Skip to content
Better HN
Top
New
Best
Ask
Show
Jobs
Search
⌘K
7x speed improvement for LLaMA in less than 10 lines of code | Better HN
7x speed improvement for LLaMA in less than 10 lines of code
(opens in new tab)
(github.com)
2 points
hack_ml
2y ago
1 comments
Share
1 comments
default
newest
oldest
brucethemoose2
2y ago
Is that 5s per token?
j
/
k
navigate · click thread line to collapse