Qwen Code is an open-source AI coding agent that lives in your terminal, maintained by the Qwen team at Alibaba. It reads your codebase, edits files, runs commands, and works through programming tasks in an agentic loop while you supervise from the command line. The project has earned over 28,000 stars on GitHub and was actively updated in October 2026.
Qwen Code is optimized for the Qwen family of models, especially Qwen3-Coder, and gives you a terminal-native workflow similar to other agentic coding CLIs. You describe a task, and the agent plans, explores the repository, makes changes across multiple files, and verifies its work by running tests and commands. It supports MCP servers for extending its toolset, works across platforms, and can be driven interactively or folded into scripted workflows. Because it is open source under Apache-2.0, teams can inspect, configure, and self-host the whole toolchain.
Qwen Code itself is free and open source. You only need access to a model, which can be a hosted Qwen API, another compatible provider, or a local deployment, depending on how you configure it.
Qwen Code is for developers who want an agentic coding partner in the terminal, especially those already invested in the Qwen model ecosystem. It suits engineers who prefer open tools they can audit and extend over closed, subscription-only coding assistants.