Skip to main content

How to Connect Claude API to Trae IDE: Complete Setup Guide (2026)

A free AI-powered coding IDE by ByteDance — get access to top-tier Claude models in under 5 minutes, accessible from anywhere with no restrictions.

ToolsTraeclaudeapiEst. read10min
2026.04.21 published
How to Connect Claude API to Trae IDE: Complete Setup Guide (2026)

How to Connect Claude API to Trae IDE: Complete Setup Guide (2026)

A free AI-powered coding IDE by ByteDance — get access to top-tier Claude models in under 5 minutes, accessible from anywhere with no restrictions.


What Is Trae?

Trae (formerly Trea) is a native AI coding IDE built by ByteDance, based on the VS Code engine. It comes with built-in AI chat, code completion, and intelligent refactoring capabilities.

Think of it as a Cursor alternative — with a similar feature set, but completely free for core functionality.

Why Connect ClaudeAPI to Trae?

Trae’s built-in models are decent but limited. By connecting Claude Sonnet / Opus via ClaudeAPI, you unlock:

  • ✅ Significantly better code generation quality
  • ✅ More reliable handling of complex refactoring tasks
  • ✅ More accurate long-context understanding

Prerequisites

Get Your ClaudeAPI Key

  1. Go to console.claudeapi.com and sign up
  2. Navigate to the API Keys page → Create New Key
  3. Copy and save it immediately (it’s only shown once)

Step 1: Download & Install Trae

Head to the official site: trae.ai

Choose your platform (Windows / macOS) and install

blogdev-guidestrae-ide-claude-api-setup-tutorial


Step 2: Configure the Model

Windows: Install the Cline Extension

blogdev-guidestrae-ide-claude-api-setup-tutorial

mac:

  1. Open Trae, click the Settings icon ⚙️ in the bottom-left corner
  2. Select “AI Configuration” or “Model Settings”
  3. Find the “Custom Provider” option

Step 3: Configure ClaudeAPI

In the custom model configuration, enter the following:

Setting Value
Provider Type Anthropic / Claude
API Base URL https://gw.claudeapi.com
API Key Your ClaudeAPI key
Default Model claude-sonnet-4-6

⚠️ If your IDE only shows an “OpenAI Compatible” option, set the Base URL to https://gw.claudeapi.com/v1


Step 4: Choose a Model

Models available through ClaudeAPI:

Model Pricing (Input / Output) Best For
claude-sonnet-4-6 $2.40 / $12 Everyday coding, code review (recommended)
claude-opus-4-7 $4 / $20 Complex architecture, hard debugging
claude-haiku-4-5-20251001 $0.80 / $4 Quick completions, simple Q&A

Our recommendation: Start with claude-sonnet-4-6 — it offers the best balance of performance and cost.

Step 5: Verify Your Setup

  1. Click Save or Test Connection

  2. Open any project file

3.Trigger the AI chat with the shortcut (usually Ctrl/Cmd + L)

  1. Enter a test prompt:
Explain the main functionality of this file
Explain the main functionality of this file

If the model responds normally, you’re all set! ✅


Core Features

1. AI Chat

Press Ctrl/Cmd + L to open the chat panel. Use it to:

  • Ask questions about your code
  • Generate code snippets
  • Get explanations of complex logic

2. Inline Edit

Select code → press → press Ctrl/Cmd + K → type your instruction:

  • “Optimize the performance of this block”
  • “Add error handling”
  • “Convert to TypeScript”

3. Autocomplete

While you code, Claude automatically provides intelligent completions in real time. Just press Tab to accept.

4. Context References

Use the @ symbol in chat to pull in context:

  • @file - Reference the current file
  • @folder - Reference an entire folder
  • @selection - Reference selected code

Advanced Configuration

Switching Models

You can quickly switch models by clicking the model name at the top of the chat panel:

  • Simple questions → Haiku (fastest response)

  • Day-to-day development → Sonnet (balanced)

  • Complex tasks → Opus (most capable)

Custom System Prompt

Some versions support setting a default System Prompt in your configuration. For example:

You are a senior software engineer proficient in Python and TypeScript.
Keep responses concise. Code should include comments and prioritize readability and maintainability.
You are a senior software engineer proficient in Python and TypeScript.
Keep responses concise. Code should include comments and prioritize readability and maintainability.

Config File Location

Trae configuration files are typically located at:

  • Windows: %APPDATA%\Trae\settings.json
  • macOS: ~/Library/Application Support/Trae/settings.json

FAQ

Q: Getting an “Invalid API Key” error?

Check the following:

  • Make sure the key was copied in full (no extra whitespace)
  • Verify the Base URL is correct(https://gw.claudeapi.com
  • Confirm your ClaudeAPI account has sufficient balance

Q: Chat responses are slow?

  • Switch to the Haiku model for the fastest response times
  • Check your network connection
  • Reduce the number of context files referenced in the conversation

Q: Autocomplete not working?

  • Make sure AI Autocomplete is enabled in Settings

  • Some file types may be disabled by default — you’ll need to enable them manually


Q: Trae vs. Cursor — which one should I pick?

Aspect Trae Cursor
Price Free Pro $20/month
Global Access Reliable anywhere Requires stable connection
Custom API Supported natively Requires extension (Roo Code)
Ecosystem Maturity Newer More mature

Bottom line: On a budget or want a zero-friction setup? Go with Trae. Want the most polished experience and don’t mind paying? Go with Cursor + Roo Code.


TL;DR — Quick Setup Checklist

Step Action
1 Sign up for ClaudeAPI and grab your API key
2 Download and install Trae
3 Go to Settings → AI Configuration → Custom Model
4 Enter your Base URL + API Key + Model name
5 Send a test message to confirm everything works

Pairing Trae + ClaudeAPI gives you a top-tier AI coding experience — accessible from anywhere with no VPN needed, and with flexible pay-as-you-go pricing.


Get Started

Sign up for ClaudeAPI →

Related Articles