newton
A GPU-accelerated physics engine for robotics simulation
Decision gist · record as of 2026-08-14
Yes, if you are a roboticist or simulation researcher with access to an NVIDIA GPU and need GPU-accelerated physics with differentiability and extensibility. The low install friction, active maintenance, permissive license, and comprehensive example suite make it a solid choice. No security vulnerabilities reported. Consider if you require CPU-only simulation on non-macOS platforms or need physics engines not covered by the provided examples.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10+; NVIDIA GPU with Maxwell architecture or newer and driver 545+, or runs CPU-only on macOS.
- Low friction installation via pip; requires Python 3.10+, NVIDIA GPU with driver 545 or newer (CUDA 12), or CPU-only on macOS.
- Active maintenance with release 3 days old.
License · maintenance · safety
Apache-2.0 (permissive) — Apache-2.0 permissive license allows commercial and private use with minimal restrictions; documentation separately licensed CC-BY-4.0.
last release 2026-08-11 (3 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 142,588 downloads/mo, #11,205 on PyPI
Alternatives
Verify before relying
pip install "newton[examples]"
python -m newton.examples basic_pendulum- Whether warp-lang dependency is automatically installed or requires separate setup.
- Performance characteristics and simulation scale limits compared to other physics engines.
- Availability and maturity of the OpenUSD integration mentioned in the description.
What it is and what it does
Newton is a GPU-accelerated physics simulation engine purpose-built for robotics and simulation research. It extends NVIDIA Warp's simulation capabilities and integrates MuJoCo Warp as its primary backend, emphasizing GPU-based computation, differentiability, and extensibility. The engine supports a wide range of simulation scenarios including rigid bodies, soft bodies, cloth, cables, and material point method (MPM) simulations, with built-in support for OpenUSD and multiple visualization backends.
The package is designed for rapid iteration in robotics research, offering examples for robot control, inverse kinematics, contact dynamics, and multi-physics coupling. It requires Python 3.10+, an NVIDIA GPU with Maxwell architecture or newer (driver 545+), or can run CPU-only on macOS. Installation is straightforward via pip, with optional examples available. The project is community-maintained under Linux Foundation stewardship and was initiated by Disney Research, Google DeepMind, and NVIDIA.
Use it for
- Simulate robot dynamics and control policies for humanoid and legged robots like Franka, H1, and ANYmal.
- Develop and test inverse kinematics solutions with GPU acceleration for real-time performance.
- Model cloth and cable interactions in manipulation tasks, such as pick-and-place with deformable objects.
- Run differentiable simulations for gradient-based optimization of robot morphologies or control parameters.
- Prototype multi-physics scenarios combining rigid bodies, soft bodies, and granular materials in a single simulation.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are a roboticist or simulation researcher with access to an NVIDIA GPU and need GPU-accelerated physics with differentiability and extensibility.
The low install friction, active maintenance, permissive license, and comprehensive example suite make it a solid choice. No security vulnerabilities reported. Consider if you require CPU-only simulation on non-macOS platforms or need physics engines not covered by the provided examples.
Install
newton on PyPI
Before you install
Low friction installation via pip; requires Python 3.10+, NVIDIA GPU with driver 545 or newer (CUDA 12), or CPU-only on macOS. Active maintenance with release 3 days old.
Requires Python 3.10+; NVIDIA GPU with Maxwell architecture or newer and driver 545+, or runs CPU-only on macOS.
License in practice
Apache-2.0 permissive license allows commercial and private use with minimal restrictions; documentation separately licensed CC-BY-4.0.
Quickstart
pip install "newton[examples]"
python -m newton.examples basic_pendulumVerify before relying
- Whether warp-lang dependency is automatically installed or requires separate setup.
- Performance characteristics and simulation scale limits compared to other physics engines.
- Availability and maturity of the OpenUSD integration mentioned in the description.
Package facts
| License | Apache-2.0 permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 1 packagewarp-lang |
| Maintenance | Actively maintained 3 days since the last release |
| First released | |
| Downloads | 142,588 / month, #11,205 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 5 - Production/StableEnvironment :: GPU :: NVIDIA CUDAEnvironment :: GPU :: NVIDIA CUDA :: 12Environment :: GPU :: NVIDIA CUDA :: 13Intended Audience :: DevelopersIntended Audience :: Science/ResearchNatural Language :: EnglishProgramming Language :: Python :: 3Programming Language :: Python :: 3 :: OnlyProgramming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Topic :: Scientific/Engineering |
Evidence: newton-1.5.0-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “NVIDIA Warp physics”
- newtonNewton is a GPU-accelerated physics simulation engine built on NVIDIA…
- mujoco-warpGPU-accelerated physics simulation for robotics using NVIDIA Warp,…
- warp-langWarp is a Python framework that JIT-compiles regular Python functions…
Give your agent the search over MCP, or paste the wish link into any chat.
More Scientific/Engineering packages
NumPy provides an N-dimensional array object and a comprehensive suite of mathematical, linear algebra, Fourier transform, and random number functions for scientific computing in Python.
pandas provides fast, flexible data structures (Series and DataFrame) for loading, cleaning, transforming, and analyzing labeled or relational data in Python.
scipy provides numerical algorithms for mathematics, science, and engineering—including optimization, integration, linear algebra, Fourier transforms, signal and image processing, and ODE solvers—built on numpy arrays.
scikit-learn provides a comprehensive Python library for supervised and unsupervised machine learning, including classification, regression, clustering, dimensionality reduction, and model evaluation tools built on NumPy and SciPy.
Install it if you need to train, evaluate, or deploy supervised or unsupervised learning models.
dill extends Python's pickle module to serialize and deserialize a much wider range of Python objects, including functions, lambdas, classes, and interpreter sessions, to byte streams for storage or network transmission.
Multiprocess is an enhanced fork of Python's standard multiprocessing library that uses dill for better serialization, allowing you to spawn processes with a threading-like API and share complex objects between them.
Install it if you use multiprocessing and encounter pickle serialization limits with lambdas or complex objects.
See also mujoco-warp · newton-actuators · quadrants · sapien · warp-lang · mjlab · mjviser · newton-usd-schemas · pybullet · pin