Rick W / Saturday, November 21, 2026 / Categories: Artificial Intelligence Fast Attention Kernels and Operator Fusion Overview This post is divided into four parts; they are: • Operator Fusion • FlashAttention-Style Kernels • Paged Attention Kernels • Fusion and CUDA Graphs with torch. Previous Article GPT-6.2 Astra and Sol: A Hands-On Performance Review Next Article The Complete Guide to LLM Evaluation Print 5