Qwen Code is an open-source terminal-based AI coding assistant maintained by the Qwen team (Alibaba Cloud / QwenLM). Built on Node.js/TypeScript and drawing architectural lineage from Google's Gemini-CLI, it connects terminal workflows directly with Qwen's suite of code-focused large language models, including Qwen-Coder series models. In daily operations, Qwen Code runs directly in the command line or terminal emulator to assist with automated refactoring, file inspection, bug fixing, and script execution. It parses local repository contexts, applies patches across project files, and integrates with external tooling and skills to execute build commands and tests. What distinguishes Qwen Code is its native alignment with open-weights Qwen-Coder models and flexible self-hosted or API-backed backend endpoints. Developers who require local execution, self-hosted LLM endpoints via vLLM or Ollama, or direct DashScope API integration can leverage a streamlined terminal interface without being locked into proprietary proprietary SaaS agents. Qwen Code is fully open source under the Apache-2.0 license, making it free to inspect, fork, and run locally against custom model deployments or cloud provider tokens.
Terminal-first workflow — Executes multi-file edits, code analysis, and terminal commands directly from the command line. - Qwen-Coder model integration — Tailored to work with QwenLM coding models via DashScope APIs or compatible self-hosted OpenAI-compatible endpoints. - Extensible skill & tool engine — Supports automated tooling integrations, custom agent skills, and session-level command executions. - Multi-platform support — Cross-platform CI coverage supporting Linux, macOS, and Windows environments. - Open-source architecture — Fully inspectable TypeScript/Node.js codebase adaptable to custom private LLM infrastructures.
Use cases
Use Case: Automated Bug Fixing and Refactoring Developers can invoke Qwen Code in their repository to analyze failing tests, review stack traces, and apply multi-file code diffs without leaving the terminal.
Use Case: Repository Exploration and Documentation Engineers onboarding to unfamiliar codebases can query project structure, decipher complex algorithms, and generate inline documentation directly within the CLI environment.
Use Case: Local and Air-Gapped Code Generation Teams running self-hosted Qwen-Coder instances via vLLM, SGLang, or Ollama can route Qwen Code agent calls through local private network endpoints to ensure internal source code privacy.
Who it is for
Qwen Code is aimed at developers and DevOps engineers who prefer lightweight terminal-native agents and want deep integration with the Qwen open-weights model ecosystem. It is well-suited for teams running on-premise or cloud-hosted Qwen-Coder instances, but less ideal for developers looking for full graphical IDE point-and-click extensions without terminal interaction.
Install guide
Installation Install Qwen Code globally via npm or clone and build from the official GitHub repository: bash npm install -g @qwenlm/qwen-code Alternatively, build from source: bash git clone https://github.com/QwenLM/qwen-code.git cd qwen-code npm install npm run build
First Run Configure your Qwen API key (DashScope or OpenAI-compatible endpoint): bash export DASHSCOPE_API_KEY="your_api_key_here" # Or set custom base URL for local inference export OPENAI_BASE_URL="http://localhost:8000/v1" Run the agent initialization: bash qwen-code
Example Session Start an interactive session within your repository to inspect and modify code: bash qwen-code "Review src/index.ts for potential memory leaks and write a unit test in tests/index.test.ts"
VTCode — VTCode is an open-source terminal-based AI coding agent built in Rust by Vinh Nguyen (vinhnx). Engineered as a high-performance CLI…
Nanocoder — Nanocoder is an open-source terminal-based AI coding assistant developed by the Nano-Collective community. Designed for software developers who prefer working…
coro-code — coro-code is an open-source, terminal-based AI coding assistant developed in Rust by Blushyes. Built around a compiled native architecture, it…
Groq Code CLI — Groq Code CLI is an open-source command-line coding assistant developed by the build-with-groq team. It integrates directly with Groq's high-speed…
ob-1 — ob-1 is an open-source terminal-based AI coding agent created by Overbrilliant. Designed for developer workflows in the shell, it orchestrates…
CodeWhale — CodeWhale is an open-source terminal-based AI coding agent developed in Rust by Hunter Bown. It provides an autonomous and semi-autonomous…