Skip to main content

OpenClaw

Prerequisites​

  1. A CloudBase environment with its Environment ID (ENV_ID)
  2. A Token Resource Pack purchased
  3. Enable the required models in Console → AI → Text Models
  4. A CloudBase API Key (Console → Environment Settings → API Key)

Configuration Steps​

  1. Run openclaw dashboard in your terminal to open the WebUI in your browser
  2. Navigate to Config → Raw tab in the left menu
  3. In the openclaw.json editor, locate or add the models.providers field with the following configuration:
{
  "models": {
    "mode": "merge",
    "providers": {
      "cloudbase": {
        "baseUrl": "https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/cloudbase",
        "apiKey": "<YOUR_CLOUDBASE_API_KEY>",
        "models": {
          "hy3-preview": {
            "label": "Hunyuan Hy3",
            "reasoning": false
          }
        }
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "cloudbase/hy3-preview"
      }
    }
  }
}

Replace <ENV_ID> with your actual CloudBase environment ID and <YOUR_CLOUDBASE_API_KEY> with your CloudBase API Key.

  1. Click Save, then click Update to apply the configuration

Thinking Mode​

The reasoning field controls whether the Hunyuan Hy3 model uses extended thinking:

  • false (default): no_think mode, faster responses
  • true: enables deep reasoning

Troubleshooting​

If the configuration looks correct but requests return 404 or 401 errors, this is typically an OpenClaw caching issue. Directly modify the local configuration file and restart the gateway to resolve it.