Cursor is serving old Claude 3.5 when you select Claude 4 - with proof

[normalnormie]

It’s actually charging you less when it doesn’t uses thinking tokens, as for point 3. that’s what auto-mode is for, that ‘complexity threshold’ is decided by Claude and not by Cursor, if you want it to think more then instruct it to think more

also check if you’re having this bug:

Claude Sonnet 4 Thinking is not thinking

Describe the Bug Frequently when using Claude Sonnet 4 thinking, the model will not used extended thinking at all, and will instead just answer immediately. It performs very poorly in these instances, and seems to generally lack chat context. For instance, in a recent query, the non-thinking Claude Sonnet 4 fixated on an unrelated issue (that I believe was mentioned in a recent chat summary in the attached context) and rapidly made several edits trying to fix a non-existent issue that was not me…