Qwen Code

An open-source AI coding agent for macOS, Windows and Linux that supports local models through Ollama or vLLM and cloud model APIs.

Alternative to:

Qwen Code is an AI coding assistant for developers who want an agent to review code, fix bugs and carry out work in a project. You can use it in a terminal, editor, desktop app or browser.

Model choice: Use locally hosted models through Ollama or vLLM, or connect to Qwen, OpenAI, Anthropic and Gemini APIs, with provider switching during a session.
Coding tasks: Built-in skills cover reviews, bug fixes and batch work, while a planning mode lets you work through proposed changes before execution.
Agent collaboration: Subagents, agent teams and dynamic workflows support dividing tasks among agents; automatic memory and skills are also available.
Tools and isolation: MCP connects external tools, sandboxing and Git worktrees support isolated coding work, and computer use supports desktop automation.
Model comparison: Agent Arena lets you compare different models working on the same task.

Editor integrations include VS Code, JetBrains and Zed. Headless operation and SDKs support scripted workflows, while Telegram, DingTalk, WeChat and Feishu integrations provide access through chat. A shared agent service can serve multiple clients.

Qwen Code runs on macOS, Windows and Linux and is open source under Apache 2.0. The agent runs on your machine; model requests go to the backend you choose. Ollama and vLLM support local model hosting, while cloud APIs send requests to their providers.