💻 Using Cursor IDE with Chinese LLMs — Complete Setup Guide

Step 1: Get Your API Key

Sign up at AI Token and get your API key. You'll have access to DeepSeek V4, V3, R1, Qwen3, and more.

Step 2: Open Cursor Settings

In Cursor, go to Cursor Settings → Models. Click "Add OpenAI Key" (yes, even though this isn't OpenAI — the API is compatible).

Step 3: Configure Custom API

Set the base URL and your API key:

Base URL: https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1
API Key:  your-api-key-here

Step 4: Override Model Name

In Cursor's model override settings, add:

# Use DeepSeek V4 as your primary coding model
deepseek-v4

# Or use DeepSeek R1 for complex reasoning
deepseek-r1

# Or Qwen3.6 Flash for fast completions
qwen3.6-flash

Step 5: Start Coding

That's it! Now use Ctrl+K (Cmd+K on Mac) as usual. Cursor will use DeepSeek V4 for code generation, completions, and chat.