NVIDIA’s home for open source projects and research across artificial intelligence, robotics, and more.

 

RAPIDS

Open GPU data science

Website›   GitHub›

TensorRT

High-performance platform for deep learning inference

Website›   GitHub›

NCCL

Optimized primitives for collective multi-gpu communication

Website›   GitHub›

Triton

Inference microservice for data center production that maximizes GPU utilization

Website›   GitHub>

DALI

Data pre-processing in deep learning applications

GitHub›

Apex

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

Website›   GitHub›

TensorFlow-TensorRT

Integration for TensorFlow with TensorRT to get upto 6x faster inference in TensorFlow with few lines of code.

Blog›   GitHub›

ONNX

TensorRT backend for ONNX

Website›   GitHub›

NVDLA

Open source Deep Learning Inference Accelerator

Website›   GitHub›

Deep Learning Examples

Tensor Cores optimized code-samples



Website›   GitHub›

Docker

Build and run Docker containers leveraging NVIDIA GPUs

Website›   GitHub›

Kubernetes

Kubernetes on NVIDIA GPUs

Website›   GitHub›

K8s Device Plugin

Enable GPU support in Kubernetes with the NVIDIA device plugin

GitHub›

Container Runtime

Support multiple Linux container runtimes via the NVIDIA Container Runtime

Website›   GitHub›

Container Runtime Library

Automatically configure GNU/Linux containers leveraging NVIDIA hardware

GitHub›

MDL (Material Definition Language)

Physically accurate material standard and SDK


Website›   GitHub›

USD (Universal Scene Description)

Extensible 3D scene description for composing virtual worlds

Website›   GitHub›

PhysX

Advanced, true-to-reality physics simulation engine

Website›   GitHub›

TF-TRT Image Classification

Image classification with native integration of NVIDIA TensorRT for TensorFlow

GitHub›

Flang

Flang is a Fortran compiler targeting LLVM

GitHub›

CUTLASS

CUDA Templates for Linear Algebra Subroutines

Blog›   GitHub›

Thrust

Parallel algorithms library

Website›   GitHub›

AmgX

Distributed multi-grid linear solver library on GPU

Website›   GitHub›

libcu++

C++ Standard Library for your entire CPU+GPU system

GitHub›

CUB

A library of collective primitives and utilities

Website›   GitHub›

nvcomp

High performance GPU data compression library

Website›   GitHub›

Open Seq2Seq

Toolkit for efficient experimentation with various sequence-to-sequence models

Blog›   GitHub›

vid2vid

High-resolution photorealistic video-to-video translation

GitHub›

Deep Recommender

Deep learning for recommender systems

GitHub›

Milano

Tool for automating hyper-parameters search for your models on a backend of your choice

GitHub›

Tacotron 2

PyTorch implementation with faster-than-realtime inference

GitHub›

Falcor

Real-time rendering framework

Website›   GitHub›