41554 blogs · 4,417,066 posts Claim your blog

Record · corpus.blog/posts/01a08724-0397-71b4-8c53-2f2cb7aa28fe

Speed Up PyTorch With Custom Kernels. But It Gets Progressively Darker

alexdremov.me · published 4 January 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at alexdremov.me Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
Links from this post: 7
Words captured: 1,897

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

https://triton-lang.org/main/index.html?ref=alexdremov.me

triton-lang.org

https://github.com/alexdremov/kernels/blob/main/src/softmax/kernel.py?ref=alexdremov.me

github.com

https://github.com/fattorib/CudaSoftmax?ref=alexdremov.me

github.com

https://pytorch.org/tutorials/recipes/compiling_optimizer.html?ref=alexdremov.me

pytorch.org

https://discuss.pytorch.org/t/how-should-i-use-torch-compile-properly/144598?ref=alexdremov.me

discuss.pytorch.org

https://pytorch.org/tutorials/recipes/torch_compile_user_defined_triton_kernel_tutorial.html?ref=alexdremov.me

pytorch.org

https://discuss.pytorch.org/t/torch-compile-with-custom-triton-kernel/192876?ref=alexdremov.me

discuss.pytorch.org