Simon Boehm

mentions 2 type Person feed RSS

// recent coverage 2 mentions

00:00

2026-09-14

rocm.blogs.amd.com

ai-chips

An Educational GEMM Ladder for Helios GPUs

AMD's Helios GPU, the Instinct MI455X, packs 432 GB of HBM4, 23 TB/s of HBM bandwidth per GPU, and 40 PFLOPs of FP4 compute, according to AMD's product brochure. A new educational blog post builds a l…

16:58

2026-08-11

philipbotros.github.io

machine-learning

How to get a matmul kernel into the top percentile (worklog)

A developer's worklog documents iterative optimization of a CUDA matrix multiplication kernel to reach the top percentile on LeetGPU, focusing on memory hierarchy and data reuse bottlenecks rather tha…

// co-occurs with top 8 entities

LeetGPU 1 Nvidia 1 Stephen Jones 1 Aleksa Gordic 1 T4 1 CUDA 1 arXiv 1 AMD 1

// topics top 5 topics

ai infrastructure 2 developer tools 2 machine learning 1 ai chips 1 ai research 1