Alibaba Cloud Model Studio:Qwen Code

API key (recommended for production)

Create or edit settings.json at the path below. Replace YOUR_API_KEY with your Model Studio API key:

• macOS / Linux: ~/.qwen/settings.json

• Windows: C:\Users\your_username\.qwen\settings.json

{
  "env": {
    "BAILIAN_API_KEY": "YOUR_API_KEY"
  },
  "modelProviders": {
    "openai": [
      {
        "id": "qwen3.5-plus",
        "name": "[Bailian] qwen3.5-plus",
        "baseUrl": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
        "envKey": "BAILIAN_API_KEY",
        "generationConfig": {
          "extra_body": {
            "enable_thinking": true
          }
        }
      },
      {
        "id": "qwen3-coder-plus",
        "name": "[Bailian] qwen3-coder-plus",
        "baseUrl": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
        "envKey": "BAILIAN_API_KEY"
      },
      {
        "id": "qwen3-coder-next",
        "name": "[Bailian] qwen3-coder-next",
        "baseUrl": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
        "envKey": "BAILIAN_API_KEY"
      }
    ]
  },
  "security": {
    "auth": {
      "selectedType": "openai"
    }
  },
  "model": {
    "name": "qwen3.5-plus"
  },
  "$version": 3
}

Qwen OAuth (1,000 free calls per day)

For faster setup without an API key:

1. Run qwen in a terminal.

2. Type /auth and select Qwen Oauth.

3. Press Enter. On the redirected page, log in to your Qwen Chat account or scan the QR code to authenticate.

After authentication, Qwen Code provides 1,000 API calls per day, separate from Model Studio's free quota.

If you use up the 1,000 daily API calls and want to continue, use /auth to switch to the OpenAI authentication method.

If this is your first time using Qwen Chat, register an account.

Configuration reference

For more modelProviders configuration options, see the Qwen Code configuration documentation.

Why does Qwen Code consume tokens quickly?

Qwen Code may make multiple API calls per request, consuming tokens quickly. To reduce consumption:

• Streamline the working directory

  Start Qwen Code in a specific project directory. Avoid directories with too many files (e.g., root directory) to reduce token consumption.

• Set a token limit

  Create .qwen/settings.json in the project root and restart Qwen Code. The sessionTokenLimit field sets the max tokens per API call:

  {
    "sessionTokenLimit": 32000
  }

  Because Qwen Code may make multiple API calls for a single request, the total token consumption can exceed the sessionTokenLimit value.

• Use compression or cleanup commands

  • /compress: Compress conversation history to free up tokens and continue.

  • /clear: Clear all conversation history and start a new session.

Q: How do I use the free quota of 1,000 calls per day?

A: Type /auth after starting Qwen Code and select Qwen Oauth. Press Enter and authenticate on the redirected page (log in to your Qwen Chat account or scan the QR code). After authentication, Qwen Code provides 1,000 API calls per day, separate from Model Studio's free quota.

If you use up the 1,000 daily API calls and want to continue, use /auth to switch to the OpenAI authentication method.

If this is your first time using Qwen Chat, register an account.

How do I switch models?

Type /model to select a different model from your settings.json. To add a new model, add its configuration to modelProviders and restart Qwen Code.

Why do I get a 401 Incorrect API key provided error?

This error usually indicates incorrect API configuration. Verify the API key in the error message matches your settings. Note: project-level settings.json overrides user-level settings.json. See Qwen Code troubleshooting for details.