Configuring TraeCode
TraeCode supports custom models, allowing you to use wylon chat models through the OpenAI Chat Completions-compatible protocol.
1. Add a model
- Open Settings → Models and select Add Model.
- Switch to Custom configuration.
- Enter the values below. Advanced display, context-window, and tool-turn options are optional.
- Select Add Model.
| Field | Value |
|---|---|
| API format | OpenAI Chat Completions |
| Full URL | Off |
| Custom request URL | https://api.wylon.cn/v1 |
| Model ID | A complete ID, such as ZhipuAI/GLM-5.3 |
| Multimodal | Off for text-only models |
| API Key | Your wylon API key |
Request URL and Full URL
Recommended: leave Full URL off and use https://api.wylon.cn/v1. Alternatively, turn it on and enter https://api.wylon.cn/v1/chat/completions.
/v1. With it off, do not enter the full /chat/completions path. Either mismatch can return 404.Model ID
Copy the exact, case-sensitive model ID from the Model Plaza or GET /v1/models.
Multimodal input
Disable image input for text-only models such as ZhipuAI/GLM-5.3.
Model family
Keep the model family at its default unless the selected model requires a matching preset. Unrelated presets may rewrite protocol fields, parameters, or thinking mode.
2. Select and verify the model
- Turn off AUTO in the chat input.
- Select your custom model.
- Send a simple message. A normal response confirms the endpoint, key, and model ID.
3. Use another model
Copy the exact ID from the Model Plaza or GET https://api.wylon.cn/v1/models. Create another custom entry, keeping the same request URL and API key while replacing the model ID and matching its declared capabilities.
