DeepSeek V4.1 Flash for Cursor

This router is running. Point Cursor's OpenAI base URL here to chat with DeepSeek V4.1 Flash, including Agent mode tool calls.

Router configuration

✓
PROXY_TOKENSet. Cursor must send it as the OpenAI API key.
✓
DEEPSEEK_API_KEYSet. Requests go to https://api.deepseek.com as deepseek-flash.
✗
OPENAI_API_KEY (optional)Not set. Only DeepSeek models work while the override is on.

Set up Cursor

Open Cursor Settings → Models.
Under API Keys, paste your PROXY_TOKEN into OpenAI API Key and enable it.
Turn on Override OpenAI Base URL and enter https://cursor.knickqueue.win/v1
Add custom models: deepseek-v4.1-flash deepseek-v4.1-flash-fast deepseek-v4.1-flash-max
Pick deepseek-v4.1-flash in the chat model picker.