Use Cursor with a custom OpenAI-compatible API

Cursor accepts an OpenAI API key plus an overridden base URL. Point both at UncensoredAI and every chat and Composer request runs on this catalog.

Base URL

https://uncensoredai.dev/v1

01
Generate an access key
Create a key on the signup page — no email or KYC. Copy the 16-digit value; it is both your sign-in and your API credential.
02
Paste the key and override the base URL
Open Cursor Settings, switch to the Models tab, paste the key into the OpenAI API Key field, enable Override OpenAI Base URL and enter the base URL below. Keep the /v1 suffix — do not add /chat/completions.
Override base URL
```
https://uncensoredai.dev/v1
```
03
Add the model ids you want
Cursor does not auto-discover models from /v1/models. Use Add model and type an exact catalog id. Repeat for each model you plan to use.
Model ids to add
```
deepseek-v4.1-flash
gpt-6-sol
glm-5.3-flash-uncensored
```
04
Verify
Click Verify next to the key field, then pick your model in the chat picker and send a message. A 401 means the key is wrong; a 404 means the model id does not match the catalog.

Commands and configuration below are shown exactly as the tool expects them, so they stay in English.

FAQ

Does Cursor support a non-OpenAI endpoint?+

Yes. Cursor supports an OpenAI API key plus an overridden base URL. Any OpenAI-compatible gateway works; only model names must be added manually.

Which Cursor features use this key?+

Chat and Composer requests routed through the OpenAI key. Tab autocomplete runs on Cursor's own infrastructure and is not covered by this override.

Why does the model picker stay empty?+

Because discovery is disabled for custom endpoints. Add each model id manually with the Add model button; the list is not fetched from /v1/models.