Skip to content

@NVIDIA NVIDIA Corporation

Change the repository type filter

All

    Repositories list

    639 repositories

    • Megatron-LM

      Public
      Ongoing research training transformer models at scale
      transformersmodel-paralarge-language-models
      Python
      •
      Other
      •3.4k•14k•330•239•Updated Dec 7, 2025Dec 7, 2025

    • tilus

      Public
      Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.
      tileprogrammingkernelcuda
      Python
      •
      Apache License 2.0
      •12•414•8•0•Updated Dec 7, 2025Dec 7, 2025

    • KAI-Scheduler

      Public
      KAI Scheduler is an open source Kubernetes Native scheduler for AI workloads at large scale
      Go
      •
      Apache License 2.0
      •114•956•23•36•Updated Dec 7, 2025Dec 7, 2025

    • torch-harmonics

      Public
      Differentiable signal processing on the sphere for PyTorch
      machine-learningsignal-processingspherepytorch
      Jupyter Notebook
      •
      Other
      •62•610•3•4•Updated Dec 7, 2025Dec 7, 2025

    • aistore

      Public
      AIStore: scalable storage for AI applications
      kuberneteshigh-performancedistributed-storagehigh-availabilityobject-storagemulti-cloudbatch-jobss3-compatiblemultipart-uploadml-training
      Go
      •
      MIT License
      •228•1.7k•0•0•Updated Dec 7, 2025Dec 7, 2025

    • Model-Optimizer

      Public
      A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
      Python
      •
      Apache License 2.0
      •206•1.6k•68•44•Updated Dec 7, 2025Dec 7, 2025

    • cuda-python

      Public
      CUDA Python: Performance meets Productivity
      Python
      •
      Other
      •227•3.1k•205•15•Updated Dec 7, 2025Dec 7, 2025

    • 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 by minimizing the downtime due to failures and interruptions.
      Python
      •
      Other
      •37•239•1•15•Updated Dec 7, 2025Dec 7, 2025

    • accelerated-computing-hub

      Public
      NVIDIA curated collection of educational resources related to general purpose GPU programming.
      Jupyter Notebook
      •
      Other
      •167•926•13•4•Updated Dec 7, 2025Dec 7, 2025

    • Fuser

      Public
      A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")
      C++
      •
      Other
      •69•364•205•212•Updated Dec 7, 2025Dec 7, 2025

    • 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 inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.
      cudapytorchmoeblackwellllm-serving
      Python
      •
      Other
      •1.9k•12k•618•453•Updated Dec 7, 2025Dec 7, 2025

    • k8s-dra-driver-gpu

      Public
      NVIDIA DRA Driver for GPUs
      Go
      •
      Apache License 2.0
      •101•504•94•27•Updated Dec 7, 2025Dec 7, 2025

    • TensorRT-Incubator

      Public
      Experimental projects related to TensorRT
      MLIR
      •19•116•37•12•Updated Dec 7, 2025Dec 7, 2025

    • k8s-nim-operator

      Public
      An Operator for deployment and maintenance of NVIDIA NIMs and NeMo microservices in a Kubernetes environment.
      Go
      •
      Apache License 2.0
      •34•140•7•29•Updated Dec 7, 2025Dec 7, 2025

    • NV-Kernels

      Public
      Ubuntu kernels which are optimized for NVIDIA server systems
      C
      •
      Other
      •43•70•0•11•Updated Dec 7, 2025Dec 7, 2025

    • JAX-Toolbox

      Public
      JAX-Toolbox
      Python
      •
      Apache License 2.0
      •67•367•80•43•Updated Dec 7, 2025Dec 7, 2025

    • TileGym

      Public
      Helpful kernel tutorials and examples for tile-based GPU programming
      Python
      •
      Other
      •9•252•0•0•Updated Dec 7, 2025Dec 7, 2025

    • go-gpuallocator

      Public
      Go Abstraction for Allocating NVIDIA GPUs with Custom Policies
      Go
      •
      Apache License 2.0
      •26•119•5•7•Updated Dec 7, 2025Dec 7, 2025

    • nvidia-container-toolkit

      Public
      Build and run containers leveraging NVIDIA GPUs
      Go
      •
      Apache License 2.0
      •446•3.9k•117•23•Updated Dec 7, 2025Dec 7, 2025

    • mig-parted

      Public
      MIG Partition Editor for NVIDIA GPUs
      Go
      •
      Apache License 2.0
      •52•231•22•16•Updated Dec 7, 2025Dec 7, 2025

    • 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
      •25•99•40•7•Updated Dec 7, 2025Dec 7, 2025

    • k8s-device-plugin

      Public
      NVIDIA device plugin for Kubernetes
      kubernetes
      Go
      •
      Apache License 2.0
      •758•3.6k•75•34•Updated Dec 7, 2025Dec 7, 2025

    • cuCollections

      Public
      datastructurescppgpucudahashmapcpp17hashsethashtable
      C++
      •
      Apache License 2.0
      •101•598•55•13•Updated Dec 7, 2025Dec 7, 2025

    • gpu-operator

      Public
      NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes
      kubernetesgpucudanvidia
      Go
      •
      Apache License 2.0
      •420•2.4k•94•67•Updated Dec 7, 2025Dec 7, 2025

    • cuopt

      Public
      GPU accelerated decision optimization
      gpuoptimizationcudalinear-programming
      Cuda
      •
      Apache License 2.0
      •100•601•72•26•Updated Dec 7, 2025Dec 7, 2025

    • cloud-native-docs

      Public
      Documentation repository for NVIDIA Cloud Native Technologies
      kubernetescontainerskubernetes-operator
      PowerShell
      •
      Apache License 2.0
      •33•31•5•10•Updated Dec 7, 2025Dec 7, 2025

    • vgpu-device-manager

      Public
      NVIDIA vGPU Device Manager manages NVIDIA vGPU devices on top of Kubernetes
      Go
      •
      Apache License 2.0
      •22•152•0•16•Updated Dec 7, 2025Dec 7, 2025

    • stdexec

      Public
      `std::execution`, the proposed C++ framework for asynchronous and parallel programming.
      C++
      •
      Apache License 2.0
      •217•2.1k•112•11•Updated Dec 6, 2025Dec 6, 2025

    • nvbench

      Public
      CUDA Kernel Benchmarking Library
      benchmarkperformancegpucudanvidiacuda-kernelskernel-benchmark
      Cuda
      •
      Apache License 2.0
      •94•773•53•8•Updated Dec 6, 2025Dec 6, 2025

    • cccl

      Public
      CUDA Core Compute Libraries
      cpphpcgpumodern-cppparallel-computingcudanvidiagpu-accelerationcuda-kernelsgpu-computing
      C++
      •
      Other
      •297•2.1k•1.1k•195•Updated Dec 6, 2025Dec 6, 2025