Hacker News new | ask | show | jobs
by Havoc 753 days ago
> 35x increase in AI inference performance compared to AMD Instinct MI300 Series

Even for marketing claims that’s pretty wild.

Still lots of trajectory left in just scale up plan it seems

1 comments

I think there is a close limit considering most of these gains are coming from the reduced memory bandwidth consumption that comes with the smaller data types. This would line up with Nvidia’s crazy graph from yesterday where data types were specified.

How much lower can these go though? 2bit? 1.58bit? 1bit? It seems that these massive gains have a very hard stop to gains that AMD and Nvidia will use to raise their stock price before it all comes to a sudden end.