Skip to content
MCP server

ModelsLab

Hosted MCP server to generate images, video, speech and music across 10,000+ models

Actively maintainedStatus and history

The repository was last updated 18 days ago and is published under MIT.

Install
Platforms
API, Web
Runs
Remotely
GitHub stars
0
Last commit
18 days ago
License
MIT
Version
1.0.0

Similar MCP servers

About ModelsLab

ModelsLab is an MCP server for design and media. It runs as a hosted remote server, so there is nothing to install locally.

ModelsLab runs two hosted MCP servers, and this entry is the Generation one at modelslab.com/mcp/v7. A second, the Agent Control Plane at modelslab.com/mcp/agents, handles account, billing, subscriptions and teams and uses a bearer token. Generation accepts either OAuth 2.1 sign-in or an API key sent as a Bearer or X-API-Key header.

Tools are grouped as discovery, image, video, audio, LLM and jobs. Image tools cover text-to-image, image-to-image and inpainting; video covers text, image and video to video plus lip-sync and motion control; audio covers text-to-speech, speech-to-text, speech-to-speech, sound and music generation, song extending and inpainting, and dubbing; chat-completion runs language models. Most generation is asynchronous: a tool returns a processing status and an id, and the client polls fetch-generation, or a type-specific fetch tool, until it reports success or error. Call list-models first to get a valid model_id, with families named such as Flux, Qwen Image, Seedream, Kling, Veo and Wan.

The README gives setup steps for Claude, Claude Code, Cursor, VS Code, Windsurf, Gemini CLI and Codex, and a mcp-remote route for stdio-only clients. Usage prices are not stated in the material.

Install

Run in a terminal

claude mcp add --transport http modelslab https://modelslab.com/mcp/v7

Add to Settings → Connectors → Add custom connector

https://modelslab.com/mcp/v7
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "modelslab": {
      "url": "https://modelslab.com/mcp/v7"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "modelslab": {
      "type": "http",
      "url": "https://modelslab.com/mcp/v7"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "modelslab": {
      "serverUrl": "https://modelslab.com/mcp/v7"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "modelslab": {
      "type": "http",
      "url": "https://modelslab.com/mcp/v7"
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Tools this server provides (23)

Taken from the project documentation; the running server may differ.

list-models
Browse available AI models.
list-providers
Browse model providers.
text-to-image
Generate an image from a prompt.
image-to-image
Transform an image.
inpaint-image
Edit part of an image.
fetch-image
Fetch a finished image.
text-to-video
Generate a video from text.
image-to-video
Animate an image.
video-to-video
Transform a video.
lip-sync
Sync lips to audio.
motion-control
Control motion in a video.
fetch-video
Fetch a finished video.
text-to-speech
Convert text to speech.
speech-to-text
Transcribe speech.
speech-to-speech
Convert speech to another voice.
sound-generation
Generate sound effects.
music-generation
Generate music.
song-extender
Extend a song.
song-inpaint
Edit part of a song.
dubbing
Dub audio into another language.
fetch-audio
Fetch finished audio.
chat-completion
Chat with language models.
fetch-generation
Poll a generation job by id.

What it does

Image generation
Text-to-image, image-to-image and inpainting.
Video creation
Text, image and video to video, lip-sync and motion control.
Audio and music
Speech, transcription, sound effects, music, song extension and dubbing.
LLM chat
Runs chat completions on hosted language models.
Model discovery
list-models and list-providers return valid model IDs.
Async job polling
Poll a returned id until generation succeeds or fails.
Separate control plane
A second server manages account, billing and teams.

Good for

  • Generate product photos or concept art from a prompt
  • Make a short video clip from an image
  • Create a voice-over or song with an assistant
  • Dub or lip-sync an existing video
  • Pick the best available model for a task

Pricing

Pricing unknown

Prices not published in the material; API key needed

Prices change. Check docs.modelslab.com before you buy.

ModelsLab FAQ

How do I authenticate?

Sign in with OAuth 2.1, or send your ModelsLab API key as a Bearer or X-API-Key header.

Why does a tool return processing?

Most generation is asynchronous; call fetch-generation with the returned id until it succeeds.

What does the second server do?

The Agent Control Plane manages account, billing, subscriptions and teams.

Which models are available?

The README says 10,000+, including Flux, Qwen Image, Seedream, Kling, Veo and Wan; list-models returns the current set.

How do I add ModelsLab to Claude Code?

Run this command in your terminal: claude mcp add --transport http modelslab https://modelslab.com/mcp/v7

What are the best alternatives to ModelsLab?

The closest alternatives to ModelsLab are HyperFrames by HeyGen, headroom, omniget, next-ai-draw-io and mcp-for-blender.

Where this comes from

  • Description: written from the project's README and the project's website.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers