Helion: A High-Level DSL for Performant and Portable ML Kernels(pytorch.org)

12 pointsby xfrvor 9 Monaten1 comments

https://pytorch.org/blog/helion/

1 comments

Archit3chvor 9 Monaten

Why does this autotuning require a specialized DSL and cannot be a part of a more general LLVM-based language or Mojo?