logoalt Hacker News

trollbridge • today at 4:52 PM • 1 reply • view on HN

Then it’s that much easier to develop a competitor.

I use Apple and AMD chips for desktop inference. nVidia is still cheaper for anything beyond that.


Replies

crostlybostly • today at 5:05 PM

Is AMD inference clunky these days or does it work pretty seamlessly with most local tools?