Repositories list

• nvcf

  Public

  NVIDIA/nvcf’s past year of commit activity
  Platform for deploying and routing GPU-accelerated inference, streaming, and batch workloads at scale.

  kubernetesgpuinference+ 3

  kubernetesgpuinferencenvidiacloud-functionsserverless-gpu

  Go

  •

  Apache License 2.0

  •4343 forks•195195 stars•155155 issues•3636 pull requests•Updated Aug 7, 2026Aug 7, 2026

• aicr

  Public

  NVIDIA/aicr’s past year of commit activity
  Tooling for optimized, validated, and reproducible GPU-accelerated AI runtime in Kubernetes

  configkubernetesmanifest+ 5

  configkubernetesmanifestairuntimegpuhelmargocd

  Go

  •

  Apache License 2.0

  •7979 forks•377377 stars•106106 issues•1111 pull requests•Updated Aug 7, 2026Aug 7, 2026

• stdexec

  Public

  NVIDIA/stdexec’s past year of commit activity
  `std::execution`, the standard C++ framework for asynchronous and parallel programming.

  C++

  •

  Apache License 2.0

  •262262 forks•2.4k2.4k stars•140140 issues•1616 pull requests•Updated Aug 7, 2026Aug 7, 2026

• NeMo-Relay

  Public

  NVIDIA/NeMo-Relay’s past year of commit activity
  Multi-language agent runtime and library for execution scope management, lifecycle events, and middleware on tool and LLM calls.

  securitymiddlewareai+ 16

  securitymiddlewareairuntimeoptimizationnvidiaagentsobservabilitycodexotel+ 9

  Rust

  •

  Apache License 2.0

  •5252 forks•113113 stars•1010 issues•1515 pull requests•Updated Aug 7, 2026Aug 7, 2026

• infra-controller

  Public
  NVIDIA Infra Controller - Hardware Lifecycle Management and multitenant networking

  Rust

  •

  Apache License 2.0

  •170170 forks•247247 stars•728728 issues•9595 pull requests•Updated Aug 7, 2026Aug 7, 2026

• dsx-github-actions

  Public
  Github Action infrastructure for DSX

  Dockerfile

  •

  Apache License 2.0

  •66 forks•55 stars•00 issues•22 pull requests•Updated Aug 7, 2026Aug 7, 2026

• mig-parted

  Public

  NVIDIA/mig-parted’s past year of commit activity
  MIG Partition Editor for NVIDIA GPUs

  Go

  •

  Apache License 2.0

  •6767 forks•260260 stars•77 issues•33 pull requests•Updated Aug 7, 2026Aug 7, 2026

• cuda-quantum

  Public

  NVIDIA/cuda-quantum’s past year of commit activity
  C++ and Python support for the CUDA Quantum programming model for heterogeneous quantum-classical workflows

  pythoncppquantum+ 6

  pythoncppquantumquantum-computinghacktoberfestquantum-programming-languagequantum-algorithmsquantum-machine-learningunitaryhack

  C++

  •

  Apache License 2.0

  •435435 forks•1.1k1.1k stars•350350 issues•161161 pull requests•Updated Aug 7, 2026Aug 7, 2026

• gpu-operator

  Public

  NVIDIA/gpu-operator’s past year of commit activity
  NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes

  kubernetesgpucuda+ 1

  kubernetesgpucudanvidia

  Go

  •

  Apache License 2.0

  •531531 forks•2.8k2.8k stars•4949 issues•4545 pull requests•Updated Aug 7, 2026Aug 7, 2026

• NVFlare

  Public
  NVIDIA Federated Learning Application Runtime Environment

  pythondecentralizedpet+ 4

  pythondecentralizedpetprivacy-protectionfederated-learningfederated-analyticsfederated-computing

  Python

  •

  Apache License 2.0

  •270270 forks•954954 stars•66 issues•2222 pull requests•Updated Aug 7, 2026Aug 7, 2026

• cuvs

  Public

  NVIDIA/cuvs’s past year of commit activity
  cuVS - a library for vector search and clustering on the GPU

  machine-learninginformation-retrievalstatistics+ 13

  machine-learninginformation-retrievalstatisticsclusteringgpudistancecudasparsenearest-neighborssimilarity-search+ 6

  Cuda

  •

  Apache License 2.0

  •217217 forks•829829 stars•536536 issues•115115 pull requests•Updated Aug 7, 2026Aug 7, 2026

• libredfish

  Public
  A Rust Crate for interacting with DTMF Redfish endpoints

  Rust

  •

  MIT License

  •3232 forks•1313 stars•00 issues•33 pull requests•Updated Aug 7, 2026Aug 7, 2026

• NemoClaw

  Public
  Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference

  typescriptnvidiasandboxing+ 4

  typescriptnvidiasandboxinghermesai-agentsopenclawopenshell

  TypeScript

  •

  Apache License 2.0

  •3k3k forks•22k22k stars•240240 issues•4848 pull requests•Updated Aug 7, 2026Aug 7, 2026

• cuda-python

  Public
  CUDA Python: Performance meets Productivity

  Cython

  •

  Apache License 2.0

  •318318 forks•3.3k3.3k stars•206206 issues•7070 pull requests•Updated Aug 7, 2026Aug 7, 2026

• Megatron-LM

  Public

  NVIDIA/Megatron-LM’s past year of commit activity
  Ongoing research training transformer models at scale

  transformersmodel-paralarge-language-models

  transformersmodel-paralarge-language-models

  Python

  •

  Other

  •4.3k4.3k forks•17k17k stars•374374 issues•748748 pull requests•Updated Aug 7, 2026Aug 7, 2026

