Topic
WGMMA
Latest news
- 13 OctAWS details kernel engineering for fused Mixture-of-Experts on SageMaker AI
An AWS Machine Learning Blog post describes kernel optimizations for fused Mixture-of-Experts (MoE) models on Amazon SageMaker AI that lower latency and cost by increasing GPU utilization. The post explains techniques such as fusing routing and expert computations, hardware-specific tuning for NVIDIA Hopper WGMMA, and other kernel-level changes.
Models · 1 source
Articles
No articles about WGMMA yet.