Remote script. The standalone installer pipes a script from an Aliyun storage bucket into your shell; npm or Homebrew avoid that.
Who it is forA developer who likes the Claude Code workflow but wants open source and freedom to switch models.
The problemMost terminal coding agents tie you to one vendor's models and pricing.
The outcomeA terminal agent working in your project with the provider of your choice, switchable at runtime.
What you need first
- Linux, macOS or Windows
- The standalone installer, or Node.js 22 or later for npm, or Homebrew
- An API key for a supported provider, or a local model via Ollama or vLLM
- Optional: VS Code, Zed or JetBrains for editor integration
Steps
- Install with
npm install -g @qwen-code/qwen-code@latest(Node.js 22+) orbrew install qwen-code - Restart your terminal, then run
cd /path/to/your-projectand start the agent with qwen - Inside the session run /auth to configure your provider and API key
- Try the README's first prompt: Explain this repository and show me where to start
- See the Authentication Guide to use a local model through Ollama or vLLM
Prompt for your agent
Install QwenLM/qwen-code, start it in this project with qwen, configure <provider> with /auth, then explain this repository and propose a plan for <task> in Plan Mode before changing any files.
Caveat. Model costs depend on the provider you choose; a local model avoids API bills but needs capable hardware.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for QwenLM/qwen-code and "Run an open-source coding agent with any model"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer