Cursor: one key for 200+ models
Cursor lets you bring your own OpenAI-compatible key and base URL for chat models. These steps follow Cursor's docs and staff answers as of October 2026.
Replace
YOUR-ENDPOINT with the endpoint shown on your API Key page after you sign up, and YOUR-KEY with your key. One key works for every model on 666 AI — Claude, GPT, Gemini, DeepSeek, Qwen and 200+ more.Steps
- Open Cursor Settings → Models.
- In the OpenAI API Key section, paste
YOUR-KEYand turn on Use OpenAI API key. - Turn on Override OpenAI Base URL and enter
https://YOUR-ENDPOINT/v1. - Add a model with the exact ID from our model page, e.g.
anthropic/claude-sonnet-5.5, and enable it in the model list. - Pick that model in chat.
Know the limits
- Your own key is used for chat models only; Tab autocomplete keeps using Cursor's models.
- The base URL override is global: while it's on, Cursor's own models may show This model does not support custom API keys. Turn the override off to use them.
- Requests go through Cursor's servers, so the endpoint must be reachable from the internet (ours is).
- Some Cursor plans don't allow your own keys — check your plan.
- Some newer models (Claude 5.x) reject the
temperatureparameter; if you see temperature is deprecated, pick another model or use Claude Code.
FAQ
Why can't I see my custom model?
Model names must match exactly, and the model must be enabled in the list. Copy the ID from its 666 AI model page.
Can I use Claude Code instead?
Yes — see our Claude Code guide; it uses the Anthropic format directly.