• cuopt

  Public

  NVIDIA/cuopt’s past year of commit activity
  GPU accelerated decision optimization

  gpuoptimizationcuda+ 1

  gpuoptimizationcudalinear-programming

  Cuda

  •

  Apache License 2.0

  •218218 forks•1k1k stars•153153 issues•4545 pull requests•Updated Aug 7, 2026Aug 7, 2026

• earth2studio

  Public

  NVIDIA/earth2studio’s past year of commit activity
  Open-source deep-learning framework for exploring, building and deploying AI weather/climate workflows.

  weatheraideep-learning+ 1

  weatheraideep-learningclimate-science

  Python

  •

  Apache License 2.0

  •240240 forks•1.1k1.1k stars•2121 issues•1717 pull requests•Updated Aug 7, 2026Aug 7, 2026

• OpenShell

  Public

  NVIDIA/OpenShell’s past year of commit activity
  OpenShell is the safe, private runtime for autonomous AI agents.

  Rust

  •

  Apache License 2.0

  •1.2k1.2k forks•8k8k stars•315315 issues•116116 pull requests•Updated Aug 7, 2026Aug 7, 2026

• TensorRT-LLM

  Public
  TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inferen…

  cudapytorchmoe+ 2

  cudapytorchmoeblackwellllm-serving

  Python

  •

  Other

  •2.6k2.6k forks•14k14k stars•612612 issues•1k1k pull requests•Updated Aug 7, 2026Aug 7, 2026

• srt-slurm

  Public

  NVIDIA/srt-slurm’s past year of commit activity
  NVIDIA Inference Benchmarks provide recipes in ready-to-use templates for evaluating platform speed. Validate your platform across specific AI use cases across…

  Python

  •

  Other

  •7575 forks•4040 stars•1010 issues•6969 pull requests•Updated Aug 7, 2026Aug 7, 2026

• OSMO

  Public

  NVIDIA/OSMO’s past year of commit activity
  The developer-first platform for scaling complex Physical AI workloads across heterogeneous compute—unifying training GPUs, simulation clusters, and edge device…

  Python

  •

  Apache License 2.0

  •4848 forks•199199 stars•7575 issues•3535 pull requests•Updated Aug 7, 2026Aug 7, 2026

• cccl

  Public

  NVIDIA/cccl’s past year of commit activity
  CUDA Core Compute Libraries

  cpphpcgpu+ 16

  cpphpcgpumodern-cppparallel-computingcudanvidiagpu-accelerationcuda-kernelsgpu-computing+ 9

  C++

  •

  Other

  •455455 forks•2.5k2.5k stars•1.7k1.7k issues•300300 pull requests•Updated Aug 7, 2026Aug 7, 2026

• skills

  Public

  NVIDIA/skills’s past year of commit activity
  Agent Skills for NVIDIA products — install into Claude Code, Codex, and other coding agents to run Physical AI, robotics, simulation, CUDA, and RAG workflows en…

  roboticssimulationcuda+ 11

  roboticssimulationcudadeveloper-toolscodexai-agentsragomniversellmcoding-agents+ 4

  Python

  •

  Apache License 2.0

  •331331 forks•2.8k2.8k stars•66 issues•77 pull requests•Updated Aug 7, 2026Aug 7, 2026

• OpenShell-Research

  Public

  NVIDIA/OpenShell-Research’s past year of commit activity
  🧪 OpenShell's Research Journal

  Python

  •

  Apache License 2.0

  •11 fork•55 stars•00 issues•11 pull request•Updated Aug 7, 2026Aug 7, 2026

• Model-Optimizer

  Public

  NVIDIA/Model-Optimizer’s past year of commit activity
  A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It comp…

  Python

  •

  Apache License 2.0

  •536536 forks•3.4k3.4k stars•8282 issues•257257 pull requests•Updated Aug 7, 2026Aug 7, 2026

• NeMo-Retriever

  Public
  NeMo Retriever Library is a scalable, performance-oriented document content and metadata extraction microservice. NeMo Retriever Library uses specialized NVIDIA…

  Python

  •

  Apache License 2.0

  •341341 forks•3k3k stars•126126 issues•9999 pull requests•Updated Aug 7, 2026Aug 7, 2026

• MatX

  Public
  An efficient C++20 GPU numerical computing library with Python-like syntax

  hpcgpucuda+ 2

  hpcgpucudagpgpugpu-computing

  C++

  •

  BSD 3-Clause "New" or "Revised" License

  •123123 forks•1.4k1.4k stars•1212 issues•66 pull requests•Updated Aug 7, 2026Aug 7, 2026

• NVSentinel

  Public
  NVSentinel is a cross-platform fault remediation service designed to rapidly remediate runtime node-level issues in GPU-accelerated computing environments

  Go

  •

  Apache License 2.0

  •105105 forks•365365 stars•4949 issues•2424 pull requests•Updated Aug 7, 2026Aug 7, 2026

• ai-cloud-validation

  Public
  Validation and management tools for NVIDIA AI Cloud environments.

  testingkubernetesvalidation+ 3

  testingkubernetesvalidationgpuslurmnvidia

  Python

  •

  Apache License 2.0

  •2323 forks•2323 stars•3939 issues•77 pull requests•Updated Aug 7, 2026Aug 7, 2026

• elements

  Public

  NVIDIA/elements’s past year of commit activity
  NVIDIA Design System and UI Agent Harness for AI/ML Factories, Robotics, and Autonomous Vehicles

  reactweb-componentcli+ 15

  reactweb-componentcliangulartypescriptuiaivueroboticsmcp+ 8

  TypeScript

  •

  Apache License 2.0

  •99 forks•3333 stars•00 issues•33 pull requests•Updated Aug 7, 2026Aug 7, 2026