Best of Atomistic Machine Learning βοΈπ§¬π
π A ranked list of awesome atomistic machine learning (AML) projects. Updated regularly.
This curated list contains 510 awesome open-source projects with a total of 240K stars grouped into 23 categories. All projects are ranked by a project-quality score, which is calculated based on various metrics automatically collected from GitHub and different package managers. If you like to add or update projects, feel free to open an issue, submit a pull request, or directly edit the projects.yaml.
The current focus of this list is more on simulation data rather than experimental data, and more on materials rather than drug design. Nevertheless, contributions from other fields are warmly welcome!
How to cite. See the button "Cite this repository" on the right side-bar.
π§ββοΈ Discover other best-of lists or create your own.
- π₯π₯π₯ Combined project-quality score
- βοΈ Star count from GitHub
- π£ New project (less than 6 months old)
- π€ Inactive project (6 months no activity)
- π Dead project (12 months no activity)
- ππ Project is trending up or down
- β Project was recently added
- π¨βπ» Contributors count from GitHub
- π Fork count from GitHub
- π Issue count from GitHub
- β±οΈ Last update timestamp on package manager
- π₯ Download count from package manager
- π¦ Number of dependent projects
Projects that focus on enabling active learning, iterative learning schemes for atomistic ML.
DP-GEN (π₯23 Β· β 380) - The deep potential generator to generate a deep-learning based model of interatomic potential energy and force field. LGPL-3.0 ML-IAP MD workflows
- [GitHub](https://github.com/deepmodeling/dpgen) (π¨βπ» 73 Β· π 180 Β· π₯ 2.1K Β· π¦ 8 Β· π 340 - 13% open Β· β±οΈ 06.04.2026):
git clone https://github.com/deepmodeling/dpgen
conda install -c deepmodeling dpgen
FLARE (π₯18 Β· β 350) - An open-source Python package for creating fast and accurate interatomic potentials. MIT C++ ML-IAP
- [GitHub](https://github.com/mir-group/flare) (π¨βπ» 44 Β· π 78 Β· π₯ 9 Β· π¦ 12 Β· π 220 - 14% open Β· β±οΈ 30.01.2026):
git clone https://github.com/mir-group/flare
Bgolearn (π₯17 Β· β 130) - [arXiv:2601.06820] Offical implement of Bgolearn. MIT materials-discovery probabilistic
- [GitHub](https://github.com/Bin-Cao/Bgolearn) (π¨βπ» 4 Β· π 18 Β· π₯ 71 Β· β±οΈ 05.04.2026):
git clone https://github.com/Bin-Cao/Bgolearn
IPSuite (π₯15 Β· β 24) - A Python toolkit for FAIR development and deployment of machine-learned interatomic potentials. EPL-2.0 ML-IAP MD workflows HTC FAIR
- [GitHub](https://github.com/zincware/IPSuite) (π¨βπ» 9 Β· π 13 Β· π¦ 14 Β· π 180 - 49% open Β· β±οΈ 13.03.2026):
git clone https://github.com/zincware/IPSuite
DP-GEN2 (π₯14 Β· β 40) - 2nd generation of the Deep Potential GENerator. LGPL-3.0 ML-IAP MD workflows
- [GitHub](https://github.com/deepmodeling/dpgen2) (π¨βπ» 19 Β· π 36 Β· π¦ 6 Β· π 46 - 39% open Β· β±οΈ 04.04.2026):
git clone https://github.com/deepmodeling/dpgen2
Show 4 hidden projects...
- flare++ (π₯12 Β· β 38 Β· π) - A many-body extension of the FLARE code. MIT C++ ML-IAP
- Finetuna (π₯10 Β· β 67 Β· π) - Active Learning for Machine Learning Potentials. MIT
- ACEHAL (π₯5 Β· β 15 Β· π) - Hyperactive Learning (HAL) Python interface for building Atomic Cluster Expansion potentials. Unlicensed Julia
- ALEBREW (π₯4 Β· β 21 Β· π) - Official repository for the paper Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic.. Custom ML-IAP MD
Projects that collect atomistic ML resources or foster communication within community.
π ACE / GRACE support - Support forum for the Atomic Cluster Expansion (ACE) and extensions.
π AI for Science Map - Interactive mindmap of the AI4Science research field, including atomistic machine learning, including papers,..
π ASE ecosystem - This is a list of software packages related to ASE or using ASE. md, ml-iap
π Atomic Cluster Expansion - Atomic Cluster Expansion (ACE) community homepage.
π CrystaLLM - Generate a crystal structure from a composition. language-models generative pretrained transformer
π GAP-ML.org community homepage ML-IAP
π matsci.org - A community forum for the discussion of anything materials science, with a focus on computational materials science..
π Matter Modeling Stack Exchange - Machine Learning - Forum StackExchange, site Matter Modeling, ML-tagged questions.
Best-of Machine Learning with Python (π₯21 Β· β 23K) - A ranked list of awesome machine learning Python libraries. Updated weekly. CC-BY-4.0 general-ml Python
- [GitHub](https://github.com/lukasmasuch/best-of-ml-python) (π¨βπ» 57 Β· π 3K Β· π 63 - 46% open Β· β±οΈ 22.03.2026):
git clone https://github.com/ml-tooling/best-of-ml-python
MatBench Discovery (π₯21 Β· β 220) - An evaluation framework for machine learning models simulating high-throughput materials discovery. MIT datasets benchmarking model-repository
- [GitHub](https://github.com/janosh/matbench-discovery) (π¨βπ» 29 Β· π 54 Β· π¦ 7 Β· π 72 - 4% open Β· β±οΈ 08.04.2026):
git clone https://github.com/janosh/matbench-discovery
pip install matbench-discovery
OpenML (π₯19 Β· β 730) - Open Machine Learning. BSD-3 datasets
- [GitHub](https://github.com/openml/OpenML) (π¨βπ» 35 Β· π 120 Β· π 960 - 40% open Β· β±οΈ 23.01.2026):
git clone https://github.com/openml/OpenML
Garden (π₯18 Β· β 39) - FAIR AI/ML Model Publishing Framework. MIT model-repository
- [GitHub](https://github.com/Garden-AI/garden) (π¨βπ» 14 Β· π 4 Β· π¦ 6 Β· π 380 - 3% open Β· β±οΈ 18.03.2026):
git clone https://github.com/Garden-AI/garden
Graph-based Deep Learning Literature (π₯17 Β· β 5K) - links to conference publications in graph-based deep learning. MIT general-ml rep-learn
- [GitHub](https://github.com/naganandy/graph-based-deep-learning-literature) (π¨βπ» 12 Β· π 770 Β· β±οΈ 07.02.2026):
git clone https://github.com/naganandy/graph-based-deep-learning-literature
AI for Science Resources (π₯14 Β· β 750) - List of resources for AI4Science research, including learning resources. GPL-3.0 license
- [GitHub](https://github.com/divelab/AIRS) (π¨βπ» 36 Β· π 89 Β· π 32 - 18% open Β· β±οΈ 30.03.2026):
git clone https://github.com/divelab/AIRS
GT4SD - Generative Toolkit for Scientific Discovery (π₯14 Β· β 370 Β· π€) - Gradio apps of generative models in GT4SD. MIT generative pretrained drug-discovery model-repository
- [GitHub](https://github.com/GT4SD/gt4sd-core) (π¨βπ» 20 Β· π 79 Β· π 120 - 11% open Β· β±οΈ 18.09.2025):
git clone https://github.com/GT4SD/gt4sd-core
Awesome Materials Informatics (π₯12 Β· β 500 Β· π€) - Curated list of known efforts in materials informatics, i.e. in modern materials science. Custom
- [GitHub](https://github.com/tilde-lab/awesome-materials-informatics) (π¨βπ» 21 Β· π 100 Β· β±οΈ 19.06.2025):
git clone https://github.com/tilde-lab/awesome-materials-informatics
Neural-Network-Models-for-Chemistry (π₯12 Β· β 190) - A collection of Neural Network Models for chemistry. MIT rep-learn
- [GitHub](https://github.com/Eipgen/Neural-Network-Models-for-Chemistry) (π¨βπ» 3 Β· π 24 Β· π 2 - 50% open Β· β±οΈ 09.04.2026):
git clone https://github.com/Eipgen/Neural-Network-Models-for-Chemistry
Awesome-Scientific-Language-Models (π₯10 Β· β 650 Β· π€) - A Comprehensive Survey of Scientific Large Language Models and Their Applications in Scientific Discovery (EMNLP24). MIT language-models general-ml pretrained multimodal
- [GitHub](https://github.com/yuzhimanhua/Awesome-Scientific-Language-Models) (π¨βπ» 9 Β· π 37 Β· β±οΈ 21.06.2025):
git clone https://github.com/yuzhimanhua/Awesome-Scientific-Language-Models
Awesome Materials & Chemistry Datasets (π₯10 Β· β 300) - A curated list of the most useful datasets in materials science and chemistry for training machine learning and AI.. MIT datasets experimental-data literature-data proprietary
- [GitHub](https://github.com/blaiszik/awesome-matchem-datasets) (π¨βπ» 9 Β· π 34 Β· π 15 - 26% open Β· β±οΈ 22.03.2026):
git clone https://github.com/blaiszik/awesome-matchem-datasets
optimade.science (π₯9 Β· β 10) - A sky-scanner Optimade browser-only GUI. MIT datasets
- [GitHub](https://github.com/tilde-lab/optimade.science) (π¨βπ» 8 Β· π 4 Β· π 26 - 26% open Β· β±οΈ 04.11.2025):
git clone https://github.com/tilde-lab/optimade.science
Awesome Neural Geometry (π₯8 Β· β 1.1K) - A curated collection of resources and research related to the geometry of representations in the brain, deep networks,.. Unlicensed educational rep-learn
- [GitHub](https://github.com/neurreps/awesome-neural-geometry) (π¨βπ» 16 Β· π 70 Β· β±οΈ 24.02.2026):
git clone https://github.com/neurreps/awesome-neural-geometry
The Collection of Database and Dataset Resources in Materials Science (π₯8 Β· β 420) - A list of databases, datasets and books/handbooks where you can find materials properties for machine learning.. Unlicensed datasets
- [GitHub](https://github.com/sedaoturak/data-resources-for-materials-science) (π¨βπ» 2 Β· π 59 Β· β±οΈ 06.03.2026):
git clone https://github.com/sedaoturak/data-resources-for-materials-science
AI for Science paper collection (π₯8 Β· β 170 Β· π€) - List the AI for Science papers accepted by top conferences. Apache-2
- [GitHub](https://github.com/AI4QC/AI_for_Science_paper_collection) (π¨βπ» 7 Β· π 17 Β· β±οΈ 24.09.2025):
git clone https://github.com/sherrylixuecheng/AI_for_Science_paper_collection
Awesome Neural SBI (π₯8 Β· β 150) - Community-sourced list of papers and resources on neural simulation-based inference. MIT active-learning
- [GitHub](https://github.com/smsharma/awesome-neural-sbi) (π¨βπ» 7 Β· π 8 Β· π 2 - 50% open Β· β±οΈ 28.01.2026):
git clone https://github.com/smsharma/awesome-neural-sbi
Charting ML Publications in Science (π₯8 Β· β 47) - Literature analysis of ML applications in materials science, chemistry, physics. MIT literature-data general-ml
- [GitHub](https://github.com/blaiszik/ml_publication_charts) (π¨βπ» 2 Β· β±οΈ 10.03.2026):
git clone https://github.com/blaiszik/ml_publication_charts
DeepModeling Projects (π₯8 Β· β 8) - DeepModeling projects. CC-BY-4.0
- [GitHub](https://github.com/deepmodeling/deepmodeling-projects) (π¨βπ» 4 Β· π 2 Β· β±οΈ 02.04.2026):
git clone https://github.com/deepmodeling/deepmodeling-projects
Awesome-Crystal-GNNs (π₯7 Β· β 120) - This repository contains a collection of resources and papers on GNN Models on Crystal Solid State Materials. MIT
- [GitHub](https://github.com/kdmsit/Awesome-Crystal-GNNs) (π¨βπ» 2 Β· π 13 Β· β±οΈ 09.03.2026):
git clone https://github.com/kdmsit/Awesome-Crystal-GNNs
Show 11 hidden projects...
- MatBench (π₯18 Β· β 190 Β· π) - Matbench: Benchmarks for materials science property prediction. MIT datasets benchmarking model-repository
- GNoME Explorer (π₯12 Β· β 1.2K Β· π) - Graph Networks for Materials Exploration Database. Apache-2 datasets materials-discovery
- MoLFormers UI (π₯9 Β· β 390 Β· π) - A family of foundation models trained on chemicals. Apache-2 transformer language-models pretrained drug-discovery
- Awesome-Graph-Generation (π₯8 Β· β 360 Β· π) - A curated list of up-to-date graph generation papers and resources. Unlicensed rep-learn
- A Highly Opinionated List of Open-Source Materials Informatics Resources (π₯8 Β· β 150 Β· π) - A Highly Opinionated List of Open Source Materials Informatics Resources. MIT
- MADICES Awesome Interoperability (π₯6 Β· β 1) - Linked data interoperability resources of the Machine-actionable data interoperability for the chemical sciences.. MIT datasets
- LAM Crystal Philately competition 2024 (π₯5 Β· β 22 Β· π) - OpenLAM Challenge crystal structure prediction https://arxiv.org/abs/2501.16358. LGPL-2.1 single-paper datasets structure-prediction materials-discovery ML-IAP UIP
- Geometric-GNNs (π₯4 Β· β 120 Β· π) - List of Geometric GNNs for 3D atomic systems. Unlicensed datasets educational rep-learn
- Does this material exist? (π₯3 Β· β 18 Β· π) - Vote on whether you think predicted crystal structures could be synthesised. MIT for-fun materials-discovery
- GitHub topic materials-informatics (π₯1) - GitHub topic materials-informatics. Unlicensed
- MateriApps (π₯1) - A Portal Site of Materials Science Simulation. Unlicensed
Datasets, databases and trained models for atomistic ML.
π Alexandria Materials Database - A database of millions of theoretical crystal structures (3D, 2D and 1D) discovered by machine learning accelerated..
π Catalysis Hub - A web-platform for sharing data and software for computational catalysis research!.
π Citrination Datasets - AI-Powered Materials Data Platform. Open Citrination has been decommissioned.
π crystals.ai - Curated datasets for reproducible AI in materials science.
