NequIP is a code for building E(3)-equivariant interatomic potentials

Allegro is an open-source code for building highly scalable and accurate equivariant deep learning interatomic potentials

The Open Forcefield Toolkit provides implementations of the SMIRNOFF format, parameterization engine, and other tools. Documentation available at http://open-forcefield-toolkit.readthedocs.io

A general cross-platform tool for preparing simulations of molecules and complex molecular assemblies

Pretrained universal neural network potential for charge-informed atomistic modeling https://chgnet.lbl.gov

[ICLR 2024] EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations

[ICLR 2023 Spotlight] Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs

Tinker: Software Tools for Molecular Design

List of molecules (small molecules, RNA, peptide, protein, enzymes, antibody, and PPIs) conformations and molecular dynamics (force fields) using generative artificial intelligence and deep learning

Build neural networks for machine learning force fields with JAX

train and use graph-based ML models of potential energy surfaces

UF3: a python library for generating ultra-fast interatomic potentials

MACE-MP models

Tinker9: Next Generation of Tinker with GPU Support

PyStokes: phoresis and Stokesian hydrodynamics in Python. github.com/rajeshrinet/pystokes

KIM-based Learning-Integrated Fitting Framework for interatomic potentials.

[TMLR 2024] Generalizing Denoising to Non-Equilibrium Structures Improves Equivariant Force Fields

Quantum to Molecular Mechanics (Q2MM)

Tracking citations of atomistic simulation engines

A flexible and performant framework for training machine learning potentials.