NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
created at May 2, 2019, 10:02 p.m.
[ARCHIVED] The C++ parallel algorithms library. See https://github.com/NVIDIA/cccl
created at March 6, 2012, 1:01 a.m.
Optimized primitives for collective multi-GPU communication
created at Nov. 14, 2015, 12:12 a.m.
[ARCHIVED] The C++ Standard Library for your entire system. See https://github.com/NVIDIA/cccl
created at Sept. 17, 2020, 6:53 p.m.
An efficient C++17 GPU numerical computing library with Python-like syntax
created at Sept. 14, 2021, 10:49 p.m.
NVIDIA NMOS (Networked Media Open Specifications) Library
created at July 12, 2024, 5:35 p.m.