Newton Physics Documentation#

Newton is a GPU-accelerated, extensible, and differentiable physics simulation engine designed for robotics, research, and advanced simulation workflows. Built on top of NVIDIA Warp and integrating MuJoCo Warp, Newton provides high-performance simulation, modern Python APIs, and a flexible architecture for both users and developers.
Key Features#
GPU-accelerated: Leverages NVIDIA Warp for fast, scalable simulation.
Multiple solver implementations - XPBD, VBD, Mujoco, Featherstone, Euler
Modular design - Easily extendable with new solvers and components
Differentiable: Supports differentiable simulation for machine learning and optimization.
Rich Import/Export: Load models from URDF, MJCF, USD, and more.
Open Source: Maintained by Disney Research, Google DeepMind, and NVIDIA.
Learn More
For a deep conceptual introduction, see the DeepWiki Newton Physics page.
High-Level Architecture#
graph TD A[ModelBuilder] -->|builds| B[Model] B --> C[State] C --> D[Solver] D --> C B --> F[Renderer] C --> F G[Importer] --> A I[Application] --> A F --> H[Visualization]
ModelBuilder: Constructs models from primitives or imported assets.
Model: Encapsulates the physical structure, parameters, and configuration.
State: Represents the dynamic state (positions, velocities, etc.).
Solver: Advances the simulation by integrating physics.
Renderer: Visualizes the simulation in real-time or offline.
Importer: Loads models from external formats (URDF, MJCF, USD).
Quick Links#
Quickstart — Get started in minutes
Changelog — Full release notes
Frequently Asked Questions (FAQ) — Frequently asked questions
Development Guide — For developers and code contributors
Note
Newton is in active development. APIs and features may change. See the Changelog for updates.