NVIDIA Hopper
AWS Details Two MoE Kernel Tweaks That Lift SageMaker AI Throughput on H200
AWS has described two kernel-level optimizations for fused Mixture-of-Experts (MoE) models on Amazon SageMaker AI, reporting a 2, 4 percent end-to-end output throughput…
For developersHelion Brings Autotuned GEMM Kernels to vLLM, Beating CUTLASS and DeepGEMM on Hopper
The PyTorch team has integrated Helion, its Python-based kernel DSL, into vLLM's linear backend. According to the team, on NVIDIA Hopper GPUs it outperforms the vLLM…