π DeepChem Models - DeepChem models on HuggingFace. model-repository pretrained language-models
π Graphs of Materials Project 20190401 - The dataset used to train the MEGNet interatomic potential. ML-IAP
π HME21 Dataset - High-temperature multi-element 2021 dataset for the PreFerred Potential (PFP).. UIP
π JARVIS-Leaderboard ( β 73 Β· π€) - A large scale benchmark of materials design methods: https://www.nature.com/articles/s41524-024-01259-w. model-repository benchmarking community-resource educational
π Materials Project - Charge Densities - Materials Project has started offering charge density information available for download via their public API.
π Materials Project Trajectory (MPtrj) Dataset - The dataset used to train the CHGNet universal potential. UIP
π matterverse.ai - Database of yet-to-be-sythesized materials predicted using state-of-the-art machine learning algorithms.
π MPF.2021.2.8 - The dataset used to train the M3GNet universal potential. UIP
π NRELMatDB - Computational materials database with the specific focus on materials for renewable energy applications including, but..
π QM9 Charge Densities and Energies - QM9 molecules calculated with VASP using Atomic Simulation Environment. ML-DFT
π QM40 Dataset - A More Realistic QM Dataset for Machine Learning in Molecular Science https://doi.org/10.1038/s41597-024-04206-y. drug-discovery
π QMugs dataset - Quantum Mechanical Properties of Drug-like Molecules https://doi.org/10.1038/s41597-022-01390-7. drug-discovery
π Quantum-Machine.org Datasets - Collection of datasets, including QM7, QM9, etc. MD, DFT. Small organic molecules, mostly.
π sGDML Datasets - MD17, MD22, DFT datasets.
π MoleculeNet - A Benchmark for Molecular Machine Learning. benchmarking
π ZINC15 - A free database of commercially-available compounds for virtual screening. ZINC contains over 230 million purchasable.. graph biomolecules
π ZINC20 - A free database of commercially-available compounds for virtual screening. ZINC contains over 230 million purchasable.. graph biomolecules
FAIR Chemistry datasets (π₯32 Β· β 2K Β· π) - Datasets OC20, OC22, etc. Formerly known as Open Catalyst Project. MIT catalysis
- [GitHub](https://github.com/facebookresearch/fairchem) (π¨βπ» 69 Β· π 450 Β· π 570 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/FAIR-Chem/fairchem
pip install fairchem-core
Meta Open Materials 2024 (OMat24) Dataset (π₯31 Β· β 2K Β· π) - Contains over 100 million Density Functional Theory calculations focused on structural and compositional diversity. CC-BY-4.0
- [GitHub](https://github.com/facebookresearch/fairchem) (π¨βπ» 69 Β· π 450 Β· π 570 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/FAIR-Chem/fairchem
pip install fairchem-core
MPContribs (π₯25 Β· β 39) - Platform for materials scientists to contribute and disseminate their materials data through Materials Project. MIT
- [GitHub](https://github.com/materialsproject/MPContribs) (π¨βπ» 29 Β· π 27 Β· π¦ 58 Β· π 120 - 30% open Β· β±οΈ 26.02.2026):
git clone https://github.com/materialsproject/MPContribs
pip install mpcontribs-client
OPTIMADE Python tools (π₯23 Β· β 89) - Tools for implementing and consuming OPTIMADE APIs in Python. MIT
- [GitHub](https://github.com/Materials-Consortia/optimade-python-tools) (π¨βπ» 34 Β· π 50 Β· π 500 - 22% open Β· β±οΈ 02.03.2026):
git clone https://github.com/Materials-Consortia/optimade-python-tools
conda install -c conda-forge optimade
load-atoms (π₯19 Β· β 49) - download and manipulate atomistic datasets. MIT data-structures
- [GitHub](https://github.com/jla-gardner/load-atoms) (π¨βπ» 5 Β· π 5 Β· π¦ 8 Β· π 35 - 14% open Β· β±οΈ 25.11.2025):
git clone https://github.com/jla-gardner/load-atoms
Open Databases Integration for Materials Design (OPTIMADE) (π₯17 Β· β 100) - Specification of a common REST API for access to materials databases. CC-BY-4.0
- [GitHub](https://github.com/Materials-Consortia/OPTIMADE) (π¨βπ» 24 Β· π 37 Β· π 260 - 30% open Β· β±οΈ 18.12.2025):
git clone https://github.com/Materials-Consortia/OPTIMADE
QH9 (π₯14 Β· β 750) - A Quantum Hamiltonian Prediction Benchmark. CC-BY-NC-SA-4.0 ML-DFT
- [GitHub](https://github.com/divelab/AIRS) (π¨βπ» 36 Β· π 89 Β· π 32 - 18% open Β· β±οΈ 30.03.2026):
git clone https://github.com/divelab/AIRS
OpenQDC (π₯14 Β· β 60 Β· π€) - Repository of Quantum Datasets Publicly Available. CC-BY-4.0
- [GitHub](https://github.com/valence-labs/OpenQDC) (π¨βπ» 10 Β· π 6 Β· π¦ 4 Β· π 50 - 18% open Β· β±οΈ 19.06.2025):
git clone https://github.com/valence-labs/openQDC
conda install -c conda-forge openqdc
OpenKIM (π₯13 Β· β 37) - The Open Knowledgebase of Interatomic Models (OpenKIM) aims to be an online resource for standardized testing, long-.. LGPL-2.1 model-repository knowledge-base pretrained
- [GitHub](https://github.com/openkim/kim-api) (π¨βπ» 27 Β· π 18 Β· π 37 - 40% open Β· β±οΈ 06.03.2026):
git clone https://github.com/openkim/kim-api
nablaDFT (π₯12 Β· β 230) - nablaDFT: Large-Scale Conformational Energy and Hamiltonian Prediction benchmark and dataset. MIT ML-DFT ML-WFT drug-discovery ML-IAP benchmarking
- [GitHub](https://github.com/AIRI-Institute/nablaDFT) (π¨βπ» 9 Β· π 25 Β· π 28 - 10% open Β· β±οΈ 31.12.2025):
git clone https://github.com/AIRI-Institute/nablaDFT
MatPES (π₯12 Β· β 53) - A foundational potential energy dataset for materials. BSD-3 UIP ML-IAP
- [GitHub](https://github.com/materialyzeai/matpes) (π¨βπ» 3 Β· π 5 Β· π 9 - 22% open Β· β±οΈ 02.03.2026):
git clone https://github.com/materialsvirtuallab/matpes
SPICE (π₯11 Β· β 190) - A collection of QM data for training potential functions. MIT ML-IAP MD
- [GitHub](https://github.com/openmm/spice-dataset) (π¨βπ» 1 Β· π 11 Β· π₯ 340 Β· π 76 - 27% open Β· β±οΈ 25.02.2026):
git clone https://github.com/openmm/spice-dataset
MPDS API (π₯11 Β· β 26) - Tutorials, notebooks, issue tracker, and website on the MPDS API: the data retrieval interface for the Materials.. CC-BY-4.0 phase-transition
- [GitHub](https://github.com/mpds-io/mpds-api) (π¨βπ» 5 Β· π 5 Β· π 36 - 22% open Β· β±οΈ 24.01.2026):
git clone https://github.com/mpds-io/mpds-api
OBELiX (π₯9 Β· β 53) - A Curated Dataset of Crystal Structures and Experimentally Measured Ionic Conductivities for Lithium Solid-State.. CC-BY-4.0 experimental-data transport-phenomena
- [GitHub](https://github.com/NRC-Mila/OBELiX) (π¨βπ» 6 Β· π 9 Β· π 2 - 50% open Β· β±οΈ 27.11.2025):
git clone https://github.com/NRC-Mila/OBELiX
AIS Square (π₯9 Β· β 15) - A collaborative and open-source platform for sharing AI for Science datasets, models, and workflows. Home of the.. LGPL-3.0 community-resource model-repository
- [GitHub](https://github.com/deepmodeling/AIS-Square) (π¨βπ» 8 Β· π 8 Β· π 6 - 83% open Β· β±οΈ 03.04.2026):
git clone https://github.com/deepmodeling/AIS-Square
polyVERSE (π₯7 Β· β 32) - polyVERSE is a comprehensive repository of informatics-ready datasets curated by the Ramprasad Group. Custom soft-matter
- [GitHub](https://github.com/Ramprasad-Group/polyVERSE) (π¨βπ» 9 Β· π 6 Β· β±οΈ 21.01.2026):
git clone https://github.com/Ramprasad-Group/polyVERSE
Visual Graph Datasets (π₯6 Β· β 5) - Datasets for the training of graph neural networks (GNNs) and subsequent visualization of attributional explanations.. MIT XAI rep-learn
- [GitHub](https://github.com/aimat-lab/visual_graph_datasets) (π¨βπ» 2 Β· π 3 Β· β±οΈ 24.03.2026):
git clone https://github.com/aimat-lab/visual_graph_datasets
The Perovskite Database Project (π₯5 Β· β 70) - Perovskite Database Project aims at making all perovskite device data, both past and future, available in a form.. Unlicensed community-resource
- [GitHub](https://github.com/Jesperkemist/perovskitedatabase) (π¨βπ» 2 Β· π 26 Β· β±οΈ 20.03.2026):
git clone https://github.com/Jesperkemist/perovskitedatabase
Show 16 hidden projects...
- ATOM3D (π₯19 Β· β 320 Β· π) - ATOM3D: tasks on molecules in three dimensions. MIT biomolecules benchmarking
- Materials Data Facility (MDF) (π₯10 Β· β 10 Β· π) - A simple way to publish, discover, and access materials datasets. Publication of very large datasets supported (e.g.,.. Apache-2
- MoleculeNet Leaderboard (π₯9 Β· β 100 Β· π) - MIT benchmarking
- 2DMD dataset (π₯9 Β· β 8 Β· π) - Code for Kazeev, N., Al-Maeeni, A.R., Romanov, I. et al. Sparse representation for machine learning the properties of.. Apache-2 material-defect
- ANI-1 Dataset (π₯8 Β· β 100 Β· π) - A data set of 20 million calculated off-equilibrium conformations for organic molecules. MIT
- GEOM (π₯7 Β· β 240 Β· π) - GEOM: Energy-annotated molecular conformations. Unlicensed drug-discovery
- ANI-1x Datasets (π₯6 Β· β 67 Β· π) - The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for organic molecules. MIT
- COMP6 Benchmark dataset (π₯6 Β· β 40 Β· π) - COMP6 Benchmark dataset for ML potentials. MIT
- SciGlass (π₯6 Β· β 15 Β· π) - The database contains a vast set of data on the properties of glass materials. MIT
- GDB-9-Ex9 and ORNL_AISD-Ex (π₯5 Β· β 10 Β· π) - Distributed computing workflow for generation and analysis of large scale molecular datasets obtained running multi-.. Unlicensed
- OPTIMADE providers dashboard (π₯5 Β· β 2 Β· π€) - A dashboard of known providers. Unlicensed
- 3DSC Database (π₯4 Β· β 26 Β· π) - Repo for the paper publishing the superconductor database with 3D crystal structures. Custom superconductors materials-discovery
- paper-data-redundancy (π₯4 Β· β 11 Β· π) - Repo for the paper Exploiting redundancy in large materials datasets for efficient machine learning with less data. BSD-3 small-data single-paper
- linear-regression-benchmarks (π₯4 Β· β 1 Β· π) - Data sets used for linear regression benchmarks. MIT benchmarking single-paper
- nep-data (π₯3 Β· β 21 Β· π) - Data related to the NEP machine-learned potential of GPUMD. Unlicensed ML-IAP MD transport-phenomena
- tmQM_wB97MV Dataset (π₯1 Β· β 9 Β· π) - Code for Applying Large Graph Neural Networks to Predict Transition Metal Complex Energies Using the tmQM_wB97MV.. Unlicensed catalysis rep-learn
Projects that focus on providing data structures used in atomistic machine learning.
dpdata (π₯28 Β· β 240) - A Python package for manipulating atomistic data of software in computational science. LGPL-3.0
- [GitHub](https://github.com/deepmodeling/dpdata) (π¨βπ» 67 Β· π 160 Β· π¦ 160 Β· π 160 - 31% open Β· β±οΈ 06.04.2026):
git clone https://github.com/deepmodeling/dpdata
conda install -c deepmodeling dpdata
Metatensor (π₯22 Β· β 98) - Self-describing sparse tensor data format for atomistic machine learning and beyond. BSD-3 ML-IAP MD Rust C-lang C++ Python
- [GitHub](https://github.com/metatensor/metatensor) (π¨βπ» 36 Β· π 25 Β· π₯ 54K Β· π¦ 14 Β· π 270 - 27% open Β· β±οΈ 01.04.2026):
git clone https://github.com/metatensor/metatensor
mp-pyrho (π₯18 Β· β 42) - Tools for re-griding volumetric quantum chemistry data for machine-learning purposes. Custom ML-DFT
- [GitHub](https://github.com/materialsproject/pyrho) (π¨βπ» 10 Β· π 10 Β· π¦ 35 Β· π 10 - 70% open Β· β±οΈ 13.10.2025):
git clone https://github.com/materialsproject/pyrho
dlpack (π₯16 Β· β 1.2K) - common in-memory tensor structure. Apache-2 C++
- [GitHub](https://github.com/dmlc/dlpack) (π¨βπ» 33 Β· π 160 Β· π 85 - 28% open Β· β±οΈ 24.01.2026):
git clone https://github.com/dmlc/dlpack
Projects and models that focus on quantities of DFT, such as density functional approximations (ML-DFA), the charge density, density of states, the Hamiltonian, etc.
