Repositories list

• NemoClaw

  Public

  NVIDIA/NemoClaw’s past year of commit activity
  Run OpenClaw more securely inside NVIDIA OpenShell with managed inference

  TypeScript

  •

  Apache License 2.0

  •2.4k2.4k forks•20k20k stars•215215 issues•151151 pull requests•Updated Apr 22, 2026Apr 22, 2026

• context-aware-rag

  Public

  NVIDIA/context-aware-rag’s past year of commit activity
  Context-Aware RAG library for Knowledge Graph ingestion and retrieval functions.

  agentvideo-summarizationvideo-understanding+ 9

  agentvideo-summarizationvideo-understandingvlmraglarge-language-modelsllmlong-video-understandingretrieval-augmented-generationgraphrag+ 2

  Python

  •

  Apache License 2.0

  •2121 forks•7777 stars•11 issue•22 pull requests•Updated Apr 22, 2026Apr 22, 2026

• cuda-quantum

  Public
  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

  •363363 forks•1k1k stars•426426 issues•121121 pull requests•Updated Apr 22, 2026Apr 22, 2026

• TransformerEngine

  Public
  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 Blackwel…

  pythonmachine-learningdeep-learning+ 6

  pythonmachine-learningdeep-learninggpucudapytorchjaxfp8fp4

  Python

  •

  Apache License 2.0

  •702702 forks•3.3k3.3k stars•230230 issues•124124 pull requests•Updated Apr 22, 2026Apr 22, 2026

• aicr

  Public
  Tooling for optimized, validated, and reproducible GPU-accelerated AI runtime in Kubernetes

  configkubernetesmanifest+ 5

  configkubernetesmanifestairuntimegpuhelmargocd

  Go

  •

  Apache License 2.0

  •3333 forks•270270 stars•2626 issues•88 pull requests•Updated Apr 22, 2026Apr 22, 2026

• nccl

  Public

  NVIDIA/nccl’s past year of commit activity
  Optimized primitives for collective multi-GPU communication

  deep-learningcppgpu+ 3

  deep-learningcppgpucudanvidiacommunications

  C++

  •

  Other

  •1.2k1.2k forks•4.6k4.6k stars•235235 issues•8989 pull requests•Updated Apr 22, 2026Apr 22, 2026

• IsaacTeleop

  Public
  The unified framework for sim & real robot teleoperation

  Python

  •

  Apache License 2.0

  •1515 forks•150150 stars•4242 issues•2121 pull requests•Updated Apr 22, 2026Apr 22, 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

  •3.9k3.9k forks•16k16k stars•353353 issues•366366 pull requests•Updated Apr 22, 2026Apr 22, 2026

• ncx-infra-controller-core

  Public

  NVIDIA/ncx-infra-controller-core’s past year of commit activity
  NCX Infra Controller - Hardware Lifecycle Management and multitenant networking

  Rust

  •

  Apache License 2.0

  •8383 forks•127127 stars•168168 issues•6464 pull requests•Updated Apr 21, 2026Apr 21, 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

  •2.3k2.3k forks•13k13k stars•582582 issues•768768 pull requests•Updated Apr 21, 2026Apr 21, 2026

• multi-storage-client

  Public
  Unified high-performance Python client for object and file stores.

  Python

  •

  Apache License 2.0

  •1313 forks•6565 stars•22 issues•11 pull request•Updated Apr 21, 2026Apr 21, 2026

• Isaac-GR00T

  Public

  NVIDIA/Isaac-GR00T’s past year of commit activity
  NVIDIA Isaac GR00T N1.7 - A Foundation Model for Generalist Robots.

  Python

  •

  Apache License 2.0

  •1.1k1.1k forks•6.8k6.8k stars•186186 issues•7777 pull requests•Updated Apr 21, 2026Apr 21, 2026

• aistore

  Public
  AIStore: scalable storage for AI applications

  kuberneteshigh-performancedistributed-storage+ 9

  kuberneteshigh-performancedistributed-storagehigh-availabilityobject-storagemulti-cloudbatch-jobss3-compatiblemultipart-uploadml-training+ 2

  Go

  •

  MIT License

  •246246 forks•1.8k1.8k stars•22 issues•11 pull request•Updated Apr 21, 2026Apr 21, 2026

• topograph

  Public
  A toolkit for discovering cluster network topology.

