Y
Hacker News
new
|
ask
|
show
|
jobs
by
hnfong
798 days ago
With quantized models you can run 70B models on 64GB RAM comfortably.