Hacker News new | ask | show | jobs
by throwaway314155 150 days ago
PyTorch MPS is about 10x faster per the README.md.
1 comments
antirez 150 days ago
I cut the difference in speed by half by taking the activations on the GPU. Time to sleep but will continue tomorrow.
link
Numerlor 150 days ago
Have you tried e.g. Mojo that can vectorize/do SIMD without having to do intrinsics everywhere?
link