v9.1.1
Menu

GPU Programming

Triton Programming

Write GPU kernels with OpenAI Triton. Start from vector addition and build up to fused elementwise ops, row reductions, softmax, LayerNorm, tiled matrix multiplication, autotuning, fused dropout, and a simplified attention kernel — all running on a real NVIDIA GPU.

26

problems

5

topics

8

free

Triton Programming cover