Qwen Code
» AI Agents » Ready-to-Use » Qwen Code URL Share on
Open-source terminal AI coding agent built around Qwen3-Coder models, with auto-memory, skills, subagents, agent teams, and multi-protocol LLM support
Key Features
Multi-Agent Single Agent Human-in-the-Loop Streaming Async Support Type Safe Short-Term Memory Long-Term Memory Shared Memory Plugin System Custom Tools MCP Protocol A2A Protocol Code Execution Web Browsing File System Access Sandboxing CLI API Server Self-Hosted
Community Feedback
Strengths
- Multi-protocol — works with OpenAI, Anthropic, Gemini, and Qwen APIs, plus local models
- Rich built-in feature set: SubAgents, Agent Teams, Auto-Memory, Auto-Skills, MCP
- Beyond-terminal reach: IDE plugins, desktop app, daemon mode, SDKs, IM bots
- Actively self-improving — project uses its own agent to file issues, review PRs, and run autofixes
Weaknesses
- Free OAuth tier discontinued April 2026, following a gradual quota reduction
- Now requires a paid Alibaba Cloud plan or third-party key for meaningful usage
- Reports of Qwen team shifting toward more proprietary development post-leadership changes
Qwen Code Details
| Organization | Alibaba Cloud / Qwen Team |
| Organization Type | Company |
| Funding | Public |
| Category | Ready-to-Use |
| Subcategory | Code generation |
| Deployment | Hybrid |
| Primary Language | TypeScript |
| Runtime | Node.js 22+ |
| License | Apache-2.0 |
| Commercial Use | Permissive |
| Install Command | npm install -g @qwen-code/qwen-code@latest |
| GitHub Stars | 26,261 |
| GitHub Forks | 2,708 |
| Open Issues | 707 |
| Contributors | 418 |
| Last Commit | 2026-07-23 |
| First Release | 2025-07-23 |
| Release Cadence | Weekly point releases |
| Maturity | Stable |
| Pricing Model | Usage based |
| Free Tier | No hosted free tier as of 2026-04-15 (Qwen OAuth free access was discontinued); OpenRouter offers a separate free-hosted Qwen3-Coder model usable with this CLI via custom provider config |
| Self-Hosted Free | Yes |
| Cost Model | free CLI; OAuth quota ended Apr 2026, BYO key |
| Community Size | Large (26K+ stars, 418 contributors) |
| Community Activity | Very active |
| Sentiment | Mixed |
| GPU Required | No |
| Research Date | 2026-07-23 |
| LLM Providers | Qwen (Qwen3-Coder, via Alibaba Cloud Model Studio / DashScope), OpenAI, Anthropic Claude, Google Gemini, Local models (Ollama, vLLM) and any third-party provider |
| API Keys Required | Alibaba Cloud Model Studio API key, or an OpenAI/Anthropic/Gemini-compatible key (Qwen OAuth free tier was discontinued 2026-04-15) |
Use Cases
- Terminal-based multi-file code generation and editing
- Headless/scripted coding tasks (qwen -p "...")
- SubAgents and Agent Teams for complex, parallelizable engineering work
- IM-bot-driven coding workflows (Telegram, DingTalk, WeChat, Feishu, WeCom, QQ)
- IDE plugin and desktop-app-based agentic coding
When to Use
Best for: Teams wanting an open-source, multi-provider terminal coding agent with deep IM/bot integrations and Qwen3-Coder support
Avoid when: You want a permanently free hosted tier — Qwen Code's free OAuth access ended in April 2026
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a