ml-compiler
Regression-aware PyTorch inference optimizer using FX, Triton, and TorchInductor
DNN Compiler for Heterogeneous SoCs
PyTorch compiler and WebGPU runtime