π IKS-PIML - Code and generated data for the paper Inverting the Kohn-Sham equations with physics-informed machine learning.. neural-operator pinn datasets single-paper
π M-OFDFT - Overcoming the Barrier of Orbital-Free Density Functional Theory in Molecular Systems Using Deep Learning.. transformer single-paper
JAX-DFT (π₯26 Β· β 38K) - This library provides basic building blocks that can construct DFT calculations as a differentiable program. Apache-2
- [GitHub](https://github.com/google-research/google-research) (π¨βπ» 860 Β· π 8.3K Β· π 2.1K - 83% open Β· β±οΈ 09.04.2026):
git clone https://github.com/google-research/google-research
MALA (π₯16 Β· β 98 Β· π€) - Materials Learning Algorithms. A framework for machine learning materials properties from first-principles data. BSD-3
- [GitHub](https://github.com/mala-project/mala) (π¨βπ» 47 Β· π 27 Β· π¦ 2 Β· π 310 - 9% open Β· β±οΈ 16.09.2025):
git clone https://github.com/mala-project/mala
QHNet (π₯14 Β· β 750) - Artificial Intelligence Research for Science (AIRS). GPL-3.0 rep-learn
- [GitHub](https://github.com/divelab/AIRS) (π¨βπ» 36 Β· π 89 Β· π 32 - 18% open Β· β±οΈ 30.03.2026):
git clone https://github.com/divelab/AIRS
SALTED (π₯14 Β· β 42) - Symmetry-Adapted Learning of Three-dimensional Electron Densities (and their electrostatic response). GPL-3.0
- [GitHub](https://github.com/andreagrisafi/SALTED) (π¨βπ» 24 Β· π 6 Β· π 11 - 27% open Β· β±οΈ 26.03.2026):
git clone https://github.com/andreagrisafi/SALTED
HamGNN (π₯13 Β· β 190) - An E(3) equivariant Graph Neural Network for predicting electronic Hamiltonian matrix. GPL-3.0 rep-learn magnetism C-lang
- [GitHub](https://github.com/QuantumLab-ZY/HamGNN) (π¨βπ» 8 Β· π 44 Β· π 79 - 83% open Β· β±οΈ 09.04.2026):
git clone https://github.com/QuantumLab-ZY/HamGNN
Q-stack (π₯13 Β· β 19) - Stack of codes for dedicated pre- and post-processing tasks for Quantum Machine Learning (QML). MIT excited-states general-tool
- [GitHub](https://github.com/lcmd-epfl/Q-stack) (π¨βπ» 8 Β· π 7 Β· π 55 - 12% open Β· β±οΈ 06.03.2026):
git clone https://github.com/lcmd-epfl/Q-stack
DeePKS-kit (π₯9 Β· β 120 Β· π€) - a package for developing machine learning-based chemically accurate energy and density functional models. LGPL-3.0 ml-functional
- [GitHub](https://github.com/deepmodeling/deepks-kit) (π¨βπ» 7 Β· π 39 Β· π 32 - 46% open Β· β±οΈ 28.04.2025):
git clone https://github.com/deepmodeling/deepks-kit
CiderPress (π₯9 Β· β 18 Β· π€) - A high-performance software package for training and evaluating machine-learned XC functionals using the CIDER.. GPL-3.0 ml-functional C-lang
- [GitHub](https://github.com/mir-group/CiderPress) (π¨βπ» 2 Β· π 3 Β· β±οΈ 09.04.2025):
git clone https://github.com/mir-group/CiderPress
ACEhamiltonians (π₯9 Β· β 17 Β· π€) - Provides tools for constructing, fitting, and predicting self-consistent Hamiltonian and overlap matrices in solid-.. MIT Julia
- [GitHub](https://github.com/ACEsuit/ACEhamiltonians.jl) (π¨βπ» 5 Β· π 7 Β· π 5 - 40% open Β· β±οΈ 17.09.2025):
git clone https://github.com/ACEsuit/ACEhamiltonians.jl
dftio (π₯8 Β· β 15) - dftio is to assist machine learning communities to transcript DFT output into a format that is easy to read or used by.. LGPL-3.0 data-structures workflows
- [GitHub](https://github.com/deepmodeling/dftio) (π¨βπ» 5 Β· π 11 Β· π 8 - 50% open Β· β±οΈ 18.12.2025):
git clone https://github.com/deepmodeling/dftio
DeepH-E3 (π₯6 Β· β 110) - General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian. MIT magnetism
- [GitHub](https://github.com/Xiaoxun-Gong/DeepH-E3) (π¨βπ» 2 Β· π 28 Β· π 40 - 67% open Β· β±οΈ 27.01.2026):
git clone https://github.com/Xiaoxun-Gong/DeepH-E3
Show 26 hidden projects...
- DM21 (π₯20 Β· β 15K Β· π) - This package provides a PySCF interface to the DM21 (DeepMind 21) family of exchange-correlation functionals described.. Apache-2
- DeepH-pack (π₯12 Β· β 320 Β· π) - Deep neural networks for density functional theory Hamiltonian. LGPL-3.0 Julia
- Grad DFT (π₯10 Β· β 110 Β· π) - GradDFT is a JAX-based library enabling the differentiable design and experimentation of exchange-correlation.. Apache-2
- NeuralXC (π₯10 Β· β 36 Β· π) - Implementation of a machine learned density functional. BSD-3
- PROPhet (π₯9 Β· β 66 Β· π) - PROPhet is a code to integrate machine learning techniques with first-principles quantum chemistry approaches. GPL-3.0 ML-IAP MD single-paper C++
- Libnxc (π₯8 Β· β 21 Β· π) - A library for using machine-learned exchange-correlation functionals for density-functional theory. MPL-2.0 C++ Fortran
- ChargE3Net (π₯7 Β· β 71 Β· π) - [npj Comp. Mat.] Higher-order equivariant neural networks for charge density prediction in materials. MIT rep-learn
- Mat2Spec (π₯7 Β· β 30 Β· π) - Density of States Prediction for Materials Discovery via Contrastive Learning from Probabilistic Embeddings. MIT spectroscopy
- DeepDFT (π₯6 Β· β 89 Β· π) - Official implementation of DeepDFT model. MIT
- scdp (scalable charge density prediction) (π₯6 Β· β 40 Β· π) - [NeurIPS 2024] source code for A Recipe for Charge Density Prediction. MIT rep-learn single-paper
- charge-density-models (π₯6 Β· β 15 Β· π) - Tools to build charge density models using [fairchem](https://github.com/FAIR-Chem/fairchem). MIT rep-learn
- KSR-DFT (π₯6 Β· β 4 Β· π) - Kohn-Sham regularizer for machine-learned DFT functionals. Apache-2
- xDeepH (π₯5 Β· β 40 Β· π) - Extended DeepH (xDeepH) method for magnetic materials. LGPL-3.0 magnetism Julia
- InfGCN for Electron Density Estimation (π₯5 Β· β 16 Β· π) - Official implementation of the NeurIPS 23 spotlight paper of InfGCN. MIT rep-learn neural-operator
- rho_learn (π₯5 Β· β 4 Β· π) - A proof-of-concept workflow for torch-based electron density learning. MIT ML-DFT rep-eng
- ML-DFT (π₯4 Β· β 27 Β· π) - A package for density functional approximation using machine learning. MIT
- DeepCDP (π₯4 Β· β 6 Β· π) - DeepCDP: Deep learning Charge Density Prediction. Unlicensed
- CSNN (π₯4 Β· β 3 Β· π) - Primary codebase of CSNN - Concentric Spherical Neural Network for 3D Representation Learning. BSD-3
- rholearn (π₯4 Β· β 3 Β· π) - Learning and predicting electronic densities decomposed on a basis and global electronic densities of states at DFT.. MIT ML-DFT rep-eng density-of-states
- gprep (π₯4 Β· π) - Fitting DFTB repulsive potentials with GPR. MIT single-paper
- ofdft_nflows (π₯3 Β· β 11 Β· π) - Nomalizing flows for orbita-free DFT. Unlicensed generative
- APET (π₯3 Β· β 6 Β· π) - Atomic Positional Embedding-based Transformer. GPL-3.0 density-of-states transformer
- MALADA (π₯3 Β· β 1 Β· π€) - MALA Data Acquisition: Helpful tools to build data for MALA. BSD-3
- A3MD (π₯2 Β· β 8 Β· π) - MPNN-like + Analytic Density Model = Accurate electron densities. Unlicensed rep-learn single-paper
- MLDensity (π₯1 Β· β 7 Β· π) - Linear Jacobi-Legendre expansion of the charge density for machine learning-accelerated electronic structure.. Unlicensed
- kdft (π₯1 Β· β 2 Β· π) - The Kernel Density Functional (KDF) code allows generating ML based DFT functionals. Unlicensed
Tutorials, guides, cookbooks, recipes, etc.
π AI for Science 101 community-resource rep-learn
π AL4MS 2023 workshop tutorials active-learning
π Quantum Chemistry in the Age of Machine Learning - Book, 2022.
Deep Learning for Molecules and Materials Book (π₯13 Β· β 730) - Deep learning for molecules and materials book. Custom
- [GitHub](https://github.com/whitead/dmol-book) (π¨βπ» 19 Β· π 130 Β· π 180 - 17% open Β· β±οΈ 20.02.2026):
git clone https://github.com/whitead/dmol-book
AI4Chemistry course (π₯13 Β· β 260) - EPFL AI for chemistry course, Spring 2023. https://schwallergroup.github.io/ai4chem_course. MIT chemistry
- [GitHub](https://github.com/schwallergroup/ai4chem_course) (π¨βπ» 10 Β· π 59 Β· π 4 - 25% open Β· β±οΈ 03.04.2026):
git clone https://github.com/schwallergroup/ai4chem_course
Geometric GNN Dojo (π₯12 Β· β 520) - New to geometric GNNs: try our practical notebook, prepared for MPhil students at the University of Cambridge. MIT rep-learn
- [GitHub](https://github.com/chaitjo/geometric-gnn-dojo) (π¨βπ» 4 Β· π 52 Β· π 9 - 22% open Β· β±οΈ 09.10.2025):
git clone https://github.com/chaitjo/geometric-gnn-dojo
COSMO Software Cookbook (π₯12 Β· β 47) - A collection of simulation recipes for the atomic-scale modeling of materials and molecules. BSD-3
- [GitHub](https://github.com/lab-cosmo/atomistic-cookbook) (π¨βπ» 20 Β· π 8 Β· π 33 - 33% open Β· β±οΈ 07.04.2026):
git clone https://github.com/lab-cosmo/software-cookbook
MLforMaterials (π₯9 Β· β 140) - Online resource for a practical course in machine learning for materials research at Imperial College London.. MIT community-resource general-ml rep-eng materials-discovery
- [GitHub](https://github.com/aronwalsh/MLforMaterials) (π¨βπ» 2 Β· π 17 Β· β±οΈ 07.02.2026):
git clone https://github.com/aronwalsh/MLforMaterials
DSECOP (π₯9 Β· β 53 Β· π€) - This repository contains data science educational materials developed by DSECOP Fellows. CCO-1.0
- [GitHub](https://github.com/GDS-Education-Community-of-Practice/DSECOP) (π¨βπ» 14 Β· π 26 Β· π 8 - 12% open Β· β±οΈ 29.04.2025):
git clone https://github.com/GDS-Education-Community-of-Practice/DSECOP
iam-notebooks (π₯9 Β· β 38) - Jupyter notebooks for the lectures of the Introduction to Atomistic Modeling. Apache-2
- [GitHub](https://github.com/ceriottm/iam-notebooks) (π¨βπ» 6 Β· π 6 Β· β±οΈ 14.02.2026):
git clone https://github.com/ceriottm/iam-notebooks
jarvis-tools-notebooks (π₯8 Β· β 96 Β· π€) - This repository is no longer maintained. For the latest updates and continued development, please visit:.. NIST
- [GitHub](https://github.com/JARVIS-Materials-Design/jarvis-tools-notebooks) (π¨βπ» 6 Β· π 40 Β· β±οΈ 10.07.2025):
git clone https://github.com/JARVIS-Materials-Design/jarvis-tools-notebooks
DeepModeling Tutorials (π₯6 Β· β 16) - Tutorials for DeepModeling projects. Unlicensed
- [GitHub](https://github.com/deepmodeling/tutorials) (π¨βπ» 12 Β· π 24 Β· π 4 - 25% open Β· β±οΈ 11.03.2026):
git clone https://github.com/deepmodeling/tutorials
Show 20 hidden projects...
- DeepLearningLifeSciences (π₯12 Β· β 400 Β· π) - Example code from the book Deep Learning for the Life Sciences. MIT
- Introduction to AI-driven Science on Supercomputers: A Student Training Series (π₯11 Β· β 240) - Unlicensed general-ml rep-learn language-models
- OPTIMADE Tutorial Exercises (π₯9 Β· β 17 Β· π) - Tutorial exercises for the OPTIMADE API. MIT datasets
- RDKit Tutorials (π₯8 Β· β 310 Β· π) - Tutorials to learn how to work with the RDKit. Custom
- BestPractices (π₯8 Β· β 200 Β· π) - Things that you should (and should not) do in your Materials Informatics research. MIT
- MAChINE (π₯7 Β· β 1 Β· π) - Client-Server Web App to introduce usage of ML in materials science to beginners. MIT
- Applied AI for Materials (π₯6 Β· β 73 Β· π) - Course materials for Applied AI for Materials Science and Engineering. Unlicensed
- Machine Learning for Materials Hard and Soft (π₯6 Β· β 40 Β· π) - ESI-DCAFM-TACO-VDSP Summer School on Machine Learning for Materials Hard and Soft. Unlicensed
- ML for catalysis tutorials (π₯6 Β· β 12 Β· π) - A jupyter book repo for tutorial on how to use OCP ML models for catalysis. MIT
- Data Handling, DoE and Statistical Analysis for Material Chemists (π₯6 Β· β 4 Β· π) - Notebooks for workshops of DoE course, hosted by the Computational Materials Chemistry group at Uppsala University. GPL-3.0
- AI4Science101 (π₯5 Β· β 100 Β· π) - AI for Science. Unlicensed
- ML-in-chemistry-101 (π₯4 Β· β 87 Β· π) - The course materials for Machine Learning in Chemistry 101. Unlicensed
- MACE-tutorials (π₯4 Β· β 54) - Another set of tutorials for the MACE interatomic potential by one of the authors. MIT ML-IAP rep-learn MD
- DSM-CORE (π₯4 Β· β 17 Β· π€) - Data Science for Materials - Collection of Open Educational Resources. Unlicensed
- chemrev-gpr (π₯4 Β· β 12 Β· π) - Notebooks accompanying the paper on GPR in materials and molecules in Chemical Reviews 2020. Unlicensed
- AI4ChemMat Hands-On Series (π₯4 Β· β 1 Β· π) - Hands-On Series organized by Chemistry and Materials working group at Argonne Nat Lab. MPL-2.0
- PiNN Lab (π₯3 Β· β 3 Β· π) - Material for running a lab session on atomic neural networks. GPL-3.0
- MLDensity_tutorial (π₯2 Β· β 12 Β· π) - Tutorial files to work with ML for the charge density in molecules and solids. Unlicensed
- LAMMPS-style pair potentials with GAP (π₯2 Β· β 4 Β· π) - A tutorial on how to create LAMMPS-style pair potentials and use them in combination with GAP potentials to run MD.. Unlicensed ML-IAP MD rep-eng
- MALA Tutorial (π₯2 Β· β 2 Β· π) - A full MALA hands-on tutorial. Unlicensed
Projects that focus on explainability and model interpretability in atomistic ML.
exmol (π₯19 Β· β 350 Β· π€) - Explainer for black box models that predict molecule properties. MIT
- [GitHub](https://github.com/ur-whitelab/exmol) (π¨βπ» 9 Β· π 46 Β· π 72 - 8% open Β· β±οΈ 08.05.2025):
git clone https://github.com/ur-whitelab/exmol
Show 3 hidden projects...
