QuantumLab-ZY/HamGNN

DeePTB: A deep learning package for tight-binding Hamiltonian with ab initio accuracy.

Graph deep learning library for materials

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

An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations

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

SchNetPack - Deep Neural Networks for Atomistic Systems

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

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

cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural networks. Also includes kernels for accelerated structure prediction.