salykova personal blog

Matrix Core Programming on AMD GPUs

In this blog post, we walk through how to use Matrix Cores in HIP kernels, with a focus on low-precision data types such as FP16, FP8, and FP4, as well as the new family of Matrix Core instructions with exponent block scaling introduced in the AMD CDNA™4 architecture. Through code examples and illustrations, we…
Points
116
HN comments
5
Submissions
1
Cited by
0 commenters

Software EngineeringSystems & Performance2025

On Hacker News

SubmittedTitle on HNPointsComments
Oct 4, 2025Matrix Core Programming on AMD GPUs by skidrow1165

More from salykova

All 2 posts →

Related posts from the same era