Skip to main content
Glama

OpenRouter MCP Server

by slyfox1186
GitHub
Python
Apache 2.0
27
OverviewInspectNewEndpointsSchemaRelated ServersReviewsScore
Need Help?View Source CodeReport Issue

OpenRouter MCP Server

Use 400+ AI models (Claude, GPT-4, Gemini, etc.) directly in Claude Code

🚀 Quick Start

Get started in 4 simple steps:

Step 1: Clone and Setup

git clone https://github.com/slyfox1186/claude-code-openrouter.git cd claude-code-openrouter

Step 2: Configure API Key

Create a .env file with your OpenRouter API key:

cat > .env << 'EOF' # OpenRouter API Configuration OPENROUTER_API_KEY=sk-or-v1-your_actual_api_key_here OPENROUTER_BASE_URL=https://openrouter.ai/api/v1 # Default Model Settings DEFAULT_MODEL=z-ai/glm-4.5 DEFAULT_TEMPERATURE=0.7 DEFAULT_MAX_TOKENS=1048576 # Tool Configuration ENABLE_WEB_SEARCH=true FORCE_INTERNET_SEARCH=true # Logging Configuration LOG_LEVEL=INFO LOG_FILE=openrouter_mcp.log # Optional: Rate limiting RATE_LIMIT_REQUESTS_PER_MINUTE=60 # Docker Compose Bake delegation for better build performance COMPOSE_BAKE=true EOF

Replace sk-or-v1-your_actual_api_key_here with your actual OpenRouter API key!

Step 3: Build and Start Container

# Make scripts executable chmod +x tools/docker_manager.py add_mcp.sh # Build and start the container python3 tools/docker_manager.py build python3 tools/docker_manager.py start

Step 4: Connect to Claude Code

# Add MCP server to Claude Code ./add_mcp.sh

Done! Now you can use any OpenRouter model in Claude Code with large file support.

Step 5: Install Claude Code Command (Optional)

For enhanced multi-model collaboration, copy the command script:

# Create Claude commands directory if it doesn't exist mkdir -p ~/.claude/commands # Copy the multi-model protocol command cp examples/call-openrouter.md ~/.claude/commands/ # Now you can use '/call-openrouter' in Claude Code for collaborative workflows

🔑 Get Your API Key

  1. Go to OpenRouter.ai
  2. Sign up and get an API key
  3. Add it to your .env file:
    OPENROUTER_API_KEY=your_key_here

🎯 What You Get

  • 400+ AI Models: DeepSeek R1, GPT-4, Gemini Pro Preview, Kimi K2, Grok 4, and hundreds more
  • Large File Support: Send multiple large files (1,048,576 token limit) without errors
  • Web Search Integration: Gemini automatically searches the internet for current information
  • Conversation Memory: Continue conversations across multiple requests with full context
  • Model Switching: Change models mid-conversation seamlessly
  • Multi-Model Collaboration: Use /call-openrouter command for structured model collaboration workflows
  • Easy Management: Interactive Docker manager for build/start/logs/shell
  • No Duplicates: Single persistent container (no more container proliferation)

💬 Usage Examples

Chat with different models:

# Use DeepSeek R1 (default) - great for reasoning and analysis openrouter-docker - chat (prompt: "Explain quantum computing concepts") # Use DeepSeek Chat v3.1 (deepseek/deepseek-chat-v3.1) - latest version with 163K context openrouter-docker - chat (model: "deepseek-v3.1", prompt: "Solve this complex problem step by step") # Use Gemini 2.5 Pro Preview with automatic web search for current info openrouter-docker - chat (model: "gemini", prompt: "What are the latest AI developments in 2025?") # Use Kimi K2 for advanced reasoning and programming tasks openrouter-docker - chat (model: "kimi", prompt: "Write a Python function to sort a dictionary by values") # Use Kimi K2 by Moonshot AI for advanced reasoning openrouter-docker - chat (model: "kimi", prompt: "Analyze this complex system architecture") # Use Grok 4 by X.AI for creative and analytical tasks openrouter-docker - chat (model: "grok", prompt: "Help me brainstorm innovative solutions") # Use GPT-5 for latest flagship performance with 400K context window openrouter-docker - chat (model: "gpt-5", prompt: "Analyze this complex business strategy") # Continue previous conversation with context openrouter-docker - chat (continuation_id: "uuid-from-previous", prompt: "Can you elaborate on that?")

