Open ABI and FFI for Machine Learning Systems

NVIDIA Math Libraries for the Python Ecosystem

Ongoing research training transformer models at scale

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

Triton-based Symmetric Memory operators and examples

PyTorch native post-training library

FlatBuffers: Memory Efficient Serialization Library

Modular visual interface for GDB in Python