Re:Outperforming LLMs with less training data and smaller mo...
by marcyb5st on Friday September 22, 2023 @06:25AM
(#37610234)
Agreed. I would love to have something with the FLOPs of a 3070/4060 and 80 gigs of even slower VRAM (not necessarily HBM/GDDR6x) that can run the XXB models.
[ Parent
| on HN ]
There are no comments on this comment yet.