kairosmaterial/prophet

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

MatterSim: A deep learning atomistic model across elements, temperatures and pressures.

ORB forcefield models from Orbital Materials

Foundation Model for Materials - FM4M

Universal machine-learning models for advanced atomistic simulations

An evaluation framework for machine learning models simulating high-throughput materials discovery.

Graph deep learning library for materials

MACE foundation models (MP, OMAT, mh-1)

MACE - Fast and accurate machine learning interatomic potentials with higher order equivariant message passing.