Tile primitives for speedy kernels

Reference implementation of real-time autoregressive wavenet inference

Code from the "CUDA Crash Course" YouTube series by CoffeeBeforeArch

Fast CUDA matrix multiplication from scratch

Causal depthwise conv1d in CUDA, with a PyTorch interface

OneSweep, implemented in CUDA, D3D12, and Unity style compute shaders. Theoretically portable to all wave/warp/subgroup sizes.

CUDA implementation of exclusive prefix sum via Blelloch's algorithm

cuda implementation of depthwise conv3d

Implementation of various Parallel Computing algorithms using CUDA C++