- Linear vs blackbox (π₯3 Β· β 2 Β· π) - Code and data related to the publication: Interpretable models for extrapolation in scientific machine learning. MIT XAI single-paper rep-eng
- MEGAN: Multi Explanation Graph Attention Student (π₯2 Β· β 12) - Minimal implementation of graph attention student model architecture. MIT rep-learn
- XElemNet (π₯2 Β· π) - Using explainable artificial intelligence (XAI) techniques to analyze ElemNet... Unlicensed rep-eng single-paper
Projects and models that focus on quantities of electronic structure methods, which do not fit into either of the categories ML-WFT or ML-DFT.
DeePTB (π₯14 Β· β 100) - DeePTB: A deep learning package for tight-binding Hamiltonian with ab initio accuracy. LGPL-3.0 ML-DFT
- [GitHub](https://github.com/deepmodeling/DeePTB) (π¨βπ» 15 Β· π 29 Β· π¦ 4 Β· π 64 - 37% open Β· β±οΈ 19.03.2026):
git clone https://github.com/deepmodeling/DeePTB
Show 5 hidden projects...
- QDF for molecule (π₯8 Β· β 230 Β· π) - Quantum deep field: data-driven wave function, electron density generation, and energy prediction and extrapolation.. MIT
- QMLearn (π₯5 Β· β 12 Β· π) - Quantum Machine Learning by learning one-body reduced density matrices in the AO basis... MIT
- q-pac (π₯5 Β· β 6 Β· π) - Kernel charge equilibration method. MIT electrostatics
- halex (π₯5 Β· β 4 Β· π) - Hamiltonian Learning for Excited States https://doi.org/10.48550/arXiv.2311.00844. Unlicensed excited-states
- e3psi (π₯3 Β· β 7 Β· π) - Equivariant machine learning library for learning from electronic structures. LGPL-3.0
General tools for atomistic machine learning.
RDKit (π₯39 Β· β 3.4K) - BSD-3 C++ cheminformatics
- [GitHub](https://github.com/rdkit/rdkit) (π¨βπ» 270 Β· π 970 Β· π¦ 3 Β· π 4.3K - 15% open Β· β±οΈ 07.04.2026):
git clone https://github.com/rdkit/rdkit
conda install -c rdkit rdkit
DeepChem (π₯33 Β· β 6.7K Β· π) - Democratizing Deep-Learning for Drug Discovery, Quantum Chemistry, Materials Science and Biology. MIT
- [GitHub](https://github.com/deepchem/deepchem) (π¨βπ» 260 Β· π 2.2K Β· π¦ 660 Β· π 2.2K - 41% open Β· β±οΈ 20.02.2026):
git clone https://github.com/deepchem/deepchem
conda install -c conda-forge deepchem
docker pull deepchemio/deepchem
Matminer (π₯31 Β· β 580) - Data mining for materials science. Custom
- [GitHub](https://github.com/hackingmaterials/matminer) (π¨βπ» 57 Β· π 210 Β· π¦ 480 Β· π 230 - 12% open Β· β±οΈ 10.02.2026):
git clone https://github.com/hackingmaterials/matminer
conda install -c conda-forge matminer
QUIP (π₯26 Β· β 390) - libAtoms/QUIP molecular dynamics framework: https://libatoms.github.io. GPL-2.0 MD ML-IAP rep-eng Fortran
- [GitHub](https://github.com/libAtoms/QUIP) (π¨βπ» 86 Β· π 130 Β· π₯ 860 Β· π¦ 46 Β· π 500 - 23% open Β· β±οΈ 31.03.2026):
git clone https://github.com/libAtoms/QUIP
docker pull libatomsquip/quip
JARVIS-Tools (π₯24 Β· β 370 Β· π€) - About JARVIS-Tools: an open-source software package for data-driven atomistic materials design. Publications:.. Custom
- [GitHub](https://github.com/usnistgov/jarvis) (π¨βπ» 16 Β· π 140 Β· π 95 - 52% open Β· β±οΈ 25.08.2025):
git clone https://github.com/usnistgov/jarvis
conda install -c conda-forge jarvis-tools
MAML (π₯20 Β· β 450) - Python for Materials Machine Learning, Materials Descriptors, Machine Learning Force Fields, Deep Learning, etc. BSD-3
- [GitHub](https://github.com/materialyzeai/maml) (π¨βπ» 39 Β· π 95 Β· π¦ 17 Β· π 76 - 14% open Β· β±οΈ 14.02.2026):
git clone https://github.com/materialsvirtuallab/maml
Molfeat (π₯20 Β· β 220 Β· π€) - molfeat - the hub for all your molecular featurizers. Apache-2 cheminformatics rep-eng rep-learn generative language-models pretrained
- [GitHub](https://github.com/datamol-io/molfeat) (π¨βπ» 19 Β· π 27 Β· π¦ 73 Β· π 61 - 27% open Β· β±οΈ 27.05.2025):
git clone https://github.com/datamol-io/molfeat
conda install -c conda-forge molfeat
Scikit-Matter (π₯19 Β· β 93) - A collection of scikit-learn compatible utilities that implement methods born out of the materials science and.. BSD-3 scikit-learn
- [GitHub](https://github.com/scikit-learn-contrib/scikit-matter) (π¨βπ» 20 Β· π 25 Β· π₯ 19 Β· π 84 - 26% open Β· β±οΈ 01.04.2026):
git clone https://github.com/scikit-learn-contrib/scikit-matter
conda install -c conda-forge skmatter
AtomAI (π₯18 Β· β 230 Β· π€) - Deep and Machine Learning for Microscopy. MIT computer-vision USL experimental-data
- [GitHub](https://github.com/pycroscopy/atomai) (π¨βπ» 6 Β· π 41 Β· π¦ 13 Β· π 20 - 55% open Β· β±οΈ 23.06.2025):
git clone https://github.com/pycroscopy/atomai
Artificial Intelligence for Science (AIRS) (π₯14 Β· β 750) - Artificial Intelligence Research for Science (AIRS). GPL-3.0 license rep-learn generative ML-IAP MD ML-DFT ML-WFT biomolecules
- [GitHub](https://github.com/divelab/AIRS) (π¨βπ» 36 Β· π 89 Β· π 32 - 18% open Β· β±οΈ 30.03.2026):
git clone https://github.com/divelab/AIRS
MLatom (π₯14 Β· β 140) - AI-enhanced computational chemistry. MIT UIP ML-IAP MD ML-DFT ML-ESM transfer-learning active-learning spectroscopy structure-optimization
- [GitHub](https://github.com/dralgroup/mlatom) (π¨βπ» 6 Β· π 17 Β· π 8 - 37% open Β· β±οΈ 09.03.2026):
git clone https://github.com/dralgroup/mlatom
MAST-ML (π₯14 Β· β 130) - MAterials Simulation Toolkit for Machine Learning (MAST-ML). MIT
- [GitHub](https://github.com/uw-cmg/MAST-ML) (π¨βπ» 19 Β· π 61 Β· π₯ 170 Β· π 220 - 14% open Β· β±οΈ 10.10.2025):
git clone https://github.com/uw-cmg/MAST-ML
Show 12 hidden projects...
- QML (π₯17 Β· β 210 Β· π) - QML: Quantum Machine Learning. MIT
- Automatminer (π₯16 Β· β 170 Β· π) - An automatic engine for predicting materials properties. Custom autoML
- XenonPy (π₯16 Β· β 150 Β· π) - XenonPy is a Python Software for Materials Informatics. BSD-3
- AMPtorch (π₯11 Β· β 61 Β· π) - AMPtorch: Atomistic Machine Learning Package (AMP) - PyTorch. GPL-3.0
- OpenChem (π₯10 Β· β 740 Β· π) - OpenChem: Deep Learning toolkit for Computational Chemistry and Drug Design Research. MIT
- JAXChem (π₯7 Β· β 81 Β· π) - JAXChem is a JAX-based deep learning library for complex and versatile chemical modeling. MIT
- uncertainty_benchmarking (π₯7 Β· β 43 Β· π) - Various code/notebooks to benchmark different ways we could estimate uncertainty in ML predictions. Unlicensed benchmarking probabilistic
- torchchem (π₯7 Β· β 38 Β· π) - An experimental repo for experimenting with PyTorch models. MIT
- Equisolve (π₯6 Β· β 5 Β· π) - A ML toolkit package utilizing the metatensor data format to build models for the prediction of equivariant properties.. BSD-3 ML-IAP
- quantum-structure-ml (π₯3 Β· β 3 Β· π) - Multi-class classification model for predicting the magnetic order of magnetic structures and a binary classification.. Unlicensed magnetism benchmarking
- ACEatoms (π₯3 Β· β 2 Β· π) - Generic code for modelling atomic properties using ACE. Custom Julia
- Magpie (π₯3) - Materials Agnostic Platform for Informatics and Exploration (Magpie). MIT Java
Projects that implement generative models for atomistic ML.
GT4SD (π₯16 Β· β 370 Β· π€) - GT4SD, an open-source library to accelerate hypothesis generation in the scientific discovery process. MIT pretrained drug-discovery rep-learn
- [GitHub](https://github.com/GT4SD/gt4sd-core) (π¨βπ» 20 Β· π 79 Β· π 120 - 11% open Β· β±οΈ 18.09.2025):
git clone https://github.com/GT4SD/gt4sd-core
SLICES and MatterGPT (π₯14 Β· β 140) - SLICES: An Invertible, Invariant, and String-based Crystal Representation [2023, Nature Communications] MatterGPT,.. LGPL-2.1 rep-eng language-models transformer materials-discovery structure-prediction
- [GitHub](https://github.com/xiaohang007/SLICES) (π¨βπ» 2 Β· π 58 Β· π¦ 7 Β· π 17 - 23% open Β· β±οΈ 03.03.2026):
git clone https://github.com/xiaohang007/SLICES
docker pull xiaohang07/slices
synspace (π₯12 Β· β 48 Β· π€) - Synthesis generative model. MIT
- [GitHub](https://github.com/whitead/synspace) (π¨βπ» 2 Β· π 4 Β· π¦ 37 Β· π 4 - 50% open Β· β±οΈ 24.04.2025):
git clone https://github.com/whitead/synspace
SchNetPack G-SchNet (π₯11 Β· β 63) - G-SchNet extension for SchNetPack. MIT
- [GitHub](https://github.com/atomistic-machine-learning/schnetpack-gschnet) (π¨βπ» 3 Β· π 11 Β· β±οΈ 13.11.2025):
git clone https://github.com/atomistic-machine-learning/schnetpack-gschnet
SiMGen (π₯9 Β· β 29 Β· π€) - Zero Shot Molecular Generation via Similarity Kernels. MIT viz
- [GitHub](https://github.com/RokasEl/simgen) (π¨βπ» 4 Β· π 5 Β· π¦ 2 Β· π 5 - 20% open Β· β±οΈ 27.08.2025):
git clone https://github.com/RokasEl/simgen
Show 12 hidden projects...
- MoLeR (π₯15 Β· β 320 Β· π) - Implementation of MoLeR: a generative model of molecular graphs which supports scaffold-constrained generation. MIT
- PMTransformer (π₯14 Β· β 120 Β· π) - Universal Transfer Learning in Porous Materials, including MOFs. MIT transfer-learning pretrained transformer
- EDM (π₯9 Β· β 560 Β· π) - E(3) Equivariant Diffusion Model for Molecule Generation in 3D. MIT
- G-SchNet (π₯8 Β· β 150 Β· π) - G-SchNet - a generative model for 3d molecular structures. MIT
- bVAE-IM (π₯8 Β· β 14 Β· π) - Implementation of Chemical Design with GPU-based Ising Machine. MIT QML single-paper
- molecular-vae (π₯7 Β· β 71 Β· π) - Pytorch implementation of the paper Automatic Chemical Design Using a Data-Driven Continuous Representation of.. MIT rep-learn cheminformatics single-paper
- cG-SchNet (π₯7 Β· β 65 Β· π) - cG-SchNet - a conditional generative neural network for 3d molecular structures. MIT
- COATI (π₯6 Β· β 120 Β· π) - COATI: multi-modal contrastive pre-training for representing and traversing chemical space. Apache-2 drug-discovery multimodal pretrained rep-learn
- rxngenerator (π₯5 Β· β 12 Β· π) - A generative model for molecular generation via multi-step chemical reactions. MIT
- MolSLEPA (π₯5 Β· β 7 Β· π) - Interpretable Fragment-based Molecule Design with Self-learning Entropic Population Annealing. MIT XAI
- Mapping out phase diagrams with generative classifiers (π₯4 Β· β 8 Β· π) - Repository for our ``Mapping out phase diagrams with generative models paper. MIT phase-transition
- descriptors-inversion (π₯4 Β· β 6 Β· π) - Local inversion of the chemical environment representations. MIT rep-eng single-paper
Machine learning interatomic potentials (aka ML-IAP, MLIAP, MLIP, MLP) and force fields (ML-FF) for molecular dynamics.
