Cursor AI acting weird for anyone else?

Hey everyone! Quick question about Cursor AI’s performance.

I’ve been testing the same code modifications using Claude 3.5 in both Cursor AI and Poe. While Poe’s Claude nailed it, Cursor AI’s version actually made my code worse. Like, significantly worse.

Has anyone else noticed inconsistencies between different platforms using the same AI model? I’m curious if this is just my experience or if others have run into similar issues with Cursor AI.

Would love to hear your experiences, especially if you’ve compared it with other platforms!

I’m not sure if it’s strange or not, but I suspect whether Cursor Chat’s Claude 3.5-Sonnet-20241022 is using the official API route. Why is the output so laggy, while the GPT-4 model runs very smoothly?

I hit return and get generation barely starts and then quits.