Our take
Activeregular commits and releases- Activity
- 91
- Community
- 93
- Issues
- 90
- Pull requests
- 64
+192 stars in the last 10 days
- Releases more often than 93% of the projects we track
- Merges more pull requests than 93% of the projects we track
- Closes issues faster than 82% of the projects we track
Qwen Code is a broad, actively maintained coding-agent framework for teams that value provider choice and multiple ways to work with an agent.
Good fit if
- You want a coding agent that can be used from a terminal, editor, desktop app, browser, or chat.
- You need to work with multiple model APIs or local models rather than commit to one provider.
- Your team is comfortable with Node.js 22 or newer and values an Apache-2.0 licensed project with active development and contributions from a broad contributor base.
Look elsewhere if
- You need a narrowly scoped terminal-only tool rather than a framework with several interfaces and integrations.
- Your environment cannot use Node.js 22 or newer.
- You require a project with a documented test setup in the provided project facts, which do not list tests.
All health signals
| Last commit | 2026-10-08 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 100+ (latest v0.25.0-nightly.20261007.8003d28042, 2026-10-07, pre-release) |
| Contributors | 100+ (top contributor: 15% of commits) |
| Issues closed, last 90 days | 24+ (typically closed in 0 days) |
| Pull requests merged, last 90 days | 88 (typically merged in 1 day) |
| Project age | 16 months |
Checked on 2026-10-09 with the GitHub API.
Overview
Qwen Code is a coding agent that works in a project context and can help with tasks such as understanding a repository, reviewing code, and making changes. It addresses the need to bring AI-assisted coding into developer workflows without tying the agent to a single model provider or interface.
It is aimed at developers and teams who want one agent framework across terminal, editor, desktop, web, chat, and automation workflows. The framework and Qwen models are open source, and the project also supports APIs from other providers and local models.
Key Features
- Agent workflows with Auto-Memory, Auto-Skills, SubAgents, Agent Teams, and MCP support.
- Multiple provider protocols, including OpenAI, Anthropic, Gemini, and Qwen APIs, plus third-party providers and local models.
- Terminal use for interactive coding sessions and headless runs in scripts, CI/CD, or batch jobs.
- Integrations for VS Code, Zed, JetBrains, desktop, browser, Telegram, DingTalk, WeChat, and Feishu.
- TypeScript, Python, and Java SDKs for building custom integrations.
- Experimental daemon mode that connects clients over HTTP and SSE, and an experimental browser UI.
Use Cases
- Developers can ask questions about an unfamiliar repository and get help locating where to start.
- Engineering teams can use a coding agent for code changes, reviews, and tests from a terminal or supported editor.
- Teams with existing model-provider choices can switch among supported APIs or connect a local model at runtime.
- Developers can automate agent tasks in scripts or CI/CD, or embed Qwen Code through an SDK.
- Teams can make coding assistance available through chat tools or a shared daemon, where those integrations suit their workflows.
What you need
Detected in the repository
- Node.js >=22.0.0 (from package.json)
- A Dockerfile, so it can run in a container
- Automated checks on GitHub Actions
License in plain words
Apache-2.0permissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the license, the NOTICE file and a note of your changes
- Share your changes: no
- Includes an explicit patent grant from the contributors:
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Install with npm, then start Qwen Code in your project:
npm install -g @qwen-code/qwen-code@latest
cd /path/to/your-project
qwen
Run /auth in the session to configure a provider and API key. See the README and documentation for other installation options and setup details.
Alternatives
- openinterpreter: Open Interpreter is a Rust terminal agent focused on low-cost models and ACP editor or Codex SDK connections, while Qwen Code spans more interfaces and providers.
- tau: Tau is a Python terminal agent with a deliberately small, layered codebase for learning or customization, unlike Qwen Code's broader multi-interface product.
- zero: Zero is a Go terminal agent emphasizing scriptable use, local sessions, and explicit side-effect controls, while Qwen Code supports additional interfaces.
- kilocode: Kilo Code centers on VS Code and JetBrains integrations with planning and debugging modes, while Qwen Code also targets desktop, browser, and chat use.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| qwen-code | TypeScript | Apache-2.0 | 28.4k | Active |
| openinterpreter | Rust | Apache-2.0 | 68.5k | Active |
| tau | Python | MIT | 2.9k | New |
| zero | Go | MIT | 1.7k | New |
| kilocode | TypeScript | MIT | 27.5k | Active |
Considerations
The project is actively developed, with recent commits and releases and contributions that are not concentrated in one person. That pace is encouraging, though teams should account for ongoing changes when adopting its integrations and workflows. The supplied project facts list CI but do not list tests, so verify the checks and validation relevant to your deployment before relying on it. Node.js 22 or newer is required; Docker is also listed among the project requirements. The README marks the web UI and daemon as experimental. The Apache-2.0 license permits broad reuse, subject to its license terms.
Found this useful?
Share it with someone who would like qwen-code.