATH, MINIMAX_API_RESOURCE_MODE). In Claude Desktop, it is added via an entry in claude_desktop_config.json with command uvx and arguments ["minimax-mcp", "-y"].
uvx minimax-mcp -y
Get an API key and pick a region
First, create an API key in the developer dashboard of the MiniMax platform. Choosing the right region matters: the global endpoint is https://api.minimax.io, and the mainland China endpoint is https://api.minimaxi.com. According to the provider, the API key and host region must match, or authentication fails with an "Invalid API key" error.
Step 1: Install uv and start the server
The server requires the Python package manager uv, which can be installed via curl -LsSf https://astral.sh/uv/install.sh | sh. After that, uvx minimax-mcp -y starts the server automatically as needed, with no separate installation step. For Node.js environments, MiniMax also offers a standalone JavaScript variant as a separate repository.
curl -LsSf https://astral.sh/uv/install.sh | sh
uvx minimax-mcp -y
Step 2: Configure your client
Add the server to your client's configuration file. For Claude Desktop, the entry in claude_desktop_config.json looks like this: command uvx, arguments ["minimax-mcp", "-y"], plus the environment variables MINIMAX_API_KEY (your API key), MINIMAX_API_HOST (your chosen region), MINIMAX_MCP_BASE_PATH (local output directory for generated files), and optionally MINIMAX_API_RESOURCE_MODE (url or local, default url). According to the provider, Windows users must additionally enable Developer Mode in Claude Desktop's settings to allow local processes to run.
First steps with audio, image, and video
Test the connection first with list_voices to see the available voices, then generate a first piece of speech with text_to_audio. For custom voices, use voice_clone to derive a voice from an audio sample, or voice_design to create a voice purely from a text description. Images come from text_to_image, and videos from generate_video based on the MiniMax-Hailuo-02 model; since video generation takes longer, check progress with query_video_generation.
Keep an eye on costs
Since there is no permanently free tier according to the provider, and every generation can incur costs based on MiniMax's pricing structure, it's worth checking the official pricing page before embedding the server into automated or high-frequency workflows.
FAQ
Do I need a paid subscription for the MiniMax MCP Server? The documentation does not describe a permanently free tier; according to the provider, using the tools incurs costs based on MiniMax's pricing structure.
Which host endpoint should I use? Use api.minimax.io for global access and api.minimaxi.com for mainland China accounts. The API key and host must belong to the same region.
Can I clone someone else's voice without consent? It's technically possible, but doing so without the recorded person's consent raises legal and ethical concerns; only use voice_clone with your own recordings or explicit permission.
Why does video generation take longer than a normal reply? Video generation runs asynchronously; generate_video starts the job, and query_video_generation then checks its processing status.
Does the server work without Python? For Node.js environments, MiniMax offers a standalone JavaScript variant as a separate repository; the standard path described here, however, uses uv/uvx.
uvx minimax-mcp -y
Authentication
API key from the MiniMax developer dashboard, which must match the chosen region (global or mainland China). There is no OAuth sign-in.
Required access permissions
The API key grants access to all audio, image, and video features enabled in the MiniMax account. There is no fine-grained permission scoping within the MCP server itself.
Transmitted or stored data
Text inputs, audio samples, and generated media are sent to the chosen MiniMax region (global or China) for processing and then enter the context of the connected AI model.
Security risks
Voice-cloning features process audio samples of real people; without the recorded person's consent there is potential for misuse. The API key and host region must match, or authentication fails. There is no free ongoing tier: according to the provider, usage incurs costs based on MiniMax's pricing structure.
License and costs
- License
- MIT
- Cost
- paid
Usage requires a MiniMax API key; the documentation does not describe a permanently free tier. According to the provider, using the tools incurs costs based on MiniMax's pricing structure for speech, image, and video generation. See MiniMax's official pricing page for current prices.
Alternatives
Not recorded yet.
At a glance
- Provider
- MiniMax
- Status
- Official server
- Deployment
- Local
- Current version
- 0.0.19
- GitHub stars
- 1,591
- Last reviewed
- 13.09.2026
Repository and documentation
Categories
Supported clients
Related guides
Guides and background related to this entry.