Remix.run Logo
steinvakt2 3 days ago

I had a 5090 some months ago but couldnt get flash attention to work. Does it now work natively? What about 5080?

sigmoid10 3 days ago | parent | next [-]

Pytorch now has native support for the Blackwell architecture:

https://pytorch.org/blog/pytorch-2-7/

SynasterBeiter 3 days ago | parent [-]

It does, but the performance is pretty bad, worse than Hopper.

zackangelo 3 days ago | parent | prev [-]

Curious what issues you were having. The kernel should compile natively if you pass nvcc the correct arch flags, although it probably won't take advantage of any new hardware features.

saagarjha 3 days ago | parent [-]

High-performance GPU code typically uses nonportable features that are not supported across generations.