fairchem (π₯32 Β· β 2K Β· π) - FAIR Chemistrys library of machine learning methods for chemistry. Formerly known as Open Catalyst Project. MIT pretrained UIP rep-learn catalysis
- [GitHub](https://github.com/facebookresearch/fairchem) (π¨βπ» 69 Β· π 450 Β· π 570 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/FAIR-Chem/fairchem
pip install fairchem-core
NequIP (π₯30 Β· β 890) - NequIP is a code for building E(3)-equivariant interatomic potentials. MIT
- [GitHub](https://github.com/mir-group/nequip) (π¨βπ» 39 Β· π 200 Β· π¦ 49 Β· π 120 - 3% open Β· β±οΈ 25.03.2026):
git clone https://github.com/mir-group/nequip
conda install -c conda-forge nequip
DeePMD-kit (π₯29 Β· β 1.9K) - A deep learning package for many-body potential energy representation and molecular dynamics. LGPL-3.0 MD workflows C++
- [GitHub](https://github.com/deepmodeling/deepmd-kit) (π¨βπ» 84 Β· π 600 Β· π₯ 69K Β· π¦ 47 Β· π 1K - 13% open Β· β±οΈ 08.04.2026):
git clone https://github.com/deepmodeling/deepmd-kit
conda install -c deepmodeling deepmd-kit
docker pull deepmodeling/deepmd-kit
TorchANI (π₯26 Β· β 540) - TorchANI 2.0 is an open-source library that supports training, development, and research of ANI-style neural network.. MIT
- [GitHub](https://github.com/aiqm/torchani) (π¨βπ» 22 Β· π 140 Β· π¦ 73 Β· π 180 - 4% open Β· β±οΈ 15.12.2025):
git clone https://github.com/aiqm/torchani
conda install -c conda-forge torchani
MatCalc (π₯25 Β· β 140) - A python library for calculating materials properties from the PES. BSD-3 workflows benchmarking UIP pretrained model-repository
- [GitHub](https://github.com/materialyzeai/matcalc) (π¨βπ» 23 Β· π 35 Β· π¦ 23 Β· π 39 - 15% open Β· β±οΈ 25.03.2026):
git clone https://github.com/materialsvirtuallab/matcalc
Metatrain (π₯25 Β· β 65) - Train, fine-tune, and manipulate machine learning models for atomistic systems. BSD-3 workflows benchmarking rep-eng rep-learn
- [GitHub](https://github.com/metatensor/metatrain) (π¨βπ» 34 Β· π 24 Β· π₯ 79 Β· π¦ 14 Β· π 300 - 31% open Β· β±οΈ 09.04.2026):
git clone https://github.com/metatensor/metatrain
MACE (π₯24 Β· β 1.1K) - MACE - Fast and accurate machine learning interatomic potentials with higher order equivariant message passing. MIT
- [GitHub](https://github.com/ACEsuit/mace) (π¨βπ» 69 Β· π 390 Β· π 580 - 20% open Β· β±οΈ 01.03.2026):
git clone https://github.com/ACEsuit/mace
TorchMD-NET (π₯22 Β· β 470) - Training neural network potentials. MIT MD rep-learn transformer pretrained
- [GitHub](https://github.com/torchmd/torchmd-net) (π¨βπ» 20 Β· π 97 Β· π₯ 190 Β· π 120 - 25% open Β· β±οΈ 17.03.2026):
git clone https://github.com/torchmd/torchmd-net
conda install -c conda-forge torchmd-net
janus-core (π₯21 Β· β 44) - Tools for machine learnt interatomic potentials. BSD-3 benchmarking workflows structure-optimization MD transport-phenomena
- [GitHub](https://github.com/stfc/janus-core) (π¨βπ» 11 Β· π 17 Β· π₯ 240 Β· π¦ 16 Β· π 310 - 15% open Β· β±οΈ 09.04.2026):
git clone https://github.com/stfc/janus-core
apax (π₯20 Β· β 36 Β· π) - A flexible and performant framework for training machine learning potentials. MIT
- [GitHub](https://github.com/apax-hub/apax) (π¨βπ» 11 Β· π 7 Β· π¦ 7 Β· π 180 - 17% open Β· β±οΈ 24.03.2026):
git clone https://github.com/apax-hub/apax
sGDML (π₯18 Β· β 170 Β· π€) - sGDML - Reference implementation of the Symmetric Gradient Domain Machine Learning model. MIT
- [GitHub](https://github.com/stefanch/sGDML) (π¨βπ» 8 Β· π 42 Β· π¦ 13 Β· π 22 - 50% open Β· β±οΈ 13.06.2025):
git clone https://github.com/stefanch/sGDML
KLIFF (π₯18 Β· β 40) - KIM-based Learning-Integrated Fitting Framework for interatomic potentials. LGPL-2.1 probabilistic workflows
- [GitHub](https://github.com/openkim/kliff) (π¨βπ» 14 Β· π 22 Β· π¦ 4 Β· π 57 - 42% open Β· β±οΈ 28.02.2026):
git clone https://github.com/openkim/kliff
conda install -c conda-forge kliff
Allegro (π₯17 Β· β 470) - Allegro is an open-source code for building highly scalable and accurate equivariant deep learning interatomic.. MIT
- [GitHub](https://github.com/mir-group/allegro) (π¨βπ» 9 Β· π 73 Β· π 50 - 2% open Β· β±οΈ 24.02.2026):
git clone https://github.com/mir-group/allegro
Autoplex (π₯16 Β· β 140) - Code for automated fitting of machine learned interatomic potentials. GPL-3.0 benchmarking workflows
- [GitHub](https://github.com/autoatml/autoplex) (π¨βπ» 15 Β· π 22 Β· π¦ 2 Β· π 140 - 25% open Β· β±οΈ 09.04.2026):
git clone https://github.com/autoatml/autoplex
Graph-PES (π₯15 Β· β 120) - train and use graph-based ML models of potential energy surfaces. MIT rep-learn UIP MD pretrained
- [GitHub](https://github.com/vldgroup/graph-pes) (π¨βπ» 5 Β· π 14 Β· π¦ 3 Β· π 17 - 17% open Β· β±οΈ 20.02.2026):
git clone https://github.com/jla-gardner/graph-pes
Neural Force Field (π₯14 Β· β 290) - Neural Network Force Field based on PyTorch. MIT pretrained
- [GitHub](https://github.com/learningmatter-mit/NeuralForceField) (π¨βπ» 45 Β· π 61 Β· π 23 - 21% open Β· β±οΈ 10.02.2026):
git clone https://github.com/learningmatter-mit/NeuralForceField
NNPOps (π₯14 Β· β 100) - High-performance operations for neural network potentials. MIT MD C++
- [GitHub](https://github.com/openmm/NNPOps) (π¨βπ» 11 Β· π 18 Β· π 66 - 42% open Β· β±οΈ 04.02.2026):
git clone https://github.com/openmm/NNPOps
conda install -c conda-forge nnpops
MLIPX - Machine-Learned Interatomic Potential eXploration (π₯14 Β· β 100) - Machine-Learned Interatomic Potential eXploration (mlipx) is designed at BASF for evaluating machine-learned.. MIT benchmarking viz workflows
- [GitHub](https://github.com/basf/mlipx) (π¨βπ» 5 Β· π 8 Β· π¦ 6 Β· π 18 - 33% open Β· β±οΈ 16.10.2025):
git clone https://github.com/basf/mlipx
wfl (π₯14 Β· β 43) - Workflow is a Python toolkit for building interatomic potential creation and atomistic simulation workflows. GPL-2.0 workflows HTC
- [GitHub](https://github.com/libAtoms/workflow) (π¨βπ» 20 Β· π 21 Β· π¦ 5 Β· π 170 - 42% open Β· β±οΈ 22.12.2025):
git clone https://github.com/libAtoms/workflow
MACE-Jax (π₯12 Β· β 94) - Equivariant machine learning interatomic potentials in JAX. MIT
- [GitHub](https://github.com/ACEsuit/mace-jax) (π¨βπ» 5 Β· π 21 Β· π 10 - 50% open Β· β±οΈ 10.02.2026):
git clone https://github.com/ACEsuit/mace-jax
aiida-mlip (π₯12 Β· β 16) - machine learning interatomic potentials aiida plugin. BSD-3 workflows structure-optimization MD
- [GitHub](https://github.com/ElliottKasoar/aiida-mlip) (π¨βπ» 7 Β· π 10 Β· β±οΈ 30.01.2026):
git clone https://github.com/ElliottKasoar/aiida-mlip
calorine (π₯12 Β· β 15) - A Python package for constructing and sampling neuroevolution potential models. https://doi.org/10.21105/joss.06264. Custom
- [PyPi](https://pypi.org/project/calorine) (π₯ 42K / month Β· π¦ 8 Β· β±οΈ 24.01.2026):
- [GitLab](https://gitlab.com/materials-modeling/calorine) (π 6 Β· π 120 - 9% open Β· β±οΈ 24.01.2026):
git clone https://gitlab.com/materials-modeling/calorine
PiNN (π₯11 Β· β 120) - A Python library for building atomic neural networks. BSD-3
- [GitHub](https://github.com/Teoroo-CMC/PiNN) (π¨βπ» 8 Β· π 39 Β· π 7 - 14% open Β· β±οΈ 26.03.2026):
git clone https://github.com/Teoroo-CMC/PiNN
tinker-hp (π₯10 Β· β 100) - Tinker-HP: High-Performance Massively Parallel Tinker for CPUs & GPUs. Custom
- [GitHub](https://github.com/TinkerTools/tinker-hp) (π¨βπ» 12 Β· π 26 Β· π 29 - 24% open Β· β±οΈ 26.01.2026):
git clone https://github.com/TinkerTools/tinker-hp
DeepMD-GNN (π₯10 Β· β 54) - DeePMD-kit plugin for various graph neural network models. LGPL-3.0 rep-learn MD UIP C++
- [GitHub](https://github.com/deepmodeling/deepmd-gnn) (π¨βπ» 8 Β· π 9 Β· π 23 - 52% open Β· β±οΈ 13.02.2026):
git clone https://github.com/deepmodeling/deepmd-gnn
ALF (π₯10 Β· β 39) - A framework for performing active learning for training machine-learned interatomic potentials. Custom active-learning
- [GitHub](https://github.com/lanl/ALF) (π¨βπ» 8 Β· π 13 Β· β±οΈ 21.03.2026):
git clone https://github.com/lanl/alf
ACEfit (π₯10 Β· β 8) - MIT Julia
- [GitHub](https://github.com/ACEsuit/ACEfit.jl) (π¨βπ» 10 Β· π 8 Β· π 60 - 38% open Β· β±οΈ 20.03.2026):
git clone https://github.com/ACEsuit/ACEfit.jl
PyNEP (π₯9 Β· β 69) - A python interface of the machine learning potential NEP used in GPUMD. MIT
- [GitHub](https://github.com/bigd4/PyNEP) (π¨βπ» 10 Β· π 17 Β· π 14 - 42% open Β· β±οΈ 27.10.2025):
git clone https://github.com/bigd4/PyNEP
ACE1.jl (π₯9 Β· β 23 Β· π€) - Atomic Cluster Expansion for Modelling Invariant Atomic Properties. Custom Julia
- [GitHub](https://github.com/ACEsuit/ACE1.jl) (π¨βπ» 9 Β· π 7 Β· π 46 - 47% open Β· β±οΈ 15.04.2025):
git clone https://github.com/ACEsuit/ACE1.jl
TurboGAP (π₯9 Β· β 20) - The TurboGAP code. Custom Fortran
- [GitHub](https://github.com/mcaroba/turbogap) (π¨βπ» 9 Β· π 13 Β· π 12 - 66% open Β· β±οΈ 08.04.2026):
git clone https://github.com/mcaroba/turbogap
GAP (π₯8 Β· β 46) - Gaussian Approximation Potential (GAP). Custom
- [GitHub](https://github.com/libAtoms/GAP) (π¨βπ» 13 Β· π 20 Β· β±οΈ 03.01.2026):
git clone https://github.com/libAtoms/GAP
Asparagus (π₯8 Β· β 12) - Program Package for Sampling, Training and Applying ML-based Potential models https://doi.org/10.48550/arXiv.2407.15175. MIT workflows sampling MD
- [GitHub](https://github.com/MMunibas/Asparagus) (π¨βπ» 11 Β· π 6 Β· β±οΈ 07.04.2026):
git clone https://github.com/MMunibas/Asparagus
Show 46 hidden projects...
