Qwen Code Setup

Qwen Code is Alibaba's terminal coding agent (a Gemini CLI fork). It speaks OpenAI-compatible, Anthropic, Gemini, and Qwen protocols -- point its OpenAI-compatible mode at ClawRouter so every request benefits from key rotation, fallback chains, and logging.

Version 1.0.18

Prerequisites

1. ClawRouter running -- clawrouter status (default: http://localhost:3030).
2. A provider created in the ClawRouter dashboard, with API keys added.
3. The proxy API key -- copy it from Settings > Proxy API Key in the dashboard.

Method 1: "Prompt for AI" (Recommended)

1. Open the provider's detail page in the ClawRouter dashboard.
2. Click "Prompt for AI" on the Base URL banner and select the Qwen Code tab.
3. Click Copy and paste the prompt to your AI agent -- or follow the generated instructions yourself.

The generated prompt contains the env-var block and the settings.json block below, pre-filled with the correct Base URL, your saved model IDs, and your real proxy API key.

Method 2: Manual Configuration

Option A -- Environment variables (simplest)

export OPENAI_API_KEY="cr_your_proxy_key"
export OPENAI_BASE_URL="http://localhost:3030/proxy/my-provider-id/v1"
export OPENAI_MODEL="model-id"

(Add the exports to your shell profile to make them permanent.)

Option B -- Settings file ~/.qwen/settings.json

Merge these keys (preserve everything else):

{
  "modelProviders": {
    "openai": [
      { "id": "model-id", "name": "model-id", "baseUrl": "http://localhost:3030/proxy/my-provider-id/v1", "envKey": "CLAWROUTER_API_KEY" }
    ]
  },
  "env": { "CLAWROUTER_API_KEY": "cr_your_proxy_key" },
  "security": { "auth": { "selectedType": "openai" } },
  "model": { "name": "model-id" }
}

Credential priority: CLI flags > shell exports > .env file > settings.json env block. If the settings file does not take effect, use Option A or run /auth inside Qwen Code.

Works with every provider: ClawRouter translates the OpenAI Chat Completions API into the provider's native format automatically.

Model Parameters (Context, Reasoning)

Per-model parameters live in the model entry's generationConfig (Option B):

{
  "id": "qwen3.5-plus",
  "name": "qwen3.5-plus",
  "baseUrl": "http://localhost:3030/proxy/my-provider-id/v1",
  "envKey": "CLAWROUTER_API_KEY",
  "generationConfig": {
    "contextWindowSize": 1000000,
    "extra_body": { "enable_thinking": true },
    "samplingParams": { "max_tokens": 65536 }
  }
}

• contextWindowSize -- the model's real context window; samplingParams.max_tokens -- output cap. Fill both from the provider's official model docs.
• extra_body -- extra request params for OpenAI-compatible APIs (e.g. Qwen's enable_thinking + thinking_budget).
• Reasoning effort is global: "model": { "reasoningEffort": "low|medium|high|xhigh|max" } or the /effort command -- the value is mapped to the provider's supported levels.

Reasoning effort value sets differ per model (low/medium/high vs low/medium/xhigh vs low/high/max) -- always verify in the provider's docs. ClawRouter translates reasoning/thinking parameters into the provider's native dialect automatically.

References: https://qwenlm.github.io/qwen-code-docs/en/users/configuration/model-providers/ and https://qwenlm.github.io/qwen-code-docs/en/users/configuration/auth/.

Verify the Connection

1. Start Qwen Code and run /model -- or send a short prompt.
2. Open the ClawRouter Logs page -- the request should appear with a successful response.

Troubleshooting

HTTP 401 "Invalid or missing API key"

• The key is missing, wrong, or was regenerated. Copy the current key from Settings > Proxy API Key. This 401 comes from ClawRouter itself, before any upstream request.

Settings not applied

• Remember the credential priority (flags > exports > .env > settings.json) -- a stale shell export wins over settings.json. Restart Qwen Code after editing.

Connection refused

• Ensure ClawRouter is running: clawrouter status. Verify the port (default: 3030) and that the base URL ends at /v1.

"Model not found" errors

• The model ID must exist on the upstream provider. Use Fetch Models in the provider's Models tab for the current list.