Public GitHub footprint of NVIDIA Corporation

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

Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills before you install them.

OpenShell is the safe, private runtime for autonomous AI agents.

Ongoing research training transformer models at scale

NVIDIA Linux open GPU kernel module source

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.

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

PersonaPlex code.

CUDA Templates and Python DSLs for High-Performance Linear Algebra

cuDF - GPU DataFrame Library

the LLM vulnerability scanner

Pytorch implementation of our method for high-resolution (e.g. 2048x1024) photorealistic video-to-video translation.

NVIDIA Isaac GR00T N1.7 - A Foundation Model for Generalist Robots.

A Python framework for GPU-accelerated simulation, robotics, and machine learning.

A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.

NVIDIA cuML: GPU-Accelerated Machine Learning

Optimized primitives for collective multi-GPU communication

Build and run containers leveraging NVIDIA GPUs

NVIDIA's CUDA platform for Rust. Host runtime crates plus Tile (cutile-rs) and SIMT (cuda-oxide) kernel programming models in idiomatic Rust.

Agent Skills for NVIDIA products — install into Claude Code, Codex, and other coding agents to run Physical AI, robotics, simulation, CUDA, and RAG workflows end to end.

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 Blackwell GPUs, to provide better performance with lower memory utilization in both training and inference.

CUDA Python: Performance meets Productivity

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

NeMo Retriever Library is a scalable, performance-oriented document content and metadata extraction microservice. NeMo Retriever Library uses specialized NVIDIA NIM microservices to find, contextualize, and extract text, tables, charts and images that you can use in downstream generative applications.

NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes

The NVIDIA NeMo Agent toolkit is an open-source library for efficiently connecting and optimizing teams of AI agents.

CUDA Core Compute Libraries

`std::execution`, the standard C++ framework for asynchronous and parallel programming.

NVIDIA curated collection of educational resources related to general purpose GPU programming.

AIStore: scalable storage for AI applications

A suite of image and video neural tokenizers

This repo contains the source code for RULER: What’s the Real Context Size of Your Long-Context Language Models?

Collection of step-by-step playbooks for setting up AI/ML workloads on NVIDIA DGX Spark devices with Blackwell architecture.

Open-source deep-learning framework for exploring, building and deploying AI weather/climate workflows.

C++ and Python support for the CUDA Quantum programming model for heterogeneous quantum-classical workflows

GPU accelerated decision optimization

NVIDIA cuDF for Apache Spark plugin - accelerate Apache Spark with GPUs

NVIDIA Federated Learning Application Runtime Environment

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

cuVS - a library for vector search and clustering on the GPU

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs

No description provided for this repository.

DreamGen: Nvidia GEAR Lab's initiative to solve the robotics data problem using world models

Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.

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

Shared capture infrastructure for teleoperation and egocentric data.

NVSentinel detects and remediates GPU faults on Kubernetes nodes

This repo contains CUDA-Q Academic materials, including self-paced Jupyter notebook modules for building and optimizing hybrid quantum-classical algorithms using CUDA-Q.

Foundation Models for Structured Data

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.

NVIDIA Kaggle Plugin gives agents end-to-end Kaggle competition workflows through a single skill, nvidia-kaggle-skill. It can gather competition context, study public writeups and notebooks, reproduce kernels locally, submit to competitions, and manage Ka

OptiX + GL based implementations of ANARI

The developer-first platform for scaling complex Physical AI workloads across heterogeneous compute—unifying training GPUs, simulation clusters, and edge devices in a simple YAML

Platform for deploying and routing GPU-accelerated inference, streaming, and batch workloads at scale.

Multi-language agent runtime and library for execution scope management, lifecycle events, and middleware on tool and LLM calls.

NeMo-Speech.cpp is a lightweight C++ inference runtime for Speech models

Auto Ontology is a Python framework for building semantic, graph-based reasoning layers over data. It enables deterministic query generation, better context retrieval, and more explainable AI for LLM and agent workflows.

TensorIR is a lightweight NVIDIA-owned MLIR compiler frontend for expressing tensor computations and lowering them to NVIDIA CUDA Tile IR.

SimReady Foundation is a central repository for defining simulation content specifications based on various runtime use cases.

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

Ubuntu kernels which are optimized for NVIDIA server systems

NVIDIA TensorRT-RTX is an SDK for high-performance AI inference on NVIDIA RTX GPUs. This repository contains Open-Source Software components of TensorRT-RTX.

The UI SDK and agent-ready developer toolkit for AI factories, robotics, and autonomous vehicles.

CloudAI Benchmark Framework

:coffee:️ Simulate NVIDIA GPU infrastructure on CPU nodes in Kubernetes

XR AI

No description provided for this repository.

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

knavigator is a development, testing, and optimization toolkit for AI/ML scheduling systems at scale on Kubernetes.

NVIDIA Cluster Readiness Engine

A Kubernetes Operator to manage Node OS customizations.

NVIDIA’s repository for enabling trustworthy AI.

nvloom is a set of tools designed to scalably test MNNVL fabrics.

NVIDIA cuDF plugin JNI For Apache Spark

Linux kernel source tree

NVIDIA NeMo Fabric

NVIDIA Omniverse NuRec agent skills automate neural reconstruction for autonomous vehicle and robotics simulation, running the reconstruction and rendering workflow end to end.

DAQIRI connects high bandwidth streaming sensor data to the NVIDIA software ecosystem and is part of NVIDIA's Holoscan platform for streaming, real time AI data processing

🧪 OpenShell's Research Journal

Deploy AI agents to investigate cluster issues and streamline governed AI factory operations.

Workflow orchestration framework for end-to-end synthetic data generation (SDG), training (SFT), and evaluation pipelines built on NVIDIA's NeMo ecosystem

Orchestrator for LLM coding agents in Docker and Kubernetes pods

AI Agent Skills for NVIDIA Broadcast app

CUDA-accelerated Python tools for astronomical and multi-dimensional data access, image subtraction, warping, candidate classification, and scientific review.

The open-source model suite for robotics data engine by NVIDIA GEAR

Kumo Engine is an accelerator for context generation over structured data, designed for teams building and running predictive models in production at scale.

Accelerated algorithm libraries for quantum-classical computing built on CUDA-Q

MCTP VDM-based Nvidia System Management API

NVIDIA Digital Health skills repo for agent-guided healthcare AI workflows, including evaluation, data generation, model adaptation, deployment guidance, and developer best practices.

A do everything Redfish, KVM, GUI, and DBus webserver for OpenBMC

Packaging and distributing agent skills and agent plugins as OCI artifacts.

D-Bus configurable sensor scanning applications

One Python client for NVIDIA's structured-data foundation models: ask a predictive question of one table or of a whole relational schema, and get an answer without training anything.

Visualization Engine & Libraries on ANARI

A machine learning compiler for GPUs, CPUs, and ML accelerators

No description provided for this repository.

YAML descriptors of standard dbus interfaces

Run-time JSON driven system configuration manager

Rust-based BUFR (Binary Universal Form for the Representation of meteorological data) file parser