- MEGNet (π₯22 Β· β 560 Β· π) - Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals. BSD-3 multifidelity
- Ultra-Fast Force Fields (UF3) (π₯15 Β· β 71 Β· π) - UF3: a python library for generating ultra-fast interatomic potentials. Apache-2
- PyXtalFF (π₯14 Β· β 93 Β· π) - Machine Learning Interatomic Potential Predictions. MIT
- n2p2 (π₯13 Β· β 240 Β· π) - n2p2 - A Neural Network Potential Package. GPL-3.0 C++
- Pacemaker (π₯13 Β· β 110 Β· π) - Python package for fitting atomic cluster expansion (ACE) potentials. Custom
- TensorMol (π₯12 Β· β 280 Β· π) - Tensorflow + Molecules = TensorMol. GPL-3.0 single-paper
- ANI-1 (π₯12 Β· β 230 Β· π) - ANI-1 neural net potential with python interface (ASE). MIT
- So3krates (MLFF) (π₯12 Β· β 140 Β· π) - Build neural networks for machine learning force fields with JAX. MIT
- SIMPLE-NN (π₯12 Β· β 48 Β· π) - SIMPLE-NN(SNU Interatomic Machine-learning PotentiaL packagE version Neural Network). GPL-3.0
- CCS_fit (π₯10 Β· β 10 Β· π) - Curvature Constrained Splines. GPL-3.0
- DimeNet (π₯9 Β· β 350 Β· π) - DimeNet and DimeNet++ models, as proposed in Directional Message Passing for Molecular Graphs (ICLR 2020) and Fast and.. Custom
- SchNet (π₯9 Β· β 290 Β· π) - SchNet - a deep learning architecture for quantum chemistry. MIT
- GemNet (π₯9 Β· β 220 Β· π) - GemNet model in PyTorch, as proposed in GemNet: Universal Directional Graph Neural Networks for Molecules (NeurIPS.. Custom
- ACE.jl (π₯9 Β· β 67 Β· π) - Parameterisation of Equivariant Properties of Particle Systems. Custom Julia
- Point Edge Transformer (PET) (π₯9 Β· β 34 Β· π) - Point Edge Transformer. MIT rep-learn transformer
- EquiformerV2 (π₯8 Β· β 340 Β· π) - [ICLR 2024] EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations. MIT pretrained UIP rep-learn
- AIMNet (π₯8 Β· β 110 Β· π) - Atoms In Molecules Neural Network Potential. MIT single-paper
- SIMPLE-NN v2 (π₯8 Β· β 43 Β· π) - SIMPLE-NN is an open package that constructs Behler-Parrinello-type neural-network interatomic potentials from ab.. GPL-3.0
- Atomistic Adversarial Attacks (π₯8 Β· β 40 Β· π) - Code for performing adversarial attacks on atomistic systems using NN potentials. MIT probabilistic
- SNAP (π₯8 Β· β 38 Β· π) - Repository for spectral neighbor analysis potential (SNAP) model development. BSD-3
- NNsforMD (π₯8 Β· β 11 Β· π) - Neural network class for molecular dynamics to predict potential energy, forces and non-adiabatic couplings. MIT
- MEGNetSparse (π₯8 Β· β 5 Β· π) - A library imlementing a graph neural network with sparse representation from Code for Kazeev, N., Al-Maeeni, A.R.,.. MIT material-defect
- PhysNet (π₯7 Β· β 120 Β· π) - Code for training PhysNet models. MIT electrostatics
- MLIP-3 (π₯6 Β· β 27 Β· π) - MLIP-3: Active learning on atomic environments with Moment Tensor Potentials (MTP). BSD-2 C++
- testing-framework (π₯6 Β· β 11 Β· π) - The purpose of this repository is to aid the testing of a large number of interatomic potentials for a variety of.. Unlicensed benchmarking
- PANNA (π₯6 Β· β 11 Β· π) - A package to train and validate all-to-all connected network models for BP[1] and modified-BP[2] type local atomic.. MIT benchmarking
- MLXDM (π₯6 Β· β 9 Β· π) - A Neural Network Potential with Rigorous Treatment of Long-Range Dispersion https://doi.org/10.1039/D2DD00150K. MIT long-range
- Alchemical learning (π₯6 Β· β 3 Β· π) - Code for the Modeling high-entropy transition metal alloys with alchemical compression article. BSD-3 rep-eng Defects & Disorder
- BPNET (π₯6 Β· β 3 Β· π€) - Fast Behler-Parrinello type neural networks in Fortran2008. MIT rep-eng Fortran
- ACE1Pack.jl (π₯6 Β· β 1 Β· π) - Provides convenience functionality for the usage of ACE1.jl, ACEfit.jl, JuLIP.jl for fitting interatomic potentials.. MIT Julia
- glp (π₯5 Β· β 26 Β· π) - tools for graph-based machine-learning potentials in jax. MIT
- NequIP-JAX (π₯5 Β· β 24 Β· π) - JAX implementation of the NequIP interatomic potential. Unlicensed
- Allegro-Legato (π₯5 Β· β 21 Β· π) - An extension of Allegro with enhanced robustness and time-to-failure. MIT MD
- TensorPotential (π₯5 Β· β 13 Β· π) - Tensorpotential is a TensorFlow based tool for development, fitting ML interatomic potentials from electronic.. Custom
- GN-MM (π₯5 Β· β 11 Β· π) - The Gaussian Moment Neural Network (GM-NN) package developed for large-scale atomistic simulations employing atomistic.. MIT active-learning MD rep-eng magnetism
- MatML (π₯4 Β· β 9) - Full MatML Docker image, including MatGL, MatCalc, MatPES and LAMMPS with ML-GNNP and ML-SNAP. BSD-3 MD UIP rep-learn pretrained
- PeriodicPotentials (π₯4 Β· π) - A Periodic table app that displays potentials based on the selected elements. MIT community-resource viz JavaScript
- Allegro-JAX (π₯3 Β· β 22 Β· π€) - JAX implementation of the Allegro interatomic potential. MIT
- ACE Workflows (π₯3 Β· π) - Workflow Examples for ACE Models. Unlicensed Julia workflows
- PyFLAME (π₯3 Β· π) - An automated approach for developing neural network interatomic potentials with FLAME.. Unlicensed active-learning structure-prediction structure-optimization rep-eng Fortran
- SingleNN (π₯2 Β· β 9 Β· π) - An efficient package for training and executing neural-network interatomic potentials. Unlicensed C++
- mag-ace (π₯2 Β· β 6 Β· π€) - Magnetic ACE potential. FORTRAN interface for LAMMPS SPIN package. Unlicensed magnetism MD Fortran
- AisNet (π₯2 Β· β 3 Β· π) - A Universal Interatomic Potential Neural Network with Encoded Local Environment Features.. MIT
- RuNNer (π₯2) - The RuNNer Neural Network Energy Representation is a Fortran-based framework for the construction of Behler-.. GPL-3.0 Fortran
- nnp-pre-training (π₯1 Β· β 6 Β· π) - Synthetic pre-training for neural-network interatomic potentials. Unlicensed pretrained MD
- mlp (π₯1 Β· β 1 Β· π) - Proper orthogonal descriptors for efficient and accurate interatomic potentials... Unlicensed Julia
Projects that use (large) language models (LMs, LLMs) or natural language procesing (NLP) techniques for atomistic ML.
π MaCBench Leaderboard - Leaderboard for multimodal language models for chemistry & materials research. community-resource benchmarking datasets
paper-qa (π₯18 Β· β 8.3K) - LLM Chain for answering questions from docs. Unlicensed ai-agent
- [GitHub]() (π 840):
git clone https://github.com/whitead/paper-qa
ChemBench (π₯17 Β· β 140 Β· π€) - How good are LLMs at chemistry?. MIT benchmarking multimodal
- [GitHub](https://github.com/lamalab-org/chembench) (π¨βπ» 15 Β· π 16 Β· π¦ 3 Β· π 340 - 16% open Β· β±οΈ 11.09.2025):
git clone https://github.com/lamalab-org/chembench
ChatMOF (π₯12 Β· β 110 Β· π€) - Predict and Inverse design for metal-organic framework with large-language models (llms). MIT generative
- [GitHub](https://github.com/Yeonghun1675/ChatMOF) (π¨βπ» 2 Β· π 22 Β· π¦ 3 Β· β±οΈ 15.05.2025):
git clone https://github.com/Yeonghun1675/ChatMOF
AtomGPT (π₯11 Β· β 50 Β· π€) - https://atomgpt.org. Custom generative pretrained transformer
- [GitHub](https://github.com/usnistgov/atomgpt) (π¨βπ» 7 Β· π 10 Β· β±οΈ 21.08.2025):
git clone https://github.com/usnistgov/atomgpt
LLaMP (π₯8 Β· β 91) - [EMNLP 25] A web app and Python API for multi-modal RAG framework to ground LLMs on high-fidelity materials.. BSD-3 multimodal RAG materials-discovery pretrained JavaScript Python
- [GitHub](https://github.com/chiang-yuan/llamp) (π¨βπ» 6 Β· π 13 Β· π 25 - 32% open Β· β±οΈ 11.11.2025):
git clone https://github.com/chiang-yuan/llamp
LLM-Prop (π₯7 Β· β 54) - A repository for the LLM-Prop implementation. MIT
- [GitHub](https://github.com/vertaix/LLM-Prop) (π¨βπ» 6 Β· π 10 Β· π 3 - 66% open Β· β±οΈ 31.01.2026):
git clone https://github.com/vertaix/LLM-Prop
LLM4Chem (π₯6 Β· β 110 Β· π€) - Official code repo for the paper LlaSMol: Advancing Large Language Models for Chemistry with a Large-Scale,.. MIT cheminformatics datasets
- [GitHub](https://github.com/OSU-NLP-Group/LLM4Chem) (π¨βπ» 2 Β· π 19 Β· β±οΈ 09.06.2025):
git clone https://github.com/OSU-NLP-Group/LLM4Chem
Show 17 hidden projects...
- OpenBioML ChemNLP (π₯17 Β· β 170 Β· π) - ChemNLP project. MIT datasets
- ChemCrow (π₯16 Β· β 890 Β· π) - Open source package for the accurate solution of reasoning-intensive chemical tasks. MIT ai-agent
- ChemDataExtractor (π₯16 Β· β 350 Β· π) - Automatically extract chemical information from scientific documents. MIT literature-data
- mat2vec (π₯12 Β· β 640 Β· π) - Supplementary Materials for Tshitoyan et al. Unsupervised word embeddings capture latent knowledge from materials.. MIT rep-learn
- gptchem (π₯12 Β· β 260 Β· π) - Use GPT-3 to solve chemistry problems. MIT
- nlcc (π₯11 Β· β 46 Β· π) - Natural language computational chemistry command line interface. MIT single-paper
- NIST ChemNLP (π₯11 Β· β 28 Β· π€) - chemnlp. MIT literature-data
- MoLFormer (π₯9 Β· β 390 Β· π) - Repository for MolFormer. Apache-2 transformer pretrained drug-discovery
- MolSkill (π₯9 Β· β 120 Β· π) - Extracting medicinal chemistry intuition via preference machine learning. MIT drug-discovery recommender
- chemlift (π₯7 Β· β 45 Β· π) - Language-interfaced fine-tuning for chemistry. MIT
- BERT-PSIE-TC (π₯6 Β· β 15 Β· π) - A dataset of Curie temperatures automatically extracted from scientific literature with the use of the BERT-PSIE.. MIT magnetism
- crystal-text-llm (π₯5 Β· β 120 Β· π) - Large language models to generate stable crystals. CC-BY-NC-4.0 materials-discovery
- SciBot (π₯5 Β· β 33 Β· π) - SciBot is a simple demo of building a domain-specific chatbot for science. Unlicensed ai-agent
- Cephalo (π₯5 Β· β 12 Β· π) - Multimodal Vision-Language Models for Bio-Inspired Materials Analysis and Design. Apache-2 generative multimodal pretrained
- MAPI_LLM (π₯5 Β· β 9 Β· π) - A LLM application developed during the LLM March MADNESS Hackathon https://doi.org/10.1039/D3DD00113J. MIT ai-agent dataset
- CatBERTa (π₯4 Β· β 28 Β· π) - Large Language Model for Catalyst Property Prediction. Unlicensed transformer catalysis
- ChemDataWriter (π₯3 Β· β 13 Β· π) - ChemDataWriter is a transformer-based library for automatically generating research books in the chemistry area. MIT literature-data
Projects that implement materials discovery methods using atomistic ML.
SMACT (π₯26 Β· β 130) - Python package to aid materials design and informatics. MIT HTC structure-prediction electrostatics
- [GitHub](https://github.com/WMD-group/SMACT) (π¨βπ» 49 Β· π 30 Β· π¦ 75 Β· β±οΈ 09.04.2026):
git clone https://github.com/WMD-group/SMACT
conda install -c conda-forge smact
MatterGen (π₯18 Β· β 1.7K) - Official implementation of MatterGen -- a generative model for inorganic materials design across the periodic table.. MIT generative structure-prediction pretrained
- [GitHub](https://github.com/microsoft/mattergen) (π¨βπ» 13 Β· π 310 Β· π 140 - 5% open Β· β±οΈ 27.02.2026):
git clone https://github.com/microsoft/mattergen
Materials Discovery: GNoME (π₯12 Β· β 1.2K Β· π) - Graph Networks for Materials Science (GNoME) and dataset of 381,000 novel stable materials. Apache-2 UIP datasets rep-learn proprietary
- [GitHub](https://github.com/google-deepmind/materials_discovery) (π¨βπ» 2 Β· π 180 Β· π 25 - 84% open Β· β±οΈ 03.03.2025):
git clone https://github.com/google-deepmind/materials_discovery
aviary (π₯12 Β· β 61) - The Wren sits on its Roost in the Aviary. MIT
- [GitHub](https://github.com/CompRhys/aviary) (π¨βπ» 6 Β· π 13 Β· π¦ 1 Β· π 34 - 11% open Β· β±οΈ 06.01.2026):
git clone https://github.com/CompRhys/aviary
BOSS (π₯12 Β· β 27) - Bayesian Optimization Structure Search (BOSS). Apache-2 probabilistic
- [PyPi](https://pypi.org/project/aalto-boss) (π₯ 580 / month Β· β±οΈ 28.11.2025):
- [GitLab](https://gitlab.com/cest-group/boss) (π 14 Β· π 39 - 17% open Β· β±οΈ 28.11.2025):
git clone https://gitlab.com/cest-group/boss
AGOX (π₯9 Β· β 18) - AGOX is a package for global optimization of atomic system using e.g. the energy calculated from density functional.. GPL-3.0 structure-optimization
- [PyPi](https://pypi.org/project/agox) (π₯ 130 / month Β· π¦ 3 Β· β±οΈ 04.02.2026):
- [GitLab](https://gitlab.com/agox/agox) (π 8 Β· π 28 - 32% open Β· β±οΈ 04.02.2026):
git clone https://gitlab.com/agox/agox
Show 7 hidden projects...
- Computational Autonomy for Materials Discovery (CAMD) (π₯7 Β· β 1 Β· π) - Agent-based sequential learning software for materials discovery. Apache-2
- MAGUS (π₯5 Β· β 100 Β· π) - Machine learning And Graph theory assisted Universal structure Searcher. Unlicensed structure-prediction active-learning
- CSPML (crystal structure prediction with machine learning-based element substitution) (π₯5 Β· β 29 Β· π) - Original implementation of CSPML. MIT structure-prediction
- SPINNER (π₯4 Β· β 15 Β· π) - SPINNER (Structure Prediction of Inorganic crystals using Neural Network potentials with Evolutionary and Random.. GPL-3.0 C++ structure-prediction
- ML-atomate (π₯4 Β· β 7 Β· π) - Machine learning-assisted Atomate code for autonomous computational materials screening. GPL-3.0 active-learning workflows
- closed-loop-acceleration-benchmarks (π₯4 Β· π) - Data and scripts in support of the publication By how much can closed-loop frameworks accelerate computational.. MIT materials-discovery active-learning single-paper
- sl_discovery (π₯3 Β· β 5 Β· π) - Data processing and models related to Quantifying the performance of machine learning models in materials discovery. Apache-2 materials-discovery single-paper
Projects that implement mathematical objects used in atomistic machine learning.
cuEquivariance (π₯24 Β· β 390) - cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used.. Apache-2 rep-learn
- [GitHub](https://github.com/NVIDIA/cuEquivariance) (π¨βπ» 8 Β· π 27 Β· π 80 - 21% open Β· β±οΈ 08.04.2026):
git clone https://github.com/NVIDIA/cuEquivariance
pip install cuequivariance
conda install -c conda-forge cuequivariance
KFAC-JAX (π₯21 Β· β 320) - Second Order Optimization and Curvature Estimation with K-FAC in JAX. Apache-2
- [GitHub](https://github.com/google-deepmind/kfac-jax) (π¨βπ» 20 Β· π 29 Β· π¦ 11 Β· π 36 - 69% open Β· β±οΈ 01.04.2026):
git clone https://github.com/google-deepmind/kfac-jax
OpenEquivariance (π₯20 Β· β 140) - OpenEquivariance: a fast, open-source GPU JIT kernel generator for the Clebsch-Gordon Tensor Product. BSD-3 rep-learn
- [GitHub](https://github.com/PASSIONLab/OpenEquivariance) (π¨βπ» 4 Β· π 9 Β· π¦ 2 Β· π 33 - 3% open Β· β±οΈ 24.03.2026):
git clone https://github.com/PASSIONLab/OpenEquivariance
pip install openequivariance
SpheriCart (π₯20 Β· β 95) - Multi-language library for the calculation of spherical harmonics in Cartesian coordinates. MIT
- [GitHub](https://github.com/lab-cosmo/sphericart) (π¨βπ» 13 Β· π 17 Β· π₯ 960 Β· π¦ 10 Β· π 47 - 34% open Β· β±οΈ 26.03.2026):
git clone https://github.com/lab-cosmo/sphericart
gpax (π₯17 Β· β 230 Β· π€) - Gaussian Processes for Experimental Sciences. MIT probabilistic active-learning
- [GitHub](https://github.com/ziatdinovmax/gpax) (π¨βπ» 6 Β· π 29 Β· π¦ 6 Β· π 43 - 23% open Β· β±οΈ 04.07.2025):
git clone https://github.com/ziatdinovmax/gpax
Polynomials4ML.jl (π₯12 Β· β 14) - Polynomials for ML: fast evaluation, batching, differentiation. MIT Julia
- [GitHub](https://github.com/ACEsuit/Polynomials4ML.jl) (π¨βπ» 12 Β· π 7 Β· π 61 - 14% open Β· β±οΈ 29.12.2025):
git clone https://github.com/ACEsuit/Polynomials4ML.jl
GElib (π₯9 Β· β 26) - C++/CUDA library for SO(3) equivariant operations. MPL-2.0 C++
- [GitHub](https://github.com/risi-kondor/GElib) (π¨βπ» 6 Β· π 3 Β· π 8 - 50% open Β· β±οΈ 21.10.2025):
git clone https://github.com/risi-kondor/GElib
COSMO Toolbox (π₯6 Β· β 8) - Assorted libraries and utilities for atomistic simulation analysis. Unlicensed C++
- [GitHub](https://github.com/lab-cosmo/toolbox) (π¨βπ» 10 Β· π 7 Β· β±οΈ 01.04.2026):
git clone https://github.com/lab-cosmo/toolbox
Show 6 hidden projects...
