We introduce TorchSim, an open-source atomistic simulation engine tailored for the Machine Learned Interatomic Potential (MLIP) era. By rewriting core atomistic simulation primitives in PyTorch, TorchSim can achieve orders of magnitude acceleration for popular MLIPs. Unlike existing molecular dynamics packages, which simulate one system at a time, TorchSim performs batched simulations that efficiently utilize modern GPUs by evolving multiple systems concurrently. TorchSim supports molecular dynamics integrators, structural relaxation optimizers, both machine-learned and classical interatomic potentials (such as Lennard-Jones, Morse, soft-sphere), batching with automatic memory management, differentiable simulation, and integration with popular materials informatics tools.
@article{arxiv.2508.06628,
title = {TorchSim: An efficient atomistic simulation engine in PyTorch},
author = {Orion Cohen and Janosh Riebesell and Rhys Goodall and Adeesh Kolluru and Stefano Falletta and Joseph Krause and Jorge Colindres and Gerbrand Ceder and Abhijeet S. Gangan},
journal= {arXiv preprint arXiv:2508.06628},
year = {2025}
}