TRAE

TRAE is an AI IDE with a built-in coding agent. Its Custom Config model option accepts an OpenAI-compatible endpoint, so you can point it to the ScitiX Model Inference API for agent and chat use. Supported models include glm-5.2 and DeepSeek-V4-Flash. For the full list and pricing, see the model catalog.

​

Configure

Create an API Key on the API Keys page first.

1. Open Settings → Models, click Add Model, and switch to the Custom Config tab.
2. Fill in the required fields.

   Field	Value
   API Format	OpenAI Chat Completions
   Custom Request URL	https://api.scitix.ai/model-api/v1
   Model ID	glm-5.2
   API Key	Your API Key created on the API Keys page.

   Leave the Full URL toggle off. TRAE appends /chat/completions to the URL you provide, so enter the base endpoint without a trailing slash. Under Advanced Settings, you can set a Display Name and Context Window. Multimodal is a separate toggle in the Add Model dialog; enable it only for a vision-capable model. Choose a model with native tool calling for the agent’s file actions.
3. Click Add Model.

​

Verify

Open the agent panel, select the ScitiX model from the model picker at the bottom, and give it a task. TRAE plans the task and edits files, asking you to keep or undo the changes. A normal reply confirms that the connection works.

​

FAQ

• Model not responding / no edits: confirm that the selected model is the ScitiX model you added, and that it supports native tool calling.