Specification Parameters
Interfaces & Systems
- Supported systems
- macOS、Linux、Windows · Corresponding installation scripts provided
- IDE integration
- VS Code 插件 Qwen Code Companion,需 VS Code 1.85.0+ · Shares settings.json with the CLI
- Graphical interface
- Qwen Code Desktop,内置 CLI 运行时 · Installers available for macOS, Windows, and Linux
Power
- Bailian CLI prerequisite
- Node.js 18+ · Required when accessing Bailian CLI capabilities
资料下载
Comparison with Similar Products
| Specifications | 本品 · Qwen Code | kimi-k3.1-pro | 火山方舟模型 API | qwen-audio-3.1-asr-flash | doubao-seed-2.1-turbo |
|---|---|---|---|---|---|
| Main controller | — | — | — | — | — |
| NPU | — | — | — | — | — |
| Memory / VRAM | — | — | — | — | — |
| Reference unit price | Request a Quote | Request a Quote | Request a Quote | Request a Quote | Request a Quote |
| Inventory | — | — | — | — | — |
Product introduction
Product forms & deployment
Qwen Code is primarily a terminal CLI. On macOS/Linux it is installed via a curl installation script; on Windows, run the bat installation script in cmd as administrator. After installation, reopen the terminal and verify with qwen --version. There are also a VS Code extension, Qwen Code Companion (requires VS Code 1.85.0 or higher), and Qwen Code Desktop, a graphical interface version with a built-in CLI runtime. All three share the same settings.json, so one configuration works across all of them.
Billing & access
After launch, configuration is done visually via /auth. The four plans use different baseUrls and API Keys: Token Plan Personal/Team points to the compatibility-mode endpoint at token-plan.cn-beijing.maas.aliyuncs.com; Coding Plan points to coding.dashscope.aliyuncs.com/v1; pay-as-you-go uses your own Standard API Key, with the baseUrl configured by region (Beijing, Singapore, US Virginia), where the URL must be replaced with the actual Workspace ID and the API Key must match the selected region.
Available models
The Token Plan Personal configuration example lists auto, qwen3.8-max, qwen3.8-flash, qwen3.7-max, qwen3.7-plus, qwen3.6-flash, glm-5.3, glm-5.2, deepseek-v4.1-flash and its snapshot versions, deepseek-v4-pro, and more; the Team version additionally includes qwen3.6-plus, kimi-k2.7-code, kimi-k2.6, kimi-k2.5, glm-5.1, glm-5, MiniMax-M2.5, and more; the Coding Plan example lists qwen3.7-plus, qwen3.6-plus, qwen3.5-plus, MiniMax-M2.5, glm-5, kimi-k2.5. Some models support toggling thinking mode via generationConfig.extra_body.enable_thinking.
Common commands
Supported commands include /model to switch models, /auth to change authentication, /init to generate the project context file QWEN.md, /clear, /compress to compress history and save tokens, /resume to restore sessions, and /stats to view statistics; the IDE extension supports only a subset of commands.
Typical use cases
After accessing the Bailian CLI, image generation and video generation capabilities can be invoked in conversation (prerequisite: Node.js 18+); a website screenshot can be used to replicate front-end code; an open source repository URL can be used to generate a promotional video. The documentation specifically warns that pasting plaintext API Keys into conversation may be retained in session logs, and recommends using environment variables or settings.json instead.
