MCP server/ai-tools

Crazyrouter MCP Server

Access 627+ AI models through one API key

xujfcn/crazyrouter-mcp ↗by xujfcnupdated
Manual setup required. The maintainer's config contains paths only you know - edit the placeholders below before adding it to Claude Code.
1

Prepare the server locally

Run this once before adding it to Claude Code.

git clone https://github.com/xujfcn/crazyrouter-mcp.git
cd crazyrouter-mcp
npm install
npm run build
2

Register it in Claude Code

claude mcp add -e "CRAZYROUTER_API_KEY=${CRAZYROUTER_API_KEY}" crazyrouter-mcp -- node /path/to/crazyrouter-mcp/dist/index.js

Replace any placeholder paths in the command with the real path on your machine.

Required:CRAZYROUTER_API_KEY
3

Make your agent remember this setup

crazyrouter-mcp's config, env vars, and the gotchas you hit — recalled in every future Claude Code, Cursor, and Codex session.

npx conare@latest

Free · one command · indexes the sessions already on disk. Set up in the browser instead →

What it does

  • Access to 627+ AI models including GPT-5, Claude Opus, and Gemini 3
  • Unified API for text, image, video, audio, and music generation
  • Supports multiple providers like OpenAI, Anthropic, Google, and DeepSeek
  • Global edge nodes for low latency
  • OpenAI, Anthropic, and Gemini protocol compatible

Tools 4

chatChat with any AI model
list_modelsList available models by category
generate_imageGenerate images with AI
generate_videoGenerate videos with AI

Environment Variables

CRAZYROUTER_API_KEYrequiredAPI key obtained from crazyrouter.com

Try it

Use the chat tool to ask GPT-5 to explain quantum computing
Use generate_image to create a sunset over mountains with DALL-E 3
Use generate_video with Kling V2 to create a 5-second clip of ocean waves
Use list_models to show me all available image generation models
Original README from xujfcn/crazyrouter-mcp

Crazyrouter MCP Server — Access 627+ AI Models with One API

One key. 627+ models. Text, image, video, audio, music, 3D — all through MCP.

An MCP (Model Context Protocol) server that gives your AI assistant access to 627+ AI models through Crazyrouter — the unified AI API gateway. Chat with GPT-5, generate images with DALL-E 3, create videos with Sora 2, compose music with Suno V4, and more — all from a single API key.


✨ Features

  • 🌐 627+ Models, One API — GPT-5, Claude Opus 4.6, Gemini 3, DeepSeek R1, Llama 4, Qwen3, Grok 4, and hundreds more
  • 🎨 Image Generation — DALL-E 3, Midjourney, Flux, Stable Diffusion, Nano Banana Pro
  • 🎬 Video Generation — Sora 2, Kling V2, Veo 3, Seedance, Pika
  • 🎵 Music Generation — Suno V4, Chirp
  • 🗣️ Text-to-Speech & Speech-to-Text — Multiple TTS/STT engines
  • 💰 Competitive Pricing — Below official rates for most models
  • 🌍 7 Global Edge Nodes — Low latency worldwide
  • 🔄 OpenAI + Anthropic + Gemini Protocol Compatible — Drop-in replacement
  • Credits Never Expire — Pay once, use anytime

🚀 Quick Start

1. Get Your API Key

Sign up at crazyrouter.com and grab your API key.

2. Install

git clone https://github.com/xujfcn/crazyrouter-mcp.git
cd crazyrouter-mcp
npm install
npm run build

3. Configure Your AI Client

<details> <summary><strong>Claude Desktop</strong></summary>

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "crazyrouter": {
      "command": "node",
      "args": ["/path/to/crazyrouter-mcp/dist/index.js"],
      "env": {
        "CRAZYROUTER_API_KEY": "your-api-key-here"
      }
    }
  }
}
</details> <details> <summary><strong>Cursor</strong></summary>

Edit .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "crazyrouter": {
      "command": "node",
      "args": ["/path/to/crazyrouter-mcp/dist/index.js"],
      "env": {
        "CRAZYROUTER_API_KEY": "your-api-key-here"
      }
    }
  }
}
</details> <details> <summary><strong>VS Code (Copilot)</strong></summary>

Add to your VS Code settings.json:

{
  "mcp": {
    "servers": {
      "crazyrouter": {
        "command": "node",
        "args": ["/path/to/crazyrouter-mcp/dist/index.js"],
        "env": {
          "CRAZYROUTER_API_KEY": "your-api-key-here"
        }
      }
    }
  }
}
</details> <details> <summary><strong>Windsurf</strong></summary>

Edit ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "crazyrouter": {
      "command": "node",
      "args": ["/path/to/crazyrouter-mcp/dist/index.js"],
      "env": {
        "CRAZYROUTER_API_KEY": "your-api-key-here"
      }
    }
  }
}
</details> <details> <summary><strong>OpenClaw</strong></summary>

Add to your OpenClaw MCP config:

{
  "mcpServers": {
    "crazyrouter": {
      "command": "node",
      "args": ["/path/to/crazyrouter-mcp/dist/index.js"],
      "env": {
        "CRAZYROUTER_API_KEY": "your-api-key-here"
      }
    }
  }
}
</details>

🛠️ Available Tools

Tool Description Key Parameters
chat Chat with any AI model model, messages, temperature
list_models List available models by category category (chat/image/video/audio/music)
generate_image Generate images with AI prompt, model, size
generate_video Generate videos with AI prompt, model

Usage Examples

Chat with any model:

"Use the chat tool to ask GPT-5 to explain quantum computing"

Generate an image:

"Use generate_image to create a sunset over mountains with DALL-E 3"

Generate a video:

"Use generate_video with Kling V2 to create a 5-second clip of ocean waves"

Discover models:

"Use list_models to show me all available image generation models"


📋 Supported Models

💬 Chat / Text Models

Provider Models
OpenAI GPT-5, GPT-5-mini, GPT-4.1, GPT-4o, o3, o4-mini
Anthropic Claude Opus 4.6, Claude Sonnet 4, Claude Haiku 3.5
Google Gemini 3 Pro, Gemini 2.5 Flash, Gemini 2.0
DeepSeek DeepSeek R1, DeepSeek V3
Meta Llama

Frequently Asked Questions

What are the key features of Crazyrouter MCP?

Access to 627+ AI models including GPT-5, Claude Opus, and Gemini 3. Unified API for text, image, video, audio, and music generation. Supports multiple providers like OpenAI, Anthropic, Google, and DeepSeek. Global edge nodes for low latency. OpenAI, Anthropic, and Gemini protocol compatible.

What can I use Crazyrouter MCP for?

Integrating diverse AI models into a single development workflow. Generating high-quality AI images and videos directly within an MCP-enabled IDE. Switching between different LLM providers for comparative analysis. Building applications that require multi-modal AI capabilities via one API.

How do I install Crazyrouter MCP?

Install Crazyrouter MCP by running: git clone https://github.com/xujfcn/crazyrouter-mcp.git && cd crazyrouter-mcp && npm install && npm run build

What MCP clients work with Crazyrouter MCP?

Crazyrouter MCP works with any MCP-compatible client including Claude Desktop, Claude Code, Cursor, and other editors with MCP support.

Conare · memory for coding agents

Turn this server into reusable context

Keep Crazyrouter MCP docs, env vars, and workflow notes in Conare so your agent carries them across sessions.

Set up free$npx conare@latest