NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
🚀 A high-performance C++ library for robust mesh input/output across numerous formats. This project serves as the ultimate "Swiss Army knife" for mesh manipulation workflows, providing blazing-fast and reliable data handling.
A C and Python library providing a shared, high-performance, vectorized, GPU-capable scene data substrate for OpenUSD data across Omniverse Libraries (physics, rendering, sensors, animation).
QR, symmetric eigendecomposition and SVD for batches of matrices on Apple Silicon GPUs, for MLX and PyTorch. Metal kernels per regime, routed to the GPU or the CPU by policies measured per Mac. C++, Python (pip: metal-linalg, metal-linalg-torch), Swift and C.