# Qwen Code (slim)

> Open-source coding agent from Alibaba's Qwen team for the terminal, with desktop, browser and editor interfaces. It runs Qwen, OpenAI, Anthropic and Gemini-compatible models or a local one, and runs headless with qwen -p.

- Full: https://www.anchorterminal.com/tools/qwen-code.md (~6,850 tokens) · this version ~1,430 tokens · JSON https://www.anchorterminal.com/tools/qwen-code.json · canonical https://www.anchorterminal.com/tools/qwen-code
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-08

**BB · 72.4/100 · rank #93 of 722 · #3 in Agent harnesses · agent-ready · confidence medium**


Assessment: Apache-2.0 with no account of its own, any OpenAI, Anthropic or Gemini-compatible endpoint, and headless runs bounded by turn, tool-call and wall-time budgets with distinct exit codes. Out of the box, Auto mode lets an LLM classifier approve tool calls, with the sandbox and folder trust both off. Usage statistics are on by default.

## Facts

- Kind: Agent harness · vendor: Alibaba (Qwen team) · category: Agent harnesses · legal entity: Qwen team, Alibaba Group (no legal entity is named in the repository or the docs) · provenance 48/100
- Packages: npm `@qwen-code/qwen-code`, npm `@qwen-code/sdk`, pypi `qwen-code-sdk`, oci `ghcr.io/qwenlm/qwen-code`
- Auth: API key · pricing: Free · x402: no · licence: Apache-2.0
- Probe metrics: not measured yet (probes haven't run)
- Interfaces: Terminal CLI, desktop app (macOS, Windows, Linux), `qwen serve` daemon and web UI (experimental), VS Code, Zed and JetBrains, chat channels, SDKs for TypeScript, Python and Java
- Install: npm (Node 22 or newer), Homebrew, a standalone install script. Stable, preview and nightly channels
- Models: Qwen, OpenAI, Anthropic and Gemini protocols. Alibaba Cloud Model Studio, DeepSeek, OpenRouter and other listed providers, Vertex AI, or a custom or local endpoint
- Approval modes: plan (read-only), default (ask), auto-edit, auto (LLM classifier, the settings default) and yolo. `permissions.allow`, `ask` and `deny` rules
- Sandbox: Off unless enabled. Linux tool sandbox with Bubblewrap or Landlock and `network` open or closed, macOS Seatbelt (default profile permissive-open allows network), Docker or Podman
- Folder trust: Disabled by default (`security.folderTrust.enabled`)
- MCP client: stdio, SSE and streamable HTTP, OAuth 2.0, per-server `trust`, `includeTools` and `excludeTools`
- Headless: `qwen -p` with text, json or stream-json output, `--continue` and `--resume`, `--json-schema`. Exit codes 41, 42, 44, 52, 53 (turn limit), 55 (budget) and 130
- Run budgets: `--max-session-turns`, `--max-wall-time` and `--max-tool-calls`, each unlimited by default
- Telemetry: Usage statistics on by default, off with `privacy.usageStatisticsEnabled` or `QWEN_USAGE_STATISTICS_ENABLED`. OpenTelemetry off by default, with prompts logged by default once on
- CI: GitHub Action qwen-code-action
- Releases in 90 days: 39 stable (10 July to 8 October 2026)

- Scores: Reliability 69, Performance pending, Schema & documentation 91, Agent ergonomics 85, Security & auth 53, Payments & pricing 60, Task success pending, Maintenance & community 88, Transparency & trust 63 · total over the 7 assessed categories
- Why: Reliability, Local-package reading. · Schema & documentation, Framework reading. · Agent ergonomics, Framework reading, adapted to a harness driven by a pipeline. · Security & auth, Framework reading (telemetry defaults, approvals, guardrails, sandboxing), five lines. · Payments & pricing, Self-hosted rule. · Maintenance & community, 0.25.0 on 5 October 2026 (30). · Transparency & trust, Apache-2.0 (30).
- Sources: 20, open questions: 8, both in the full twin
- Capabilities: agent.harness, agent.mcp-client, agent.multi-agent
- JSON: https://www.anchorterminal.com/api/v1/tools/qwen-code.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/qwen-code.svg` or a link to https://www.anchorterminal.com/tools/qwen-code from a page on qwenlm.github.io or one of its subdomains, or the README of github.com/QwenLM/qwen-code, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Pass `--approval-mode` on every run. The settings default is `auto`, and one docs page says headless runs default to asking, so don't rely on either
2. Add `--max-session-turns`, `--max-wall-time` and `--max-tool-calls`. All three are unlimited by default. Exit 53 is the turn limit and 55 a budget
3. `--yolo` doesn't turn on a sandbox. Add `--sandbox`, or on Linux set `tools.executionSandbox` with `network` set to `closed`
4. Set `QWEN_USAGE_STATISTICS_ENABLED=false` to stop usage statistics
5. Authenticate with `OPENAI_API_KEY`, `OPENAI_BASE_URL` and `OPENAI_MODEL` in CI. The browser sign-in is discontinued

## Connect

```bash
npm install -g @qwen-code/qwen-code@latest   # or: brew install qwen-code
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| goose | BB | 73.9 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/goose.min.md |
| Gemini CLI | BB | 72 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/gemini-cli.min.md |
| OpenHands | BB | 70.8 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/openhands.min.md |
| OpenCode | B | 67.7 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/opencode.min.md |
| Claude Code | C | 61.9 | agent.harness, agent.mcp-client, agent.multi-agent | https://www.anchorterminal.com/tools/claude-code.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)