Before I do — just to be sure — is there any public documentation or API example on how to explicitly use gpt-4o-mini or cursor-small in requests? Since these are marked as “No Limit”, I’d love to switch to them temporarily while the premium quota resets.
Also, if you don’t see the free models, they might be disabled in the Cursor settings. Open them using Cmd/Ctrl + Shift + J. On the Models tab, you’ll see a list of models.
However, if the gpt-4o-mini model is missing from the list, try adding it manually.