Fast and memory-efficient exact attention

PjRt plugin and Python APIs for MPMD workflows in Jax

Experimental projects related to TensorRT

PyTorch compiler that accelerates training and inference. Get built-in optimizations for performance, memory, parallelism, and easily write your own.

A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")

Home for OctoML PyTorch Profiler

[DEPRECATED] Moved to ROCm/rocm-libraries repo