Stylized diagram of a GPU die with attention tile matrices overlaid in cyan and amber.

Optimizing Triton Kernels for Custom Attention Patterns on H100 GPUs

A working engineer’s guide to squeezing the H100 with custom Triton kernels: tile sizing, swizzle patterns, pipeline stages, and the warp-specialization tricks that make non-vanilla attention fly.

September 5, 2026 · 11 min · 2308 words · martinuke0
Feedback