  Go

  •

  Apache License 2.0

  •2020 forks•117117 stars•66 issues•33 pull requests•Updated Apr 21, 2026Apr 21, 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

  •379379 forks•2.3k2.3k stars•1.3k1.3k issues•256256 pull requests•Updated Apr 21, 2026Apr 21, 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

  •7272 forks•260260 stars•3232 issues•1515 pull requests•Updated Apr 21, 2026Apr 21, 2026

• NeMo-Retriever

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

  Python

  •

  Apache License 2.0

  •315315 forks•2.9k2.9k stars•126126 issues•7171 pull requests•Updated Apr 21, 2026Apr 21, 2026

• cudaqx

  Public
  Accelerated libraries for quantum-classical computing built on CUDA-Q.

  C++

  •

  Other

  •5858 forks•9696 stars•3333 issues•2323 pull requests•Updated Apr 21, 2026Apr 21, 2026

• cuCollections

  Public
  datastructurescppgpu+ 5

  datastructurescppgpucudahashmapcpp17hashsethashtable

  C++

  •

  Apache License 2.0

  •107107 forks•636636 stars•5454 issues•1414 pull requests•Updated Apr 21, 2026Apr 21, 2026

• cuEquivariance

  Public

  NVIDIA/cuEquivariance’s past year of commit activity
  cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and …

  Python

  •2929 forks•387387 stars•1717 issues•22 pull requests•Updated Apr 21, 2026Apr 21, 2026

• NVFlare

  Public
  NVIDIA Federated Learning Application Runtime Environment

  pythondecentralizedpet+ 4

  pythondecentralizedpetprivacy-protectionfederated-learningfederated-analyticsfederated-computing

  Python

  •

  Apache License 2.0

  •249249 forks•922922 stars•1515 issues•1717 pull requests•Updated Apr 21, 2026Apr 21, 2026

• trt-samples-for-hackathon-cn

  Public
  Simple samples for TensorRT programming

  Python

  •

  Apache License 2.0

  •349349 forks•1.7k1.7k stars•6565 issues•22 pull requests•Updated Apr 21, 2026Apr 21, 2026

• srt-slurm

  Public
  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

  •2020 forks•1515 stars•44 issues•88 pull requests•Updated Apr 21, 2026Apr 21, 2026

• tilus

  Public
  Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.

  tileprogrammingkernel+ 1

  tileprogrammingkernelcuda

  Python

  •

  Apache License 2.0

  •2424 forks•477477 stars•55 issues•11 pull request•Updated Apr 21, 2026Apr 21, 2026

• ncx-infra-controller-rest

  Public
  NCX Infra Controller - Hardware Lifecycle Management (REST API)

  Go

  •

  Apache License 2.0

  •3232 forks•3434 stars•3030 issues•1313 pull requests•Updated Apr 21, 2026Apr 21, 2026

• nvidia-resiliency-ext

  Public
  NVIDIA Resiliency Extension is a python package for framework developers and users to implement fault-tolerant features. It improves the effective training time…

  Python

  •

  Other

  •5050 forks•284284 stars•44 issues•1919 pull requests•Updated Apr 21, 2026Apr 21, 2026

• fleet-intelligence-agent

  Public
  NVIDIA Fleet Intelligence Agent - Host agent for GPU telemetry collection and attestation

  Go

  •

  Apache License 2.0

  •11 fork•1414 stars•00 issues•33 pull requests•Updated Apr 21, 2026Apr 21, 2026

• makani

  Public
  Massively parallel training of machine-learning based weather and climate models

  Python

  •

  Other

  •7272 forks•373373 stars•55 issues•33 pull requests•Updated Apr 21, 2026Apr 21, 2026

• OSMO

  Public
  The developer-first platform for scaling complex Physical AI workloads across heterogeneous compute—unifying training GPUs, simulation clusters, and edge device…

  TypeScript

  •

  Apache License 2.0

  •3434 forks•146146 stars•6161 issues•1616 pull requests•Updated Apr 21, 2026Apr 21, 2026

• physicsnemo

  Public
  Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods

  machine-learningdeep-learningphysics+ 3

  machine-learningdeep-learningphysicspytorchnvidia-gpunvidia-warp

  Python

  •

  Apache License 2.0

  •644644 forks•2.7k2.7k stars•2121 issues•3838 pull requests•Updated Apr 21, 2026Apr 21, 2026