Coding agent

Qwen Code

Open coding agent with persistent memory, delegation, and opt-in isolation.

Status
Active
License
Apache-2.0
Evidence
Documented + code-verifiable, 25 sources
Product record checked
2026-08-22

At a glance

An Apache-2.0 terminal coding agent with persistent auto-memory, headless execution, hosted and local providers, MCP, specialized subagents, built-in desktop Computer Use, IDE integrations, checkpoints, and opt-in macOS or container isolation.

Good choice if

  • Terminal and CI workflows that can explicitly enable a first-party sandbox path
  • Teams switching among Qwen, OpenAI-compatible, Anthropic, Gemini, and local endpoints
  • Delegated coding tasks that benefit from constrained fork profiles, background workflows, MCP, skills, and IDE integrations
  • Long-running Goals that need resumable continuation and durable evidence checkpoints rather than a fixed turn ceiling

Check before choosing

  • The sandbox is off by default; `--yolo` does not enable it, and the default macOS profile allows network access while mainly restricting writes
  • Checkpointing is also opt-in and restores tracked project state and conversation, not shell-created external side effects
  • Auto-memory is enabled by default and can persist user or project knowledge; managed project memory in qwen serve remains Git-root-scoped by default, while exact-workspace isolation is opt-in
See 2 more considerations
  • Opt-in background workflow runs are limited to interactive TUI sessions, remain blocked on parent approval, are cancelled when the owning process exits, and now support cooperative pause and resume but not restart recovery, remote workflows, or tracing
  • The qwen serve daemon is an alpha local API: loopback starts without authentication, non-loopback binding fails closed without a bearer token, and the docs do not yet claim production-grade multi-client or network-failure guarantees

Capability support

Documented first-class product support, checked against the sources below.

External tools (MCP)
DocumentedProduct-supported MCP integrationSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Reusable skills
OptionalSkills bundled through optional Qwen Code extensionsSource · checked 2026-07-27Extension support does not establish compatibility or safety for every third-party package.
Local models
DocumentedLocal or self-hosted model pathSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Agent parallelism
DocumentedDelegated or parallel agent workflowSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Runs without an open UI
DocumentedNon-interactive or automation surfaceSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Browser control
DocumentedBuilt-in or product-supported browser controlSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Isolated execution
OptionalOptional OS or container sandboxSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Undo file changes
DocumentedProduct-supported file or session rollbackSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.

Getting started

Install `@qwen-code/qwen-code`, configure a provider, and enable Seatbelt, Docker, or Podman before running unattended tasks.

Open official documentation

Classification and operating model

Category fit and technical mechanisms are evidence records, not product-quality scores.

Category fit
Qualifies, 4/4 criteria
Operating model
7/7 layers documented
Inspect category criteria and operating mechanismsFirst-party records

Why it qualifies as a coding harness

This confirms category fit, not product quality. Every required criterion links back to first-party evidence.

Qualifies4 of 4 required criteria evidenced
  • Adaptive agent loop

    Documented

    The system repeatedly observes results and chooses the next action instead of following a fixed one-pass graph.

  • Repository tool execution

    Documented

    The system can use tools to inspect and change a repository or its execution environment.

  • Task-aware context management

    Documented

    The runtime assembles, updates, compacts, retrieves, or persists task-relevant context while work proceeds.

  • Model-independent runtime control

    Documented

    Permissions, budgets, interruption, policy, or stop controls operate outside the model's own text generation.

Membership establishes category fit only. It does not score quality, safety, autonomy, model capability, or benchmark performance. · Read the membership rule.

How it works under the hood

Seven mechanisms mapped from first-party records. These labels describe what the harness provides, not how intelligent its model is.

7/7layers documented
  • Execution & isolationSandbox availableDocumented mechanism, not a performance score.
  • Tooling & integrationsExtensible + browserDocumented mechanism, not a performance score.
  • Context & statePersistent stateDocumented mechanism, not a performance score.
  • Lifecycle & recoveryCheckpoint/rewindDocumented mechanism, not a performance score.
  • ObservabilityStructured tracesDocumented mechanism, not a performance score.
  • VerificationTool-assistedDocumented mechanism, not a performance score.
  • Governance & permissionsPolicy controlsDocumented mechanism, not a performance score.

Measured and public context

Configuration-specific measurements and source-native activity stay separate from general product capability.

Inspect code audit, measured configurations, and ecosystem signalsContext, not a product score

Public code audit

5/5public artifacts present
Security policy
Present at inspected commit
CI workflow
Present at inspected commit
Automated tests
Present at inspected commit
Evaluation assets
Present at inspected commit
Contributor documentation
Present at inspected commit

The v0.21.0 tree contains 2,062 test-like files, 39 workflows, security and contributor documentation, and one codegraph eval fixture. The fixture is project-owned development evidence rather than an immutable comparative result; provider services and hosted model behavior remain outside the source audit.

Inspect commit 5610eb405212f807a482214ddd28a259da7855d3, checked 2026-07-27

Measured configurations

No benchmark run passes the full metadata admission policy for this harness yet. Missing data is not scored as zero.

Benchmark policy and all runs
Context, not quality

Public ecosystem signals

Source-native observations for exact mapped artifacts and reviewed stable release trains. Different units and populations stay separate, and missing coverage is never treated as zero.

View this harness in Usage
  • Latest stable releasev0.23.0Released 2026-09-03; 41 stable releases in 90 daysOpen release
  • OpenRouter 30d tokens151.22B#23 coding app; 2026-08-09 to 2026-09-07Open app page
  • Homebrew 30d events4.98KFormula: qwen-codeOpen artifact
  • npm last-month downloads277.57KPackage: @qwen-code/qwen-codeOpen artifact
  • Release asset downloads82.05K47 stable releases; 235 matched assetsOpen artifact
  • VS Code installs349.75KExtension: qwenlm.qwen-code-vscode-ide-companionOpen artifact
  • Open VSX downloads712.98KExtension: qwenlm/qwen-code-vscode-ide-companion; latest 0.23.1-preview.2Open artifact
  • GitHub stars27.71KFull-source repository; 3.01K forksOpen artifact

Routing, package retrievals, release downloads, editor installs, and repository interest observe different populations. They are never added together and never affect capability evidence, classification, or measured results.

Interpretation rulesSignals checked

First-party evidence

Each capability claim links to the first-party record that supports it.

25 first-party sourcesProduct record checked

Product and interfaces

4 sources
View 3 more sources

Execution and control

2 sources
View 1 more source

Agents, state and recovery

11 sources
View 10 more sources

Automation and extensions

4 sources
View 3 more sources

Enterprise and operations

1 source

Releases and public code audit

3 sources
View 2 more sources