| Amp |
Atomistic machine-learning package for interatomic potential construction. |
Code/ML, Code/Lib |
| DeepModeling |
Open-source platform and community registry for AI-assisted molecular simulations. |
App, Code/ML |
| DeePMD-kit |
Deep learning package for many-body potential energy representation and molecular dynamics. |
Code/ML, Code/Lib |
| SchNetPack |
Deep neural networks for atomistic systems. |
Code/ML, Code/Lib |
| FieldSchNet |
Neural network predicting atomic forces and electric field properties. |
Code/ML, Code/Lib |
| MALA |
Materials learning algorithms framework combining machine learning and density functional theory. |
Code/ML, Code/Lib |
| RuNNer |
Neural network potential package for high-dimensional systems. |
Code/ML, Code/Lib |
| RuNNer ASE |
Atomic Simulation Environment interface for the RuNNer neural network potential package. |
Code/ML, Code/Lib |
| sGDML |
Symmetric gradient domain machine learning framework for molecular force fields. |
Code/ML, Code/Lib |
| SpookyNet |
Neural network potential incorporating electronic degrees of freedom and nonlocal effects. |
Code/ML, Code/Lib |
| MB-pol |
Data-driven many-body potential model for water simulations. |
Code/ML, Code/Lib |
| MBX |
Energy and force calculator for data-driven many-body simulations. |
Code/ML, Code/Lib |
| SchNarc |
Excited-state molecular dynamics combining SchNet neural networks and SHARC dynamics. |
Code/ML, Code/Lib |
| SPaiNN |
Spin-adapted neural network bridging machine learning and photoinduced dynamics. |
Code/ML, Code/Lib |
| MACE |
Machine learning interatomic potentials with higher-order equivariant message passing. |
Code/ML, Code/Lib |
| MACE-OFF |
Machine learning interatomic potentials optimized for organic molecules. |
Code/ML, Code/Lib |
| MACE-MP |
Foundation interatomic potentials trained on materials project datasets. |
Code/ML, Code/Lib |
| TACE |
Tensor atomic cluster expansion interatomic potentials. |
Code/ML, Code/Lib |
| Cartesian MACE |
Cartesian version of the MACE interatomic potential. |
Code/ML, Code/Lib |
| PE-MACE |
Potential-embedded MACE interatomic potential. |
Code/ML, Code/Lib |
| GRACE |
Machine learning interatomic potentials based on the GRACE tensor potential formulation. |
Code/ML, Code/Lib |
| MatGL |
Graph deep learning library for materials properties and interatomic potentials. |
Code/ML, Code/Lib |
| e3nn |
Modular PyTorch framework for Euclidean neural networks. |
Code/Lib, Code/ML |
| Equiformer |
Equivariant graph attention transformer for 3D atomistic graphs. |
Code/ML, Code/Lib |
| EquiformerV2 |
Equivariant transformer with improved scaling to higher-degree representations. |
Code/ML, Code/Lib |
| Nequix |
Foundation model for materials property prediction and phonon fine-tuning. |
Code/ML, Code/Lib |
| NequIP |
Equivariant interatomic potentials for molecular dynamics simulations. |
Code/ML, Code/Lib |
| Cartesian NequIP |
Cartesian version of the NequIP equivariant interatomic potential. |
Code/ML, Code/Lib |
| Allegro |
Scalable equivariant interatomic potentials for molecular dynamics. |
Code/ML, Code/Lib |
| Cartesian Allegro |
Cartesian version of the Allegro interatomic potential. |
Code/ML, Code/Lib |
| TorchMD-Net |
Neural network potentials for molecular systems and molecular dynamics. |
Code/ML, Code/Lib |
| CACE |
Cartesian atomic cluster expansion potentials for molecular simulations. |
Code/ML, Code/Lib |
| LES |
Latent Ewald summation for long-range interactions in neural network potentials. |
Code/ML, Code/Lib |
| SO3krates-LR |
Equivariant transformer potential for long-range interactions in molecular simulations. |
Code/ML, Code/Lib |
| MatterSim |
Deep learning atomistic model across elements, temperatures, and pressures. |
Code/ML, Code/Lib |
| SevenNet |
Graph neural network potential package supporting multi-GPU parallel molecular dynamics. |
Code/ML, Code/Lib |
| Nanocluster Transformers |
Transformer-based model for evaluating structural and energetic properties of transition-metal-doped nanoclusters. |
Code/ML, Code/Lib |
| PET-MAD |
Point edge transformer models for atomistic simulations. |
Code/ML, Code/Lib |
| UPET |
Point edge transformer architectures for universal machine learning potentials. |
Code/ML, Code/Lib |
| PRAPs |
Automated workflow for training moment tensor potentials and interfacing with MLIP software. |
Code/ML, Code/WF |
| n2p2 |
Neural network potential package for atomistic molecular dynamics simulations. |
Code/ML, Code/Lib |
| NPS |
Neural phase simulator for thermodynamic property modeling. |
Code/ML, Code/Lib |
| GEARS-H |
Hamiltonian converter for GPAW linear combination of atomic orbitals. |
Code/ML, Code/Lib |
| Polymers |
Multimodal machine learning for predicting properties of low-modulus polymers under data scarcity. |
Code/ML, Code/Lib |
| ELECTRA_2025 |
Charge density prediction using equivariant neural networks. |
Code/ML, Code/Lib |
| DistMLIP |
Distributed inference library for fast, large-scale atomistic simulation. |
Code/ML, Code/Lib |
| DeepH-pack |
Deep neural networks for predicting density functional theory Hamiltonians. |
Code/ML, Code/Lib |
| HamGNN |
Equivariant graph neural network for predicting electronic Hamiltonian matrices. |
Code/ML, Code/Lib |
| Zatom |
General-purpose model for three-dimensional molecules and materials. |
