News for hackers. Stuff that matters.

A comment on Making AMD GPUs competitive for LLM inference. Parent comment

Re:Making AMD GPUs competitive for LLM inference by zorgmonkey on Wednesday August 09, 2023 @06:34PM (#37069289)
How does the performance with Vulkan compare to the ROCm performance on the same hardware?
[ Parent | on HN ]
Re:Making AMD GPUs competitive for LLM inference by junrushao1994 on Thursday August 10, 2023 @12:05AM (#37071816)
We haven't done any comparison them yet, but generally we believe Vulkan as a more generic cross-vendor API should be slower than ROCm. Same for CUDA vs Vulkan.
[ Parent | on HN ]