Qwen Max MCP Server

Example Use Cases

• 1

  Generate long-form or multi-step text with Qwen models via the MCP protocol and Claude Desktop integration.

• 2

  Access to very large context windows (up to 1,000,000 tokens with Qwen-Turbo or 131,072 with Qwen-Plus) for document-level reasoning or long conversation history.

• 3

  Configurable generation controls (max_tokens, temperature) and robust error handling when calling Qwen model endpoints through a local MCP server.

Description

This repository implements a Model Context Protocol (MCP) server in Node.js/TypeScript to proxy requests to Alibaba's Qwen models (Max, Plus, Turbo) and integrate them with Claude Desktop. It supports configurable parameters (max_tokens, temperature), error handling, and large context windows across available Qwen models. The server requires a Dashscope API key to call the model endpoints and is intended to be run locally or deployed alongside Claude Desktop. It is an unofficial community implementation focused on reliability and type-safe integration.