Any Chat Completions MCP Server

Integrate Claude with any OpenAI SDK compatible chat completion API. Use multiple LLMs and agents at once - OpenAI, Perplexity, Groq, xAI, PyroPrompts and more.

{
  "mcpServers": {
    "chat-openai": {
      "command": "npx",
      "args": ["@pyroprompts/any-chat-completions-mcp"],
      "env": {
        "AI_CHAT_KEY": "your-openai-api-key",
        "AI_CHAT_NAME": "OpenAI",
        "AI_CHAT_MODEL": "gpt-4o",
        "AI_CHAT_BASE_URL": "https://api.openai.com/v1"
      }
    }
  }
}

What is MCP?

The Model Context Protocol (MCP) is an open standard that enables AI assistants like Claude to securely connect with external data sources and tools. This MCP server acts as a bridge, allowing Claude to communicate with any OpenAI-compatible chat completion API.

Key Features

• Universal Compatibility - Works with any OpenAI SDK compatible API
• Multiple LLM Support - Use OpenAI, Perplexity, Groq, xAI, PyroPrompts simultaneously
• Easy Configuration - Simple environment variable setup
• Claude Desktop Integration - Seamless integration with Claude Desktop
• LibreChat Support - Also works with LibreChat
• Simple Setup - Easy NPX installation with configuration files

Supported Providers

Installation Methods

Method 1: Manual NPX Installation (Recommended)

Add this configuration to your Claude Desktop config file:

{
  "mcpServers": {
    "chat-openai": {
      "command": "npx",
      "args": [
        "@pyroprompts/any-chat-completions-mcp"
      ],
      "env": {
        "AI_CHAT_KEY": "your-openai-api-key",
        "AI_CHAT_NAME": "OpenAI",
        "AI_CHAT_MODEL": "gpt-4o",
        "AI_CHAT_BASE_URL": "https://api.openai.com/v1"
      }
    }
  }
}

Method 2: Smithery

Alternatively, you can use Smithery for automatic configuration:

npx -y @smithery/cli install any-chat-completions-mcp-server --client claude

Multiple Provider Setup

You can configure multiple LLM providers simultaneously by adding multiple server configurations. This allows you to use different models for different tasks within the same Claude conversation:

{
  "mcpServers": {
    "chat-openai": {
      "command": "npx",
      "args": ["@pyroprompts/any-chat-completions-mcp"],
      "env": {
        "AI_CHAT_KEY": "your-openai-key",
        "AI_CHAT_NAME": "OpenAI",
        "AI_CHAT_MODEL": "gpt-4o",
        "AI_CHAT_BASE_URL": "https://api.openai.com/v1"
      }
    },
    "chat-perplexity": {
      "command": "npx",
      "args": ["@pyroprompts/any-chat-completions-mcp"],
      "env": {
        "AI_CHAT_KEY": "your-perplexity-key",
        "AI_CHAT_NAME": "Perplexity",
        "AI_CHAT_MODEL": "sonar",
        "AI_CHAT_BASE_URL": "https://api.perplexity.ai"
      }
    },
    "chat-groq": {
      "command": "npx",
      "args": ["@pyroprompts/any-chat-completions-mcp"],
      "env": {
        "AI_CHAT_KEY": "your-groq-key",
        "AI_CHAT_NAME": "Groq",
        "AI_CHAT_MODEL": "llama-3.1-70b-versatile",
        "AI_CHAT_BASE_URL": "https://api.groq.com/openai/v1"
      }
    }
  }
}

How It Works

Once configured, you'll see chat tools for each provider in Claude Desktop. You can then:

• Ask Claude to consult other LLMs - "Ask OpenAI what it thinks about this code"
• Compare responses - Get different perspectives from multiple models
• Use specialized models - Use Perplexity for research, Groq for speed, etc.
• Leverage unique capabilities - Each model has different strengths

Use Cases

GitHub Repository

The complete source code, documentation, and examples are available on GitHub:

LibreChat Integration

This MCP server also works with LibreChat. Add this configuration to your LibreChat MCP settings:

chat-perplexity:
  type: stdio
  command: npx
  args:
    - -y
    - @pyroprompts/any-chat-completions-mcp
  env:
    AI_CHAT_KEY: "your-perplexity-key"
    AI_CHAT_NAME: Perplexity
    AI_CHAT_MODEL: sonar
    AI_CHAT_BASE_URL: "https://api.perplexity.ai"
    PATH: '/usr/local/bin:/usr/bin:/bin'

Development

Want to contribute or modify the server? Here's how to get started with development:

git clone https://github.com/ferrants/any-chat-completions-mcp.git

cd any-chat-completions-mcp

npm install

npm run build

For development with auto-rebuild:

npm run watch

Debugging

Since MCP servers communicate over stdio, debugging can be challenging. Use the MCP Inspector for debugging:

npm run inspector

Getting Started

1. Choose your installation method: Manual NPX configuration (recommended) or Smithery
2. Get API keys for the providers you want to use
3. Configure Claude Desktop with your provider settings
4. Restart Claude Desktop to load the new MCP server
5. Start chatting with multiple LLMs through Claude!

This MCP server opens up powerful possibilities for multi-model AI workflows. Check out the GitHub repository for detailed documentation, examples, and contribution guidelines.