Code/ML, Code/Lib |
| PAIPAI |
Interstitial prediction in alloys using machine learning. |
Code/ML, Code/Lib |
| MIST |
Molecular insight SMILES transformer foundation models for chemical property prediction. |
Code/ML, Code/Lib |
| CLOUD |
Empirical scaling laws and electrochemical property prediction models for battery systems. |
Code/ML, Code/Lib |
| Charge3Net |
Equivariant neural networks for charge density prediction in materials. |
Code/ML, Code/Lib |
| mPFDNN |
Material-property-field-based deep neural network in Hopfield framework. |
Code/ML, Code/Lib |
| Uni2D |
Universal machine learning interatomic potential for two-dimensional materials. |
Code/ML, Code/Lib |
| MolCryst |
Foundation models and datasets optimized for molecular crystals. |
Code/ML, Code/Lib |
| AIMNet2 |
Fast machine-learned interatomic potentials for molecular dynamics simulations. |
Code/ML, Code/Lib |
| VibroML |
Machine learning vibrational analysis and stability estimation. |
Code/ML, Code/Lib |
| PlatonicRep |
Alignment of latent representations from foundation machine learning interatomic potentials. |
Code/ML, Code/Lib |
| CGAANet_proto |
Coarse-grained graph architectures for all-atom force predictions. |
Code/ML, Code/Lib |
| Multitask-finetuning |
Multi-task fine-tuning of pretrained atomistic models for out-of-distribution generalization. |
Code/ML, Code/Lib |
| QT-Net |
JAX-based non-equivariant graph neural network for atomic-scale molecular property prediction. |
Code/ML, Code/Lib |
| TEGNet |
Neural network predicting thermoelectric generator performance and design optimization. |
Code/ML, Code/Lib |
| TriForces |
Augmentation framework for atomistic graph neural networks to achieve transferable representations. |
Code/ML, Code/Lib |
| spin-mlip |
Workflow for active learning of spin-dependent machine learning interatomic potentials. |
Code/ML, Code/Lib |
| PolySet |
Stochastic ensemble representation framework for polymer chains. |
Code/Lib, Code/ML |
| PolyGraphPy |
Unified Python framework for atomistic simulation and machine learning-driven polymer design. |
Code/ML, Code/Lib |
| TECSA-GNN |
Multiscale graph neural network for predicting thermoelectric transport properties in inorganic crystals. |
Code/ML, Code/Lib |
| GCNN |
Graph convolutional neural network for atomistic potential energy prediction. |
Code/ML, Code/Lib |
| StructLLM |
Explainable synthesizability prediction of inorganic crystal polymorphs using Large Language Models. |
Code/ML, Code/Lib |
| MOFSimplify |
Predicting the stability of metal-organic frameworks. |
App, Code/ML |
| CGCNN |
Crystal Graph Convolutional Neural Networks for predicting material properties from crystal structure. |
Code/ML, Code/Lib |
| AutoPot |
Automated and massively parallelized construction of machine-learning interatomic potentials. |
Code/ML, Code/WF |
| BEE-NET |
Bootstrapped Ensemble of Equivariant Graph Neural Networks (BEE-NET) for predicting the Eliashberg spectral function and superconducting critical temperature. |
Code/ML, Code/Lib |
| materials_discovery |
Graph networks for active learning and materials discovery of stable inorganic crystals. |
Code/ML, Code/Lib |
| novel-materials-screening |
Materials Project screening tool for discovering novel materials to be synthesized by the autonomous laboratory (A-Lab). |
Code/ML, Code/Lib |
| s4 |
Solid-state synthesis science analyzer utilizing thermodynamics, features, and machine learning. |
Code/ML, Code/Lib |
| MatDeepLearn |
Graph neural network framework for predicting materials properties. |
Code/ML, Code/Lib |
| periodicwave |
Neural network variational Monte Carlo for solids. |
Code/ML, Code/Sim |
| XRD-AutoAnalyzer |
Automated phase identification of X-ray diffraction patterns using deep learning. |
Code/ML, Code/Lib |
| MoMa |
Modular learning framework for materials property prediction. |
Code/ML, Code/Lib |
| w2v_transformers_materials_embeddings |
Composition embeddings from Word2Vec to transformers for filtering combinatorial electrocatalysts. |
Code/ML, Code/Lib |
| volume_prediction_workflow |
High-throughput screening workflow to predict volume changes upon ion intercalation in battery materials. |
Code/ML, Code/WF |
| polymon |
Unified machine learning framework for polymer property prediction. |
Code/ML, Code/Lib |
| PolyGraphMT |
Multi-task and multi-fidelity machine learning framework for polymer property prediction. |
Code/ML, Code/Lib |
| Zeolite Classifier |
Web interface for classifying zeolite structures. |
App, Code/ML |
| Battery-Capacity-Prediction-Using-Regression |
Lithium-ion battery capacity prediction using machine learning regression models. |
Code/ML, Code/Lib |
| SR-CGCNN |
Shared recurrent convolution in crystal graph neural networks for materials property prediction. |
Code/ML, Code/Lib |
| HIP |
Deep learning framework for predicting molecular Hessians from interatomic potentials. |
Code/ML, Code/Lib |
| HIP-MACE |
Integration of MACE equivariant interatomic potentials with the HIP Hessian prediction framework. |
Code/ML, Code/Lib |
| UMA |
Interactive molecular simulation playground for the Universal Model for Atoms (UMA). |
App, Edu |
| CarNet |
Equivariant atomistic machine learning using Cartesian natural tensor networks. |
Code/ML, Code/Lib |