# NVIDIA Omniverse Libraries Omniverse is NVIDIA’s collection of libraries and microservices for physical AI development. ## Get Started With Omniverse Libraries Leveraging NVIDIA’s deep expertise in accelerated computing and AI, [NVIDIA Omniverse™ libraries](https://www.nvidia.com/en-us/omniverse/) enable software makers to integrate pre-built functionality into their solutions. ## Omniverse Core Libraries and SDKs ### ovrtx (Early Access) GPU-accelerated, physically based rendering and sensor-simulation libraries built on NVIDIA RTX™. [Get Started](https://github.com/nvidia-omniverse/ovrtx) ### ovphysx (Early Access) USD-native, open source multiphysics library for scalable robotics and digital twin simulation. [Get Started](https://github.com/NVIDIA-Omniverse/PhysX/tree/main/ovphysx) ### ovstorage (Early Access) Cloud-native APIs for managing, storing, and synchronizing OpenUSD asset data. [Get Started](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/storage_apis) ### Kit Framework SDK Core libraries, templates, and developer tooling for building custom physical AI applications. [Explore Docs](https://docs.nvidia.com/omniverse/index.html#sdks) ### OpenUSD Libraries and SDKs ### OpenUSD Open source framework and standards that enable 3D interoperability, simulation, and collaboration in virtual worlds. [Learn More](https://developer.nvidia.com/openusd) ### OpenUSD Exchange SDK Libraries and modules to build interoperable OpenUSD data exchange solutions. [Explore Docs](https://docs.omniverse.nvidia.com/usd/code-docs/usd-exchange-sdk/latest/index.html) ### SimReady SDK Simulation-ready data framework and content pipeline for 3D pipelines **Coming Soon** ### Additional Physical AI Simulation Libraries ### Newton Physics An open source, extensible physics engine built on NVIDIA Warp and OpenUSD. [Learn More About Newton Physics Engine](https://developer.nvidia.com/newton-physics) ### NuRec APIs, libraries, and [datasets](https://huggingface.co/datasets/nvidia/PhysicalAI-Autonomous-Vehicles-NuRec) for 3D Gaussian-based neural simulation from real-world data. [Get Started](https://docs.nvidia.com/nurec/index.html#) ### PhysX Open source multi-physics SDK for scalable robotics simulation and modeling. [Go to GitHub](https://github.com/NVIDIA-Omniverse/PhysX) ## Get Started With Developer Blueprints ### “DSX” Gigawatt-Scale AI Factories Optimize design, simulation, and operations of AI factories with NVIDIA libraries. [Try Blueprint](https://build.nvidia.com/nvidia/omniverse-dsx-blueprint-for-ai-factories) [Watch Video](https://www.youtube.com/watch?v=rsBobT9INP4) ### “GR00T-Mimic” Synthetic Robot Manipulation Generate exponential synthetic manipulation data for robots. [Try Blueprint](https://build.nvidia.com/nvidia/isaac-gr00t-synthetic-manipulation) [Watch Video](https://www.youtube.com/watch?v=waZ08Z3uimk&t=50s) ### “MEGA” Multi-Robot Industrial Fleets Automation Simulate complex industrial robot fleets before real-world deployment. [Try Blueprint](https://build.nvidia.com/nvidia/mega-multi-robot-fleets-for-industrial-automation) [Watch Video](https://www.youtube.com/watch?v=IuWk0C3MzBQ) ### Physical AI Data Factory Accelerate robotics, vision AI, and autonomous vehicle development with physical AI data. [Read the Press Release](https://nvidianews.nvidia.com/news/nvidia-announces-open-physical-ai-data-factory-blueprint-to-accelerate-robotics-vision-ai-agents-and-autonomous-vehicle-development) ### Digital Twin for Interactive Fluid Simulation AI-powered virtual wind tunnel for real-time aerodynamic simulation and design. [Try Blueprint](https://build.nvidia.com/nvidia/digital-twins-for-fluid-simulation) [Watch Video](https://www.youtube.com/watch?v=RIxrN7yedyQ) ### Digital Twins for Smart Cities Build, test, and optimize visual AI agents in SimReady digital twins to monitor city-scale operations. [Try Early Access](https://docs.nvidia.com/vss/3.1.0/smartcity-docs/smartcity-toc.html) [Watch Video](https://www.youtube.com/watch?v=BJpBp2gwk2Y) ## Omniverse Learning Resources ### Build and Orchestrate End-to-End SDG Workflows with NVIDIA Isaac Sim and NVIDIA OSMO This post explores creating a simulated environment, generating synthetic data with MobilityGen on OSMO, scaling data augmentation and more. ### Synthetic Manipulation Motion Generation for Robotics Synthetic data from accurate simulations speeds up data collection and adds diversity, solving the challenge of limited real-world data for training generalizable robot learning models. ### Developers Build Fast and Reliable Robot Simulations with NVIDIA Omniverse Libraries NVIDIA Omniverse libraries and Cosmos WFMs, powered by OpenUSD, give developers tools to build physically accurate virtual worlds and AI agents that understand the real world. ### Research Advances in AI-Assisted Material Generation for Physical AI NVIDIA Cosmos WFMs and rendering tech enable AI-assisted creation of realistic, simulation-ready digital twins—accelerating physical AI training and testing at scale. ### Neural Reconstruction for Robotics and Autonomous Vehicles NVIDIA Omniverse NuRec uses multi-sensor data and neural reconstruction to bring the real world into simulation, creating photorealistic environments for testing robotics and AVs. ### Test Multi-Robot Fleets for Industrial Automation Physical AI is set to revolutionize manufacturing, logistics, and supply chains with advanced automation, intelligence, and autonomy across factories and industrial facilities. ### How to Instantly Render Real-World Scenes in Interactive Simulation With NVIDIA Omniverse NuRec and 3DGUT, turn sensor data into photorealistic 3D scenes and deploy them in Isaac Sim or CARLA—instantly, without days or weeks of work. ### Accelerating AV Simulation with Neural Reconstruction and WFMs Autonomous vehicle (AV) stacks are evolving from a hierarchy of discrete building blocks to end-to-end architectures built on foundation models. ### NVIDIA Omniverse Foundational Technology Montage NVIDIA Omniverse is a platform for integrating OpenUSD, RTX, and PhysicalAI into existing tools and workflows for industrial and robotics use. ### Industrial AI and Robotics Conference Sessions - GTC Paris Discover how AI, robotics, digital twins, and advanced simulation technologies are powering smarter, safer, and more efficient industrial automation. ### Action Graph Quickstart In this tutorial, you'll use OmniGraph in Omniverse USD Composer to move a mesh in response to a key press. ### Defect Detection This guide will point you to resources that will help you build your own tools to easily create synthetic data for your use case.