Qwen Code is an open-source AI coding agent that runs in your terminal, editor, desktop app, browser and chat apps. It works with several model providers and can run headless in scripts and CI.
AgentFree
Step 1: Install Qwen Code with the install script, npm (Node.js 22 or newer) or Homebrew.
Step 2: Open a terminal in your project directory and run qwen.
Step 3: Run /auth and either sign in with a Qwen subscription or configure a provider with an API key.
Step 4: Describe your task in plain language, for example asking it to explain the repository.
Step 5: Approve actions as they come up and follow progress in the session.
Step 6: For automation, run qwen -p with a prompt in scripts, or use an SDK.
Yes. It supports the OpenAI, Anthropic, Gemini and Qwen APIs, any third-party provider, and local models through Ollama or vLLM. You can switch provider at runtime.
Yes. Headless mode runs with qwen -p followed by a prompt, which suits scripts, CI/CD and batch jobs. SDKs for TypeScript, Python and Java are also available.
By default the daemon listens on loopback without bearer authentication, so any local process that can reach it can use the full API. On shared or untrusted hosts, set QWEN_SERVER_TOKEN and start it with --require-auth.
Yes. Connect Telegram, DingTalk, WeChat or Feishu, then run qwen channel start.
Bundled Qwen Code skill for reviewing changed code for correctness, security, quality, and performance.
Bundled Qwen Code helper skill for official usage, configuration, features, and troubleshooting guidance.
SigmanticAISigmanticAI is an AI teammate for chip design teams that generates RTL and verification IP, debugs failures and works with your existing EDA tools. It runs as a command-line agent installed with pip.