Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
emaadm
87 days ago
|
parent
|
context
|
favorite
| on:
We reverse-engineered Flash Attention 4
There's some interesting work in NeurIPS this year on fused kernels for MoE too:
https://flash-moe.github.io/
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: