HK
Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Top
New
Best
Ask
Show
Jobs
Request
1.
▲
Why do math libraries produce different results across platforms?
github.com/RegularJoe-CEO
3 comments
5 months ago
luxiedge
3 points
2.
▲
Constant 14ms attention: 512→524K tokens (24.5x faster than FlashAttention)
github.com/RegularJoe-CEO
1 comment
5 months ago
luxiedge
1 points
3.
▲
Show HN: O(1) memory attention – 512K tokens in 3.85 GB (eval binary)
github.com/RegularJoe-CEO
discuss
5 months ago
luxiedge
1 points