Standard Compute
Qwen Code
#11 of 21 · 8.0/10 editorialCoding AgentOpen Source

Qwen Code Review 2026: Ratings, Pros & Cons, Alternatives

Qwen Code is Alibaba's Qwen team's open-source coding agent. It was originally based on Gemini CLI v0.8.2, stopped syncing with upstream at v0.1 and has been developed on its own since, with frequent releases (v0.25.2 on October 15, 2026). It ships subagents, agent teams, skills, hooks, MCP, plan mode, LSP and git worktrees, plus VS Code, JetBrains and Zed plugins, a desktop app and Telegram, DingTalk, WeChat and Feishu channels. Unlike Gemini CLI it speaks the OpenAI, Anthropic and Gemini APIs, so any compatible provider works through modelProviders in ~/.qwen/settings.json. The free Qwen OAuth tier ended on April 15, 2026: the tool is still free, and the models come from Alibaba's Coding Plan or Token Plan, a built-in third-party provider, a local model or a custom provider.

Visit Qwen Code →Pricing: Free (Apache 2.0), bring your own providerUpdated 2026-10-15
What it costs to run Qwen Code

4 options compared, with prices and limits.

Best API for Qwen Code →See the Qwen Code integration →

How Qwen Code Scores

Output Quality
8.0/10
Autonomy
8.0/10
Reliability
7.5/10
Speed
8.0/10
Value
9.0/10
Ease of Use
7.5/10

Bars show editorial scores from hands-on testing. Percentages showQwen Code's live community win rate in head-to-head votes, shown once a category has 10 or more votes (not enough yet, vote in a comparison below).

Qwen Code Pros, Cons, and Known Issues

Pros

  • Free and open source (Apache 2.0), with the agent features of Claude Code: subagents, agent teams, skills, hooks, MCP, plan mode, LSP and git worktrees
  • Multi-protocol: OpenAI, Anthropic and Gemini APIs, so any compatible provider, router or local model works
  • Switch models mid-session with /model; settings.json edits reload without a restart
  • Runs beyond the terminal: VS Code, JetBrains and Zed plugins, a desktop app, SDKs and chat channels
  • Strong with Qwen models: the Qwen team reports 76.4% to 77.8% on SWE-bench Verified with Qwen 3.7 Max

Cons

  • No free models any more: the Qwen OAuth tier ended on April 15, 2026, so you need a provider key
  • Tuned for Qwen models first; other providers work but get less attention in the docs
  • Setup for anything beyond the /auth menu means editing modelProviders in ~/.qwen/settings.json

Known issues users report

  • Older guides still describe the free Qwen OAuth login; cached tokens stop working and new requests are rejected
  • security.auth.apiKey and security.auth.baseUrl are deprecated, and the qwen auth command was removed; use /auth or modelProviders
  • Configs that wrap a provider's models in { protocol, models } from early previews are silently skipped; use the plain array form
  • The built-in web_search tool only works out of the box with Alibaba's Token Plan or a Model Studio key

Key Features

OpenAI, Anthropic and Gemini protocols
Subagents, agent teams & skills
MCP, plan mode & LSP
VS Code, JetBrains & Zed plugins
Desktop app & chat channels
Open source (Apache 2.0)

Runs on: macOS · Linux · Windows · VS Code / JetBrains / Zed · Desktop app|Best for: Developers who want a free, Claude Code-style terminal agent that runs Qwen models next to any other provider

Qwen Code vs the Alternatives: Vote in Any Matchup

Qwen CodeVSClaude CodeQwen CodeVSHermesQwen CodeVSCodexQwen CodeVSGemini CLIQwen CodeVSOpenClawQwen CodeVSKilo CodeQwen CodeVSCursorQwen CodeVSCopilotQwen CodeVSClineQwen CodeVSWindsurfQwen CodeVSAiderQwen CodeVSContinueQwen CodeVSOpenCodeQwen CodeVSPiQwen CodeVSompQwen CodeVSRoo CodeQwen CodeVSDevinQwen CodeVSAmpQwen CodeVSJulesQwen CodeVSAntigravity

Qwen Code runs on Standard Compute

Standard Compute

Current frontier models Claude Fable 5.1 · GPT-6 Astra

More intelligence
per dollar.

Each step goes to the cheapest model that can do it, on providers we vet for price, uptime and speed, with automatic fallback when one fails.

Compliance
Opt-in rules keep a company, team or key on European or US providers
Privacy
We don’t train on your prompts or store them

Fixed monthly plans from $19 · cancel anytime · Qwen Code setup →

7-day fair refund Free trial, no card Billing by Stripe3,400+ agent users

Qwen Code: Frequently Asked Questions

Is Qwen Code free?

The tool is: Qwen Code is open source (Apache 2.0) and free to install. The free Qwen OAuth model tier ended on April 15, 2026, so the models now come from a provider you choose in /auth: Alibaba's Coding Plan (Pro, $50 a month on the international site), Token Plan or a Model Studio key, a built-in third-party provider such as OpenRouter or DeepSeek, a local model, or a Custom Provider.

Can Qwen Code use Claude, GPT or other non-Qwen models?

Yes. Qwen Code speaks the OpenAI, Anthropic and Gemini APIs. Declare models under modelProviders in ~/.qwen/settings.json with an id, a baseUrl and an envKey, or use /auth → Custom Provider, then switch between them with /model.

How is Qwen Code related to Gemini CLI?

Qwen Code was originally based on Gemini CLI v0.8.2. From Qwen Code v0.1 it stopped syncing with upstream and became its own multi-protocol agent, so unlike Gemini CLI it accepts OpenAI- and Anthropic-compatible providers.