ModelRunner MCP server
Commit Sep 7, 2026Run over 100 AI models for image, video, audio, and 3D generation through a single API connection.
Also known as modelrunner/modelrunner-mcp, modelrunner-mcp
- Stars
- 0
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 1.5.0
- Last commit
- Sep 7, 2026
- License
- MIT
What is ModelRunner MCP?
ModelRunner runs over 100 AI models for image, video, audio, and 3D generation through a single API connection.
Summary generated from the project's README
About ModelRunner MCP server
Generating images, videos, audio, and 3D assets from within your AI assistant is what ModelRunner enables. This remote MCP server connects Claude, Cursor, and other clients to a massive catalog of over 100 models from various providers. Instead of managing multiple individual subscriptions, you use a single pay-per-use account to access diverse capabilities like text-to-video, music generation, and speech-to-text.
Setup is handled through OAuth 2.1, meaning you authorize via a browser window rather than pasting API keys into your configuration files. The server is hosted remotely at a specific endpoint, so there is nothing to install or self-host locally. You can use it directly in Claude Code, Cursor, or via mcp-remote for desktop clients.
Developers can use tools like search, list_models, and run_model to interact with the catalog. The system also supports creating wrappers, which are simplified interfaces built over base models using prompt templates and specific settings. Results are returned directly to your conversation as hosted URLs.
What you can do with it
- 1Explore available generative modelsAsk your assistant to find specific AI models for a creative task by searching the catalog.
- 2Generate multimedia assetsInstruct your AI client to create a video or an image using specific models like Kling or Stable Audio.
- 3Build custom model interfacesCreate simplified reusable interfaces called wrappers to standardize how you interact with complex base models.
Written from the project's README. Every use case is checked against a quote from it.
How to install ModelRunner MCP
Formats verified 10 Nov 2026Install ModelRunner MCP in Claude Code
$ claude mcp add --transport http modelrunner-mcp https://mcp.modelrunner.run/mcpInstall ModelRunner MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"modelrunner-mcp": {
"url": "https://mcp.modelrunner.run/mcp"
}
}
}Install ModelRunner MCP in VS Code
.vscode/mcp.json
{
"servers": {
"modelrunner-mcp": {
"type": "http",
"url": "https://mcp.modelrunner.run/mcp"
}
}
}Install ModelRunner MCP in Codex
~/.codex/config.toml
[mcp_servers.modelrunner-mcp] url = "https://mcp.modelrunner.run/mcp"
Install ModelRunner MCP in Gemini CLI
~/.gemini/settings.json
{
"mcpServers": {
"modelrunner-mcp": {
"httpUrl": "https://mcp.modelrunner.run/mcp"
}
}
}README
View on GitHubModelRunner MCP Server
Official Model Context Protocol (MCP) server for ModelRunner — run 100+ AI models across image, video, audio, and 3D generation from Claude, Cursor, and any MCP client.
- Endpoint:
https://mcp.modelrunner.run/mcp(remote server, Streamable HTTP) - Auth: OAuth 2.1 — your browser authorizes once; no API key is pasted into config
- Registry name:
ai.modelrunner/mcp - Setup guide & full tool reference: modelrunner.ai/mcp
Unlike single-vendor MCP servers, one connection exposes the entire ModelRunner catalog — models from multiple AI providers behind one endpoint and one pay-per-use billing account. Your assistant can search the catalog, inspect a model's input schema, run inference, and get results back as hosted URLs directly in the conversation.
Maintaining an open-source project? ModelRunner gives active open-source projects free monthly API credits — apply at modelrunner.ai/oss-program.
Quick start
Claude Code
claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp
Cursor / VS Code (native remote URL)
{
"mcpServers": {
"modelrunner": { "url": "https://mcp.modelrunner.run/mcp" }
}
}
Claude Desktop or any stdio-only client (via mcp-remote)
{
"mcpServers": {
"modelrunner": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
}
}
}
On first use your AI assistant opens a browser window to sign in to ModelRunner — that's the whole setup.
Tools
| Group | Tools |
|---|---|
| Discover & inspect | search, list_models, list_wrappers, recommended_models, recommended_wrappers, get_model, get_wrapper, get_model_raw_schema, get_wrapper_raw_schema, get_wrapper_prompt_template, get_wrapper_field_mappings |
| Run & fetch | run_model, wait_for_request, get_request, list_my_requests |
| Upload files | create_upload_url, upload_file |
| Author wrappers | wrapper_authoring_guide, create_wrapper, patch_wrapper, preview_wrapper, list_my_wrappers, delete_wrapper |
"Wrappers" are reusable, simplified interfaces you can build over base models — a few friendly inputs, a prompt template, and baked-in settings — and they run through the same run_model-style tools.
Models & capabilities
The catalog covers text-to-image, image-to-image, text-to-video, image-to-video, video-to-video, music generation, speech-to-text, and image-to-3D, including models like Kling, HiDream, Hunyuan Image, Stable Audio, and Rodin. Browse by capability:
- Text-to-Image API · Image-to-Image API
- Text-to-Video API · Image-to-Video API · Video-to-Video API
- Music Generation API · Speech-to-Text API
- Image-to-3D API
Full catalog: modelrunner.ai/models
Pricing
Free to connect. Model runs are pay-per-use from prepaid credits — no subscription. Every model's exact price (per image, per second of video, per megapixel, …) is shown on its model page and on the pricing page.
About this repository
This repository documents the hosted ModelRunner MCP server. The server itself is first-party, run by ModelRunner as a remote MCP endpoint — there is nothing to install or self-host; the configs above are the entire integration.
- Landing page: modelrunner.ai/mcp
- Docs: modelrunner.ai/docs
- Model catalog: modelrunner.ai/models
License
ModelRunner MCP questions
How do I add ModelRunner MCP to Claude Code?
Run claude mcp add --transport http modelrunner-mcp https://mcp.modelrunner.run/mcp in your terminal. Add --scope user to make it available in every project.
How do I add ModelRunner MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How do I authenticate with ModelRunner?
The server uses OAuth 2.1, meaning you authorize via your browser once and do not need to paste an API key into your configuration.
What kind of AI models are available?
The catalog includes models for text-to-image, image-to-image, text-to-video, image-to-video, video-to-video, music generation, speech-to-text, and image-to-3D.
Is there a subscription fee for using the models?
No, there is no subscription; model runs are pay-per-use based on prepaid credits.
Do I need to host the server myself?
No, the server is a remote endpoint provided by ModelRunner, so there is nothing to install or self-host.