Skip to main content

Trae

Prepare API Access​

Have the three items below ready before you configure anything:

API Key

The key string you generated when creating your API Key.

Base URL

https://api.watt-cloud.com/v1

The trailing /v1 is required — do not drop it.

Model Name

Take the complete model identifier from the Model Marketplace, for example DeepSeek-V4-Pro.

Installing Trae​

Grab the installer from the official Trae website and complete the installation.

Connect Trae to Watt Cloud​

Create a Custom Model​

  1. Open Settings:
    • In IDE Mode, use the settings icon in the upper right corner.
    • In SOLO Mode, use the settings icon at the upper right of the chat panel.
  2. From the left navigation bar, pick "+ Add Model".

Choose the Provider​

Because the API is OpenAI-compatible, "OpenAI" is the provider to pick from the list.

Should none of the listed options fit, "OpenRouter" works as well — so does any provider that accepts custom endpoints.

Enter the Credentials​

  1. Click "Use Other Model" in the list, then type in the model ID — for instance GLM-5.1, DeepSeek-V4-Pro, or MiniMax-M2.7.
  2. Provide your API key (the string you created earlier).

Custom Request URL​

  1. Under "Custom Request URL", put https://api.watt-cloud.com/v1/chat/completions. This field has to be filled in: leave it empty and the provider's default URL is used instead.

  2. To complete the configuration, click "Add Model".

Verify the Connection​

To confirm the API key is valid, Trae calls the API automatically:

  • Connection Successful → the custom model has been added
  • Connection Failed → the popup shows error messages and logs; read them to locate the problem

Activate the Model​

Open the model panel and make sure the model you just added is enabled; anything not enabled won't show up in the selection list.

Switch to Your Model​

At the lower right of the AI chat input box, click the model name currently shown, open the model list, and pick the model you added.

Managing Custom Models​

The Models panel is where custom models get managed:

  • Edit Model — the edit icon opens the settings for provider, model name, API key, and more
  • Delete Model — click the delete icon, then confirm, and the model is removed
  • Disable Model — switch a model off temporarily without deleting it; disabled models stay out of the selection list