- lie-nn (π₯9 Β· β 36 Β· π) - Tools for building equivariant polynomials on reductive Lie groups. MIT rep-learn
- LapJAX (π₯8 Β· β 75 Β· π) - A JAX based package designed for efficient second order operators (e.g., laplacian) computation. MIT
- EquivariantOperators.jl (π₯6 Β· β 18 Β· π) - This package is deprecated. Functionalities are migrating to Porcupine.jl. MIT Julia
- cnine (π₯4 Β· β 5) - Cnine tensor library. Unlicensed C++
- torch_spex (π₯3 Β· β 2 Β· π) - Spherical expansions in PyTorch. Unlicensed
- Wigner Kernels (π₯1 Β· β 2 Β· π) - Collection of programs to benchmark Wigner kernels. Unlicensed benchmarking
Projects that simplify the integration of molecular dynamics and atomistic machine learning.
JAX-MD (π₯28 Β· β 1.4K) - Differentiable, Hardware Accelerated, Molecular Dynamics. Apache-2
- [GitHub](https://github.com/jax-md/jax-md) (π¨βπ» 45 Β· π 230 Β· π¦ 82 Β· π 190 - 29% open Β· β±οΈ 05.04.2026):
git clone https://github.com/jax-md/jax-md
TorchSim (π₯24 Β· β 440 Β· π) - Torch-native, batchable, atomistic simulations. MIT HTC UIP ML-IAP structure-optimization
- [GitHub](https://github.com/TorchSim/torch-sim) (π¨βπ» 33 Β· π 90 Β· π 160 - 13% open Β· β±οΈ 09.04.2026):
git clone https://github.com/Radical-AI/torch-sim
pip install torch-sim-atomistic
GPUMD (π₯22 Β· β 740) - GPUMD is a highly efficient general-purpose molecular dynamic (MD) package and enables machine-learned potentials.. GPL-3.0 ML-IAP C++ electrostatics
- [GitHub](https://github.com/brucefan1983/GPUMD) (π¨βπ» 57 Β· π 180 Β· π 270 - 6% open Β· β±οΈ 08.04.2026):
git clone https://github.com/brucefan1983/GPUMD
mlcolvar (π₯22 Β· β 140) - A unified framework for machine learning collective variables for enhanced sampling simulations. MIT sampling
- [GitHub](https://github.com/luigibonati/mlcolvar) (π¨βπ» 14 Β· π 35 Β· π¦ 11 Β· π 110 - 11% open Β· β±οΈ 09.04.2026):
git clone https://github.com/luigibonati/mlcolvar
FitSNAP (π₯17 Β· β 180) - Software for generating machine-learning interatomic potentials for LAMMPS. GPL-2.0
- [GitHub](https://github.com/FitSNAP/FitSNAP) (π¨βπ» 24 Β· π 65 Β· π₯ 15 Β· π 84 - 25% open Β· β±οΈ 17.10.2025):
git clone https://github.com/FitSNAP/FitSNAP
conda install -c conda-forge fitsnap3
OpenMM-ML (π₯17 Β· β 160) - High level API for using machine learning models in OpenMM simulations. MIT ML-IAP
- [GitHub](https://github.com/openmm/openmm-ml) (π¨βπ» 8 Β· π 40 Β· π¦ 2 Β· π 72 - 33% open Β· β±οΈ 25.03.2026):
git clone https://github.com/openmm/openmm-ml
conda install -c conda-forge openmm-ml
Psiflow (π₯15 Β· β 140) - scalable molecular simulation. MIT ML-IAP active-learning sampling
- [GitHub](https://github.com/molmod/psiflow) (π¨βπ» 5 Β· π 15 Β· π 59 - 16% open Β· β±οΈ 30.03.2026):
git clone https://github.com/molmod/psiflow
pair_allegro (π₯15 Β· β 61) - LAMMPS pair styles for NequIP and Allegro deep learning interatomic potentials. MIT ML-IAP rep-learn
- [GitHub](https://github.com/mir-group/pair_nequip_allegro) (π¨βπ» 6 Β· π 10 Β· π 49 - 20% open Β· β±οΈ 10.03.2026):
git clone https://github.com/mir-group/pair_allegro
DMFF (π₯12 Β· β 190) - DMFF (Differentiable Molecular Force Field) is a Jax-based python package that provides a full differentiable.. LGPL-3.0 C++
- [GitHub](https://github.com/deepmodeling/DMFF) (π¨βπ» 17 Β· π 47 Β· π 33 - 39% open Β· β±οΈ 07.04.2026):
git clone https://github.com/deepmodeling/DMFF
pair_nequip (π₯10 Β· β 44 Β· π€) - LAMMPS pair style for NequIP. MIT ML-IAP rep-learn
- [GitHub](https://github.com/mir-group/pair_nequip) (π¨βπ» 3 Β· π 14 Β· π 33 - 39% open Β· β±οΈ 25.04.2025):
git clone https://github.com/mir-group/pair_nequip
PACE (π₯10 Β· β 31) - The LAMMPS ML-IAP `pair_style pace`, aka Atomic Cluster Expansion (ACE), aka ML-PACE,.. Custom
- [GitHub](https://github.com/ICAMS/lammps-user-pace) (π¨βπ» 8 Β· π 16 Β· π 11 - 45% open Β· β±οΈ 03.12.2025):
git clone https://github.com/ICAMS/lammps-user-pace
MUSE (π₯5 Β· β 7 Β· π€) - A python package for fast building amorphous solids and liquid mixtures from @materialsproject computed structures and.. MIT ML-IAP Defects & Disorder
- [GitHub](https://github.com/chiang-yuan/muse) (π¨βπ» 2 Β· π¦ 1 Β· β±οΈ 15.05.2025):
git clone https://github.com/chiang-yuan/muse
Show 3 hidden projects...
- openmm-torch (π₯17 Β· β 220 Β· π) - OpenMM plugin to define forces with neural networks. Custom ML-IAP C++
- SOMD (π₯4 Β· β 17) - Molecular dynamics package designed for the SIESTA DFT code. AGPL-3.0 ML-IAP active-learning
- interface-lammps-mlip-3 (π₯3 Β· β 5 Β· π) - An interface between LAMMPS and MLIP (version 3). GPL-2.0
Projects that focus on probabilistic, Bayesian, Gaussian process and adversarial methods for atomistic ML, for optimization, uncertainty quantification (UQ), etc.
thermo (π₯5 Β· β 17) - Data-driven risk-conscious thermoelectric materials discovery. MIT materials-discovery experimental-data active-learning transport-phenomena
- [GitHub](https://github.com/janosh/thermo) (π¨βπ» 2 Β· π 4 Β· β±οΈ 12.11.2025):
git clone https://github.com/janosh/thermo
Projects that focus on reinforcement learning for atomistic ML.
Show 2 hidden projects...
- ReLeaSE (π₯11 Β· β 370 Β· π) - Deep Reinforcement Learning for de-novo Drug Design. MIT drug-discovery
- CatGym (π₯5 Β· β 13 Β· π) - Surface segregation using Deep Reinforcement Learning. GPL
Projects that offer implementations of representations aka descriptors, fingerprints of atomistic systems, and models built with them, aka feature engineering.
cdk (π₯27 Β· β 580) - The Chemistry Development Kit. LGPL-2.1 cheminformatics Java
- [GitHub](https://github.com/cdk/cdk) (π¨βπ» 170 Β· π 180 Β· π₯ 98K Β· π 330 - 2% open Β· β±οΈ 03.04.2026):
git clone https://github.com/cdk/cdk
<dependency>
<groupId>org.openscience.cdk</groupId>
<artifactId>cdk-bundle</artifactId>
<version>[VERSION]</version>
</dependency>
DScribe (π₯24 Β· β 460 Β· π€) - DScribe is a python package for creating machine learning descriptors for atomistic systems. Apache-2
- [GitHub](https://github.com/SINGROUP/dscribe) (π¨βπ» 18 Β· π 96 Β· π¦ 290 Β· π 110 - 12% open Β· β±οΈ 27.09.2025):
git clone https://github.com/SINGROUP/dscribe
conda install -c conda-forge dscribe
ChemML (π₯19 Β· β 170) - ChemML is a machine learning and informatics program suite for the chemical and materials sciences. BSD-3 cheminformatics active-learning workflows
- [GitHub](https://github.com/hachmannlab/chemml) (π¨βπ» 17 Β· π 34 Β· π₯ 14 Β· π¦ 8 Β· π 13 - 53% open Β· β±οΈ 20.03.2026):
git clone https://github.com/hachmannlab/chemml
MODNet (π₯15 Β· β 110 Β· π€) - MODNet: a framework for machine learning materials properties. MIT pretrained small-data transfer-learning
- [GitHub](https://github.com/ppdebreuck/modnet) (π¨βπ» 11 Β· π 34 Β· π¦ 11 Β· π 64 - 50% open Β· β±οΈ 02.05.2025):
git clone https://github.com/ppdebreuck/modnet
Featomic (π₯15 Β· β 79) - Computing representations for atomistic machine learning. BSD-3 Rust C++
- [GitHub](https://github.com/metatensor/featomic) (π¨βπ» 19 Β· π 18 Β· π₯ 850 Β· π 89 - 49% open Β· β±οΈ 31.03.2026):
git clone https://github.com/metatensor/featomic
GlassPy (π₯15 Β· β 38) - Python module for scientists working with glass materials. GPL-3.0
- [GitHub](https://github.com/drcassar/glasspy) (π¨βπ» 2 Β· π 8 Β· π¦ 7 Β· π 16 - 43% open Β· β±οΈ 17.03.2026):
git clone https://github.com/drcassar/glasspy
pySIPFENN (π₯15 Β· β 24) - Python python toolset for Structure-Informed Property and Feature Engineering with Neural Networks. It offers unique.. LGPL-3.0 material-defect Defects & Disorder pretrained transfer-learning
- [GitHub](https://github.com/PhasesResearchLab/pySIPFENN) (π¨βπ» 5 Β· π 5 Β· π₯ 120 Β· π¦ 7 Β· π 8 - 62% open Β· β±οΈ 20.01.2026):
git clone https://github.com/PhasesResearchLab/pySIPFENN
conda install -c conda-forge pysipfenn
SISSO (π₯11 Β· β 350) - A data-driven method combining symbolic regression and compressed sensing for accurate & interpretable models. Apache-2 Fortran
- [GitHub](https://github.com/rouyang2017/SISSO) (π¨βπ» 3 Β· π 95 Β· π 78 - 23% open Β· β±οΈ 26.01.2026):
git clone https://github.com/rouyang2017/SISSO
BenchML (π₯11 Β· β 15) - ML benchmarking and pipeling framework. Apache-2 benchmarking
- [GitHub](https://github.com/capoe/benchml) (π¨βπ» 9 Β· π 6 Β· π 13 - 23% open Β· β±οΈ 28.10.2025):
git clone https://github.com/capoe/benchml
PDynA (π₯10 Β· β 51) - Python package to analyse the structural dynamics of perovskites. MIT MD
- [GitHub](https://github.com/WMD-group/PDynA) (π¨βπ» 4 Β· π 5 Β· π¦ 2 Β· β±οΈ 14.01.2026):
git clone https://github.com/WMD-group/PDynA
MOLPIPx (π₯9 Β· β 50) - Differentiable version of Permutationally Invariant Polynomial (PIP) models in JAX and Rust. Apache-2 Python Rust
- [GitHub](https://github.com/ChemAI-Lab/molpipx) (π¨βπ» 13 Β· π 1 Β· β±οΈ 27.03.2026):
git clone https://github.com/ChemAI-Lab/molpipx
ElemNet (π₯7 Β· β 100) - Deep Learning the Chemistry of Materials From Only Elemental Composition for Enhancing Materials Property Prediction. Unlicensed single-paper
- [GitHub](https://github.com/NU-CUCIS/ElemNet) (π¨βπ» 4 Β· π 35 Β· π 8 - 50% open Β· β±οΈ 13.01.2026):
git clone https://github.com/NU-CUCIS/ElemNet
fplib (π₯6 Β· β 8 Β· π€) - libfp is a library for calculating crystalline fingerprints and measuring similarities of materials. MIT C-lang single-paper
- [GitHub](https://github.com/Rutgers-ZRG/libfp) (π¨βπ» 2 Β· π 1 Β· π¦ 2 Β· β±οΈ 22.09.2025):
git clone https://github.com/zhuligs/fplib
soap_turbo (π₯5 Β· β 8) - soap_turbo comprises a series of libraries to be used in combination with QUIP/GAP and TurboGAP. Custom Fortran
- [GitHub](https://github.com/libAtoms/soap_turbo) (π¨βπ» 4 Β· π 8 Β· π 8 - 62% open Β· β±οΈ 22.01.2026):
git clone https://github.com/libAtoms/soap_turbo
Show 17 hidden projects...
