hayek

joined 2 years ago
sorted by: new top controversial old
90
Nvidia released a paper about a 100KB text-to-image model that only trained for 4 minutes but claims to be better than bigger models (research.nvidia.com)
 

They also claim that it only takes about 8 seconds to generate various good images.

Place is going exactly as expected in c/[email protected]
[–] [email protected] 28 points 2 years ago* (last edited 2 years ago) (3 children)

Didn‘t know they brought it back. They probably need to fix their engagement numbers for the IPO…