Attach files for analysis:

# Send multiple large code files to any model (1M+ tokens supported!) openrouter-docker - chat (model: "gemini", files: ["/path/to/main.py", "/path/to/config.json"], prompt: "Review this codebase for security vulnerabilities") # Analyze documentation with web search for current best practices openrouter-docker - chat (model: "gemini", files: ["/path/to/README.md"], prompt: "Summarize this project and compare with 2025 industry standards") # Programming help with Kimi K2 openrouter-docker - chat (model: "kimi", files: ["/path/to/broken_script.py"], prompt: "Debug this code and suggest improvements") # Control web search behavior manually when needed openrouter-docker - chat (model: "gemini", force_internet_search: false, prompt: "Explain basic programming concepts without external references")

Multi-Model Collaboration:

# Use the collaborative workflow command in Claude Code /call-openrouter # This initiates a structured workflow: # 1. Gemini Pro 2.5 creates initial proposal # 2. DeepSeek R1 refines and improves the plan # 3. Kimi K2 provides tie-breaking if needed # 4. Automatic implementation of the agreed solution

🤖 Available Models

Just use simple names:

  • gemini → Google Gemini 2.5 Pro Preview (with web search)
  • deepseek → DeepSeek R1 (reasoning & analysis)
  • deepseek-v3.1 → DeepSeek Chat v3.1 (latest version with 163K context)
  • kimi → Moonshot Kimi K2 (advanced reasoning)
  • grok → X.AI Grok 4 (creative & analytical)
  • gpt-5 → OpenAI GPT-5 (latest flagship model with 400K context)
  • Plus 400+ other models available by full name

🛠️ Management Commands

Check status:

python3 tools/docker_manager.py status

View logs:

python3 tools/docker_manager.py logs

Restart container:

python3 tools/docker_manager.py restart

Interactive shell:

python3 tools/docker_manager.py shell

Manual Docker commands:

# Check container status docker ps | grep openrouter # View logs directly docker logs openrouter # Manual restart docker restart openrouter

⚠️ Troubleshooting

Container not running?

python3 tools/docker_manager.py restart

MCP connection issues?

claude mcp remove openrouter-docker ./add_mcp.sh

Build issues?

python3 tools/docker_manager.py stop python3 tools/docker_manager.py build python3 tools/docker_manager.py start

Large file attachment errors (400 Bad Request)?

  • Fixed with 1,048,576 token limit (1M+ tokens supported)
  • Check logs: python3 tools/docker_manager.py logs
  • Verify container is running: python3 tools/docker_manager.py status

Still having issues?

  • Check your API key in .env file
  • Make sure Docker is running and accessible
  • Run interactive mode: python3 tools/docker_manager.py
  • Open an issue

📄 License

Apache 2.0 License - see LICENSE

Made with ❤️ for Claude Code users

This server cannot be installed

-
security - not tested
A
license - permissive license
-
quality - not tested

How are these scores calculated?

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

A Model Context Protocol server that provides unified access to 400+ AI models from 30+ providers through OpenRouter's API, enabling seamless integration with Claude Code.

  1. 🚀 Quick Start
    1. Step 1: Clone and Setup
    2. Step 2: Configure API Key
    3. Step 3: Build and Start Container
    4. Step 4: Connect to Claude Code
    5. Step 5: Install Claude Code Command (Optional)
  2. 🔑 Get Your API Key
    1. 🎯 What You Get
      1. 💬 Usage Examples
        1. 🤖 Available Models
          1. 🛠️ Management Commands
            1. ⚠️ Troubleshooting
              1. 📄 License

                Related MCP Servers

                • OpenAI Agents MCP Server

                  lroolle
                  -
                  security
                  F
                  license
                  -
                  quality
                  A Model Context Protocol server that enables Claude users to access specialized OpenAI agents (web search, file search, computer actions) and a multi-agent orchestrator through the MCP protocol.
                  Last updated -
                  9
                  • Linux
                  • Apple

                • Zen MCP Server

                  BeehiveInnovations
                  -
                  security
                  F
                  license
                  -
                  quality
                  A Model Context Protocol server that gives Claude access to multiple AI models (Gemini, OpenAI, OpenRouter) for enhanced code analysis, problem-solving, and collaborative development through AI orchestration with conversations that continue across tasks.
                  Last updated -
                  6,861
                  • Apple

                • MCP AI Bridge

                  fakoli
                  A
                  security
                  A
                  license
                  A
                  quality
                  A secure Model Context Protocol server that enables Claude Code to connect with OpenAI and Google Gemini models, allowing users to query multiple AI providers through a standardized interface.
                  Last updated -
                  3
                  2
                  MIT License

                • Zen MCP Enhanced

                  199-mcp
                  -
                  security
                  F
                  license
                  -
                  quality
                  An enhanced Model Context Protocol server that enables Claude to seamlessly collaborate with multiple AI models (Gemini, OpenAI, local models) for code analysis and development tasks, maintaining context across conversations.
                  Last updated -
                  51
                  31
                  • Apple
                  • Linux

                View all related MCP servers

                New MCP Servers

                MCP directory API

                We provide all the information about MCP servers via our MCP API.

                curl -X GET 'https://glama.ai/api/mcp/v1/servers/slyfox1186/claude-code-openrouter'

                If you have feedback or need assistance with the MCP directory API, please join our Discord server