Use uncensored models in Cursor

1. Install Cursor

Download it from cursor.com and sign in. Your own API key needs a paid Cursor plan that allows one.

2. Add your key

Create a key on the API keys page. In Cursor Settings (Cmd+Shift+J, or Ctrl+Shift+J), open Models, then API Keys. Paste the key as your OpenAI API key and turn it on. Turn on Override OpenAI Base URL and enter:

https://api.silkcompute.com/v1

3. Add the models

On the same page, add each model you want as a custom model:

kimi-k3.1-pro-uncensored
hy4-uncensored
deepseek-v4.1-flash-1017-uncensored
glm-5.3-uncensored
glm-5.3-flash-uncensored

4. Pick a model

Open a new chat and pick the model instead of Auto.

Troubleshooting

"Model name is not valid" means the key is off. Turn your OpenAI API key back on; Cursor sometimes turns it off after an update.

Notes

While the base URL override is on, Cursor's own models don't work. Turn it off to use them.

Cursor doesn't send reasoning effort for custom models, so each model runs at its default. Tab completion always uses Cursor's own models.

Cursor supports custom models in chat. Its Agent mode can fail with them.