Accelerate Computational Engineering and Robotics With NVIDIA Warp
Learn how NVIDIA Warp, an open-source CUDA-X library for accelerating computational engineering, simulation, and robotics workflows, is leveraged by developers and lead to 10M downloads. In this livestream, we’ll introduce NVIDIA Warp, why developers are using it, and how it accelerates engineering and robotics applications. We’ll walk through practical use cases, including an industrial engineering example and a robotics workflow, to showcase how Warp can help teams build and test high-performance simulation code faster. We’ll also look at the types of developers using Warp today, from researchers and robotics engineers to simulation teams and computational engineering developers. By the end, viewers will understand where NVIDIA Warp applies, how it can fit into their own workflows, and how to download NVIDIA Warp and start building. You’ll learn: - How NVIDIA Warp turns Python functions into high-performance CPU and GPU kernels - How Warp interoperates with PyTorch, JAX, and other array-based frameworks - How automatic differentiation and differentiable simulation work in Warp - How Warp supports geometry processing, sparse computation, FEM, and scientific workloads - How to move Warp code from Python development into robotics deployment environments - How Newton builds on Warp for extensible robot physics and multiphysics simulation - How Warp and Newton integrate with NVIDIA Isaac Sim and Isaac Lab - How developers are using Warp across robotics, CAE, semiconductor, CFD, research, and AI physics ---------------------------------------------------------------------------- Chapters: 0:00 NVIDIA Warp and Newton Livestream Introduction 2:42 Meet the NVIDIA Warp Team 4:31 NVIDIA Warp Origin Story 10:06 From DFlex to NVIDIA Warp 15:42 NVIDIA Warp Technical Deep Dive 18:54 PyTorch, JAX, and Framework Interoperability 24:20 Warp Execution Model and Tile-Based Programming 26:39 Automatic Differentiation and Differentiable Simulation 28:23 Geometry Processing and Warp Data Structures 29:24 Warp Compilation Pipeline 31:03 Deploying Warp From Python to Robotics Hardware 32:23 Warp With Isaac Sim and Isaac Lab 34:30 Native C++ and CUDA Inside Warp 35:37 Gradient and Warp Debugging Tools 38:49 NVIDIA Warp Developer Use Cases 42:08 Newton Physics for Robot Learning 44:45 Newton Multiphysics Simulation 45:26 Differentiable Physics in Newton 47:09 Collision, Contact, and Tactile Simulation 48:35 Warp-Based Rendering for Robot Learning 50:30 Robotics and Industrial Warp Use Cases 53:42 Warp Sim Data for Computational Engineering 55:14 CFD and AI Physics With Warp 56:48 Granular Simulation and Deformable Terrain 58:29 GPU-Accelerated CAD Geometry Processing 1:00:01 NVIDIA Warp in Academic Research 1:02:06 Where Warp Ends and Newton Begins 1:07:34 Zero-Copy Memory and Framework Interoperability 1:09:07 Hardware Requirements and Large-Scale Simulation 1:13:24 Warp and Newton in Isaac Lab 1:15:14 Custom Contact Models for Robot Learning 1:17:02 Differentiable Contact and Collision 1:19:00 Key Takeaways From the NVIDIA Warp Team 1:23:26 NVIDIA Developer and Community News 1:29:35 Closing and Next Steps ---------------------------------------------------------------------------- List of Resources - NVIDIA Warp - https://developer.nvidia.com/warp-python - NVIDIA Warp GitHub - https://github.com/nvidia/warp - NVIDIA Warp Workshop - https://www.nvidia.com/en-us/on-demand/session/gtc26-dlit81837/ - Autodesk Research Brings Warp Speed to Computational Fluid Dynamics on NVIDIA GH200 - https://developer.nvidia.com/blog/autodesk-research-brings-warp-speed-to-computational-fluid-dynamics-on-nvidia-gh200/ - NVIDIA Isaac LEAPP - https://github.com/nvidia-isaac/leapp Continue the discussion and ask follow-up questions in the NVIDIA Omniverse Discord thread: https://discord.com/channels/827959428476174346/1544789795958427658 ---------------------------------------------------------------------------- 📆 Check out the full calendar for all of our upcoming events → https://nvda.ws/3JqaWnA ---------------------------------------------------------------------------- ⬇️Get Started → https://nvda.ws/4cZAZO1 👀Explore OpenUSD → https://nvda.ws/3CeozBQ 👥Join the Community → https://nvda.ws/3ZMfc6e #NVIDIA #NVIDIAWarp #NewtonPhysics #NVIDIAIsaac #IsaacSim #IsaacLab #Robotics #PhysicalAI #ComputationalEngineering #GPUComputing #DifferentiableSimulation #CUDA #Simulation #OpenUSD #DeveloperTools