Qwen Code

Qwen Code is an open source AI coding agent from the Qwen team at Alibaba. It runs in the terminal and reads your codebase, edits files, runs shell commands, and works through multi-step tasks on your local repository.

The project started from Google's Gemini CLI v0.8.2. Since v0.1 it has been developed independently, with its own features and support for many model providers.

Key capabilities:

• Multi-provider models: works with OpenAI, Anthropic, Gemini and Qwen APIs, any compatible third-party provider, and local models through Ollama or vLLM. You can switch models at runtime.
• Agent features: subagents, agent teams, skills, auto-memory and MCP server support.
• More than a terminal: IDE plugins for VS Code, Zed and JetBrains, a desktop app, a web UI, and SDKs for building on top of the agent.
• Chat integrations: run the agent from Telegram, DingTalk, WeChat or Feishu.

Qwen Code is optimized for Qwen models, and many of those are open-weight, so you can run both the agent and the model on your own hardware. It is a good fit for developers who want a Claude Code alternative without being tied to one model vendor.

Install it with the standalone script, npm (@qwen-code/qwen-code, Node.js 22+) or Homebrew.