News for hackers. Stuff that matters.

A comment on Llama.cpp: Full CUDA GPU Acceleration. Parent comment

Re:Llama.cpp: Full CUDA GPU Acceleration by jahav on Tuesday June 13, 2023 @07:08AM (#36308400)
True, it's better just to use OpenSYCL that stores intermediate device agnostic form and complies it as needed to specific card.

I don't understand why isn't SYCL more widely used.

[ Parent | on HN ]
Re:Llama.cpp: Full CUDA GPU Acceleration by SubjectToChange on Wednesday June 14, 2023 @08:40PM (#36334921)
SYCL is still very early on in development and I don’t see it really picking up until support is up-streamed to the llvm project, at the very least. That said, I am a firm believer in the single source philosophy. There just isn’t a tractable alternative.
[ Parent | on HN ]