Connect Agent Tools

TRAE

TRAE is an AI coding assistant with built-in models and custom model integration. It natively supports code generation, debugging, and full-stack development assistance. The official build, however, has limited model choices, queueing during peak hours, network restrictions, and higher costs — you cannot freely switch among Claude, GPT, Gemini, and other mainstream models. AutoRouter solves this by removing model lock-in, delivering near-instant responses without queues, supporting multiple models, local payment methods, and lower call costs, so TRAE can adapt to different models and development scenarios.

📦 Prerequisites

What you'll need

  • TRAE installed (macOS / Windows / Linux)
  • An AutoRouter API key (generated in the console)
  • A model ID exactly matching one exposed by your AutoRouter console (e.g. claude-fable-5)

🚀 Step 1: Download and install TRAE

  1. Visit the TRAE website: https://www.trae.cn
  2. Download the installer for your platform

Download TRAE

  1. Complete the installation

🔧 Step 2: Sign in

TRAE requires an account to unlock AI features (code generation, Q&A, etc.). Sign-in is free — no paid subscription required:

  1. Launch TRAE and open the login screen. You can sign in with a phone number or email (phone is recommended for users in mainland China)
  2. Enter your phone number or email, request a verification code, enter the code, then click Login
  3. First-time sign-in does not require a separate registration — after verification, an account is created automatically and you enter the main interface

🔑 Step 3: Get an API Key

  1. Open the AutoRouter API Keys page and click Create API Key. See Quick Start for details.
  2. After creation, copy and store the API Key securely — you will need it in the next steps.

Create API Key

⚙️ Step 4: Configure a custom model in TRAE

Click your username and open Settings:

Open Settings

  1. Open the Models menu and add a model.

Add Model

  1. Choose Custom Config.

Select Custom Config

  1. Fill in the following fields:

    • API endpoint: https://api.autorouter.top/v1/chat/completions
    • API Key: your AutoRouter API Key
    • Model name: claude-fable-5 (or any other model ID available in your AutoRouter console)

Fill in model details

  1. Submit the model.

Submit model

✅ Step 5: Verify the connection

After you click Add Model, TRAE calls the provider endpoint to validate the API key.

Verify connection

  • If the connection succeeds, the custom model is added and available in chat.
  • If it fails, an error message is shown — use it to troubleshoot.

❓ Troubleshooting

IssueFix
Connection failed / invalid API keyConfirm the endpoint is https://api.autorouter.top/v1/chat/completions and the API Key is correct
Model not foundConfirm the model name exactly matches an ID in your AutoRouter console
Cannot find Custom ConfigIn Settings → Models → Add Model, choose Custom Config
Model not available in chatEnsure the model toggle is on, then switch to it in the chat model selector

OpenCode

OpenCode is an open-source terminal AI coding agent with both TUI and CLI modes. It's compatible with 75+ AI service providers and, via its built-in @ai-sdk/anthropic and @ai-sdk/openai-compatible adapters, can seamlessly connect to both Anthropic-protocol models and OpenAI-compatible models served by AutoRouter.

WorkBuddy

WorkBuddy is a full-scene desktop AI agent from Tencent. It integrates with QQ / WeCom ecosystems, supports local computer control and multi-model switching, and offers commercial service via Token quota plans for both individuals and teams. Through custom models (OpenAI-compatible protocol), you can seamlessly connect models served by AutoRouter. This guide uses gpt-5-nano as an example.

On this page

📦 Prerequisites🚀 Step 1: Download and install TRAE🔧 Step 2: Sign in🔑 Step 3: Get an API Key⚙️ Step 4: Configure a custom model in TRAE✅ Step 5: Verify the connection❓ Troubleshooting