Trae

Connect the Trae IDE to Routero so its model calls run through your gateway — attributed, budgeted, and logged — while Trae keeps handling repository access and file edits locally.

Trae connects through a custom model entry speaking the OpenAI Chat Completions protocol (/chat/completions).

What you must set

Setting Value
Model ID a model Routero serves, e.g. deepseek/deepseek-v4-flash
API Key a Routero virtual key
Base URL https://api.routero.dcsmartvision.com/v1

1. Create a virtual key

In the dashboard, open API Keys and create a virtual key to use as Trae’s API key. Scope it to the developer’s team, restrict it to the approved models, and optionally attach a budget so individual spend is attributable.

2. Add a custom model in Trae

  1. In Trae’s AI chat panel, open the model selector and click Add Model (also available under Settings → model management).
  2. Choose Custom Model (OpenAI-compatible).
  3. Fill in the fields from the table above — the Base URL must end in /v1.
  4. Save, then select the new model in the chat panel and send a test prompt.

Trae’s SOLO mode uses a separate custom-endpoint entry (Settings → AI services → custom API endpoint) — point it at the same base URL and key.

Trae’s agent and builder modes drive tools through function calls, so pick models whose deployments accept tool-call payloads.

→ Calling the API for the base URL and auth model. → Cursor, ZCode, and Kimi Code for the other coding tools.