Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Failed to load latest commit information.
Latest commit message
Commit time

Stanford Nimble Logo


Stanford Nimble

pip3 install nimblephysics


Read our docs and the paper.

Use physics as a non-linearity in your neural network. A single timestep, nimble.timestep(state, controls), is a valid PyTorch function.

Forward pass illustration

We support an analytical backwards pass, that works even through contact and friction.

Backpropagation illustration

It's as easy as:

from nimble import timestep

# Everything is a PyTorch Tensor, and this is differentiable!!
next_state = timestep(world, current_state, control_forces)

Nimble started life as a fork of the popular DART physics engine, with analytical gradients and a PyTorch binding. We've worked hard to maintain as much backwards compatability as we can, so many simulations that worked in DART should translate directly to Nimble.

Check out our website for more information.

Installing on Arm64 Macs

We don't yet publish Arm64 binaries to PyPI from our CI system, so you may not be able to pip3 install nimblephysics from a new Arm64 Mac. We will endeavor to manually push binaries occassionally, but until GitHub Actions supports using Arm64 Mac runners, the solution is to clone this repo, then run

  • ci/mac/
  • ci/mac/ That will install the dependencies you need, and then build and install the Python package. Please create Issues if you run into problems, and we'll do our best to fix them.