amidabuddha/unichat-mcp-server

Overview

This Python server implements the Model Context Protocol (MCP) to send requests to multiple AI vendors including OpenAI, MistralAI, Anthropic, xAI, Google AI and DeepSeek. It supports both tool-based and predefined prompt workflows. Each vendor requires its own API key.

Best for

Best for
Developers building MCP-based applications that need to access multiple AI models

Use cases

• Routing AI requests from a single MCP client to different providers
• Switching between AI models without modifying client code
• Using predefined prompts for consistent tasks across vendors

How to use

Install

npx -y @smithery/cli install unichat-mcp-server --client claude

Tested with

Claude Desktop, ChatGPT

Example client config

{\n  "unichat-mcp-server": {\n    "command": "uv",\n    "args": [\n      "--directory",\n      "{{your source code local directory}}/unichat-mcp-server",\n      "run",\n      "unichat-mcp-server"\n    ],\n    "env": {\n      "UNICHAT_MODEL": "SELECTED_UNICHAT_MODEL",\n      "UNICHAT_API_KEY": "YOUR_UNICHAT_API_KEY"\n    }\n  }\n}

Notes

This Python server implements the Model Context Protocol (MCP) to send requests to multiple AI vendors including OpenAI, MistralAI, Anthropic, xAI, Google AI and DeepSeek. It supports both tool-based and predefined prompt workflows. Each vendor requires its own API key.

38 stars on GitHub. Last updated 2026-03-30. Licensed MIT.

Use cases

• Routing AI requests from a single MCP client to different providers
• Switching between AI models without modifying client code
• Using predefined prompts for consistent tasks across vendors

Pros

• Supports six major AI providers in one server
• Standard MCP protocol enables easy integration with MCP clients
• Written in Python, widely accessible to developers

Cons

• Requires separate API keys for each vendor
• Low GitHub star count (38) suggests limited community and testing
• May lack advanced features found in vendor-specific SDKs

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.