Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
antirez
10 days ago
|
parent
|
context
|
favorite
| on:
Flux 2 Klein pure C inference
I cut the difference in speed by half by taking the activations on the GPU. Time to sleep but will continue tomorrow.
Numerlor
10 days ago
[–]
Have you tried e.g. Mojo that can vectorize/do SIMD without having to do intrinsics everywhere?
reply
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: