Flux RouterDocumentation

Qwen Code

Add FluxRouter to Qwen Code through its OpenAI-compatible provider at the /v1 base URL.

Qwen Code talks to FluxRouter through its OpenAI-compatible provider, which uses the /v1 endpoint. Your Flux key is the only credential it needs.

For abilities as well as models, such as reading pages and making images, add the Flux tools server with the same key if your version of the tool supports remote MCP servers.

Manual config

The fastest path is environment variables:

bash
export OPENAI_BASE_URL=https://api.fluxrouter.ai/v1
export OPENAI_API_KEY=sk-YOUR-FLUX-KEY
export OPENAI_MODEL=flux-auto

To persist the provider, add an entry to the modelProviders.openai array in ~/.qwen/settings.json:

json
{
  "modelProviders": {
    "openai": [
      {
        "name": "FluxRouter",
        "baseURL": "https://api.fluxrouter.ai/v1",
        "apiKey": "sk-YOUR-FLUX-KEY",
        "model": "flux-auto"
      }
    ]
  }
}

The baseURL ends at /v1 (not /v1/chat/completions).

Test it

Run qwen and send a prompt. A reply confirms the connection. Check your FluxRouter dashboard to confirm the request landed.

Notes

flux-auto is the default and routes each request for you. If you want to pin a single lane, set the model to flux-fast, flux-standard, or flux-reasoning.