Models

SpookyNet is a Python research package described as learning force fields with electronic degrees of freedom and nonlocal effects, with a cited publication and MIT-licensed repository code.

Molecular Property Prediction · Quantum Chemistry

PhysNet is a TensorFlow implementation of a neural network for predicting molecular energies, forces, dipole moments and partial charges, with a configurable training workflow and an example dataset.

Molecular Property Prediction · Quantum Chemistry

TorchMD-Net implements trainable neural network potentials as PyTorch modules, with TensorNet architectures, custom molecular datasets, pretrained checkpoint loading and molecular dynamics integrations.

Molecular Property Prediction · Computational Chemistry

DeePMD-kit is a toolkit for training and fine-tuning Deep Potential interatomic models from quantum-mechanical reference data, then exporting them for inference and molecular dynamics.

Computational Chemistry · Materials Discovery

TorchANI is a PyTorch library for developing, training and using ANI-style neural network interatomic potentials, with optional C++ and CUDA extensions for descriptors and inference.

Molecular Property Prediction · Quantum Chemistry

Allegro implements an E(3)-equivariant interatomic potential as a NequIP extension, with documented GPU acceleration options and a separate plugin for LAMMPS simulations.

Computational Chemistry · Materials Discovery

NequIP is an open-source framework for building E(3)-equivariant interatomic potentials, with training, pre-trained foundation potentials, and integration with ASE and LAMMPS.

Computational Chemistry · Materials Discovery

MACE provides higher-order equivariant machine-learning interatomic potentials, with workflows for training on atomistic data, evaluating configurations, and using or fine-tuning pretrained models.

Computational Chemistry · Materials Discovery

SchNetPack is a Python toolkit for building and training atomistic neural networks, with SchNet and PaiNN representations, quantum-chemical property outputs, and molecular dynamics components.

Molecular Property Prediction · Quantum Chemistry

AiZynthFinder is a Python retrosynthetic planning toolkit that uses neural-network-guided search to propose routes from target molecules to precursors in a configured stock.

Retrosynthesis

Chemprop is a PyTorch-based toolkit for training and evaluating message-passing neural networks for molecular property prediction, with CLI workflows, Python modules and task-specific notebooks.

Molecular Property Prediction