Latent Space AI cover image

Investors Bet Big as Cursor Raises Another $2.3B

Latent Space AI

00:00

Technical Implementation: Avoiding CUDA Libraries

Jaeden notes Cursor implemented Composer kernels with PXT instead of CUDA, achieving reported 3x performance gains on NVIDIA chips.

Play episode from 05:17
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app