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

Allegro is a code for building highly scalable E(3)-equivariant 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

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

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

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

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

[TMLR 2024 J2C Certification] Generalizing Denoising to Non-Equilibrium Structures Improves Equivariant Force Fields

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

A flexible and performant framework for training machine learning potentials.

Quantum to Molecular Mechanics (Q2MM)

JAX implementation of the NequIP neural network interatomic potential

Cartesian Atomic Moment Potentials -- CAMP

Optimization tool for calibrating coarse-grained force fields of lipids, relying on the simultaneous usage of reference AA trajectories (bottom-up) and experimental data (top-down)

Cartesian Atomic Moment Potentials -- CAMP

Data and scripts relevant to an evaluation of force field methods for conformer scoring

(ARCHIVED) Official PyTorch implementation of "Comprehensive Molecular Representation from Equivariant Transformer" paper https://arxiv.org/abs/2308.10752. Made in Cardiff University.