- CatLearn (π₯16 Β· β 120 Β· π) - GPL-3.0 surface-science
- ElementEmbeddings (π₯15 Β· β 51 Β· π) - Python package to interact with high-dimensional representations of the chemical elements. MIT XAI USL viz
- Librascal (π₯12 Β· β 83 Β· π) - A scalable and versatile library to generate representations for atomic-scale learning. LGPL-2.1
- CBFV (π₯12 Β· β 29 Β· π) - Tool to quickly create a composition-based feature vector. Unlicensed
- SkipAtom (π₯10 Β· β 28 Β· π) - Distributed representations of atoms, inspired by the Skip-gram model. MIT
- cmlkit (π₯9 Β· β 33 Β· π) - tools for machine learning in condensed matter physics and quantum chemistry. MIT benchmarking
- NICE (π₯7 Β· β 12 Β· π) - NICE (N-body Iteratively Contracted Equivariants) is a set of tools designed for the calculation of invariant and.. MIT
- SISSO++ (π₯7 Β· β 6 Β· π) - C++ Implementation of SISSO with python bindings. Apache-2 C++
- milad (π₯6 Β· β 34 Β· π) - Moment Invariants Local Atomic Descriptor. GPL-3.0 generative
- SA-GPR (π₯6 Β· β 23 Β· π) - Public repository for symmetry-adapted Gaussian Process Regression (SA-GPR). LGPL-3.0 C-lang
- SOAPxx (π₯6 Β· β 7 Β· π) - A SOAP implementation. GPL-2.0 C++
- pyLODE (π₯6 Β· β 3 Β· π) - Pythonic implementation of LOng Distance Equivariants. Apache-2 electrostatics
- AMP (π₯6 Β· π) - Amp is an open-source package designed to easily bring machine-learning to atomistic calculations. Unlicensed
- MXenes4HER (π₯5 Β· β 7 Β· π) - Predicting hydrogen evolution (HER) activity over 4500 MXene materials https://doi.org/10.1039/D3TA00344B. GPL-3.0 materials-discovery catalysis scikit-learn single-paper
- automl-materials (π₯4 Β· β 5 Β· π) - AutoML for Regression Tasks on Small Tabular Data in Materials Design. MIT autoML benchmarking single-paper
- magnetism-prediction (π₯4 Β· β 2 Β· π€) - DFT-aided Machine Learning Search for Magnetism in Fe-based Bimetallic Chalcogenides. Apache-2 magnetism single-paper
- ML-for-CurieTemp-Predictions (π₯3 Β· β 2 Β· π) - Machine Learning Predictions of High-Curie-Temperature Materials. MIT single-paper magnetism
General models that learn a representations aka embeddings of atomistic systems, such as message-passing neural networks (MPNN).
Deep Graph Library (DGL) (π₯36 Β· β 14K Β· π€) - Python package built to ease deep learning on graph, on top of existing DL frameworks. Apache-2
- [GitHub](https://github.com/dmlc/dgl) (π¨βπ» 300 Β· π 3K Β· π¦ 4.2K Β· π 3K - 20% open Β· β±οΈ 31.07.2025):
git clone https://github.com/dmlc/dgl
conda install -c dglteam dgl
PyG Models (π₯34 Β· β 24K) - Representation learning models implemented in PyTorch Geometric. MIT general-ml
- [GitHub](https://github.com/pyg-team/pytorch_geometric) (π¨βπ» 560 Β· π 4K Β· π¦ 11K Β· π 4K - 31% open Β· β±οΈ 08.04.2026):
git clone https://github.com/pyg-team/pytorch_geometric
e3nn (π₯29 Β· β 1.2K) - A modular framework for neural networks with Euclidean symmetry. MIT
- [GitHub](https://github.com/e3nn/e3nn) (π¨βπ» 38 Β· π 180 Β· π¦ 610 Β· π 180 - 17% open Β· β±οΈ 13.02.2026):
git clone https://github.com/e3nn/e3nn
conda install -c conda-forge e3nn
MatGL (Materials Graph Library) (π₯28 Β· β 530) - Graph deep learning library for materials. BSD-3 ML-IAP pretrained multifidelity
- [GitHub](https://github.com/materialyzeai/matgl) (π¨βπ» 25 Β· π 110 Β· π¦ 95 Β· π 160 - 2% open Β· β±οΈ 08.04.2026):
git clone https://github.com/materialsvirtuallab/matgl
docker pull materialsvirtuallab/matgl
SchNetPack (π₯27 Β· β 910) - SchNetPack - Deep Neural Networks for Atomistic Systems. MIT
- [GitHub](https://github.com/atomistic-machine-learning/schnetpack) (π¨βπ» 43 Β· π 250 Β· π¦ 110 Β· π 280 - 1% open Β· β±οΈ 17.03.2026):
git clone https://github.com/atomistic-machine-learning/schnetpack
e3nn-jax (π₯23 Β· β 230) - jax library for E3 Equivariant Neural Networks. Apache-2
- [GitHub](https://github.com/e3nn/e3nn-jax) (π¨βπ» 8 Β· π 19 Β· π¦ 83 Β· π 26 - 15% open Β· β±οΈ 01.04.2026):
git clone https://github.com/e3nn/e3nn-jax
ALIGNN (π₯21 Β· β 310 Β· π€) - Atomistic Line Graph Neural Network https://scholar.google.com/citations?user=9Q-tNnwAAAAJ.. Custom
- [GitHub](https://github.com/usnistgov/alignn) (π¨βπ» 8 Β· π 110 Β· π 87 - 67% open Β· β±οΈ 25.08.2025):
git clone https://github.com/usnistgov/alignn
Uni-Mol (π₯17 Β· β 1.1K Β· π€) - Official Repository for the Uni-Mol Series Methods. MIT pretrained
- [GitHub](https://github.com/deepmodeling/Uni-Mol) (π¨βπ» 20 Β· π 170 Β· π₯ 23K Β· π 230 - 48% open Β· β±οΈ 29.05.2025):
git clone https://github.com/deepmodeling/Uni-Mol
HydraGNN (π₯16 Β· β 100) - Distributed PyTorch implementation of multi-headed graph convolutional neural networks. BSD-3
- [GitHub](https://github.com/ORNL/HydraGNN) (π¨βπ» 18 Β· π 38 Β· π¦ 3 Β· π 56 - 30% open Β· β±οΈ 03.04.2026):
git clone https://github.com/ORNL/HydraGNN
hippynn (π₯13 Β· β 94) - python library for atomistic machine learning. Custom workflows
- [GitHub](https://github.com/lanl/hippynn) (π¨βπ» 20 Β· π 34 Β· π¦ 4 Β· π 41 - 34% open Β· β±οΈ 04.03.2026):
git clone https://github.com/lanl/hippynn
Compositionally-Restricted Attention-Based Network (CrabNet) (π₯13 Β· β 17 Β· π€) - Predict materials properties using only the composition information!. MIT
- [GitHub](https://github.com/sparks-baird/CrabNet) (π¨βπ» 6 Β· π 7 Β· π¦ 16 Β· π 19 - 84% open Β· β±οΈ 04.06.2025):
git clone https://github.com/sparks-baird/CrabNet
SE(3)-Transformers (π₯9 Β· β 580) - code for the SE3 Transformers paper: https://arxiv.org/abs/2006.10503. MIT single-paper transformer
- [GitHub](https://github.com/FabianFuchsML/se3-transformer-public) (π¨βπ» 2 Β· π 76 Β· π 29 - 37% open Β· β±οΈ 03.04.2026):
git clone https://github.com/FabianFuchsML/se3-transformer-public
UVVisML (π₯8 Β· β 35 Β· π€) - Predict optical properties of molecules with machine learning. MIT optical-properties single-paper probabilistic
- [GitHub](https://github.com/learningmatter-mit/uvvisml) (π¨βπ» 1 Β· π 10 Β· π 2 - 50% open Β· β±οΈ 30.07.2025):
git clone https://github.com/learningmatter-mit/uvvisml
DeeperGATGNN (π₯7 Β· β 65) - Scalable graph neural networks for materials property prediction. MIT
- [GitHub](https://github.com/usccolumbia/deeperGATGNN) (π¨βπ» 3 Β· π 8 Β· β±οΈ 02.02.2026):
git clone https://github.com/usccolumbia/deeperGATGNN
Crystalframer (π₯7 Β· β 18) - The official code respository for Rethinking the role of frames for SE(3)-invariant crystal structure modeling (ICLR.. MIT transformer single-paper
- [GitHub](https://github.com/omron-sinicx/crystalframer) (π¨βπ» 3 Β· π 3 Β· π₯ 11 Β· β±οΈ 16.10.2025):
git clone https://github.com/omron-sinicx/crystalframer
Show 48 hidden projects...
- dgl-lifesci (π₯24 Β· β 800 Β· π) - Python package for graph neural networks in chemistry and biology. Apache-2
- NVIDIA Deep Learning Examples for Tensor Cores (π₯20 Β· β 15K Β· π) - State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and.. Custom educational drug-discovery
- DIG: Dive into Graphs (π₯20 Β· β 2K Β· π) - A library for graph deep learning research. GPL-3.0
- escnn (π₯18 Β· β 520 Β· π) - Equivariant Steerable CNNs Library for Pytorch https://quva-lab.github.io/escnn/. Custom
- kgcnn (π₯17 Β· β 120 Β· π) - Graph convolutions in Keras with TensorFlow, PyTorch or Jax. MIT
- Graphormer (π₯15 Β· β 2.4K Β· π) - Graphormer is a general-purpose deep learning backbone for molecular modeling. MIT transformer pretrained
- benchmarking-gnns (π₯14 Β· β 2.7K Β· π) - Repository for benchmarking graph neural networks (JMLR 2023). MIT single-paper benchmarking
- Crystal Graph Convolutional Neural Networks (CGCNN) (π₯13 Β· β 850 Β· π) - Crystal graph convolutional neural networks for predicting material properties. MIT
- xtal2png (π₯13 Β· β 39 Β· π) - Encode/decode a crystal structure to/from a grayscale PNG image for direct use with image-based machine learning.. MIT computer-vision
- Neural fingerprint (nfp) (π₯12 Β· β 62 Β· π) - Keras layers for end-to-end learning with rdkit and pymatgen. Custom
- FAENet (π₯12 Β· β 34 Β· π) - Frame Averaging Equivariant GNN for materials modeling. MIT
- pretrained-gnns (π₯10 Β· β 1.1K Β· π) - Strategies for Pre-training Graph Neural Networks. MIT pretrained
- GDC (π₯10 Β· β 280 Β· π) - Graph Diffusion Convolution, as proposed in Diffusion Improves Graph Learning (NeurIPS 2019). MIT generative
- Atom2Vec (π₯10 Β· β 37 Β· π) - Atom2Vec: a simple way to describe atoms for machine learning. MIT
- GATGNN: Global Attention Graph Neural Network (π₯9 Β· β 85 Β· π) - Pytorch Repository for our work: Graph convolutional neural networks with global attention for improved materials.. MIT
- ai4material_design (π₯9 Β· β 8 Β· π) - Code for Kazeev, N., Al-Maeeni, A.R., Romanov, I. et al. Sparse representation for machine learning the properties of.. Apache-2 pretrained material-defect
- molecularGNN_smiles (π₯8 Β· β 340 Β· π) - The code of a graph neural network (GNN) for molecules, which is based on learning representations of r-radius.. Apache-2
- Equiformer (π₯8 Β· β 280 Β· π) - [ICLR 2023 Spotlight] Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs. MIT transformer
- graphite (π₯8 Β· β 110 Β· π) - A repository for implementing graph network models based on atomic structures. MIT
- GNNOpt (π₯8 Β· β 33 Β· π) - Universal Ensemble-Embedding Graph Neural Network for Direct Prediction of Optical Spectra from Crystal Structures. MIT optical-properties single-paper
- T-e3nn (π₯8 Β· β 19 Β· π) - Time-reversal Euclidean neural networks based on e3nn. MIT magnetism
- tensorfieldnetworks (π₯7 Β· β 160 Β· π) - Rotation- and translation-equivariant neural networks for 3D point clouds. MIT
- DTNN (π₯7 Β· β 78 Β· π) - Deep Tensor Neural Network. MIT
- Graph-Aware-Transformers (π₯7 Β· β 69 Β· π) - Graph-Aware Attention for Adaptive Dynamics in Transformers. Apache-2 transformer graph-data pretrained single-paper
- Cormorant (π₯7 Β· β 60 Β· π) - Codebase for Cormorant Neural Networks. Custom
- AdsorbML (π₯7 Β· β 44 Β· π) - MIT surface-science single-paper
- escnn_jax (π₯7 Β· β 32 Β· π) - Equivariant Steerable CNNs Library for Pytorch https://quva-lab.github.io/escnn/. Custom
- CGAT (π₯7 Β· β 31 Β· π) - Crystal graph attention neural networks for materials prediction. MIT
- Crystalformer (π₯7 Β· β 28 Β· π) - The official code respository for Crystalformer: Infinitely Connected Attention for Periodic Structure Encoding (ICLR.. MIT transformer single-paper
- Geom3D (π₯6 Β· β 130 Β· π) - Geom3D: Geometric Modeling on 3D Structures, NeurIPS 2023. MIT benchmarking single-paper
- matsciml (π₯6 Β· β 130 Β· π) - Open MatSci ML Toolkit is a framework for prototyping and scaling out deep learning models for materials discovery.. MIT workflows benchmarking
- MACE-Layer (π₯6 Β· β 47 Β· π) - Higher order equivariant graph neural networks for 3D point clouds. MIT
- charge_transfer_nnp (π₯6 Β· β 37 Β· π) - Graph neural network potential with charge transfer. MIT electrostatics
- FieldSchNet (π₯6 Β· β 24 Β· π) - Deep neural network for molecules in external fields. MIT
- GLAMOUR (π₯6 Β· β 24 Β· π) - Graph Learning over Macromolecule Representations. MIT single-paper
- ML4pXRDs (π₯6 Β· β 3 Β· π) - Contains code to train neural networks based on simulated powder XRDs from synthetic crystals. MIT XRD single-paper
- Autobahn (π₯5 Β· β 30 Β· π) - Repository for Autobahn: Automorphism Based Graph Neural Networks. MIT
- CraTENet (π₯5 Β· β 18 Β· π) - An attention-based deep neural network for thermoelectric transport properties. MIT transport-phenomena
- SCFNN (π₯5 Β· β 15 Β· π) - Self-consistent determination of long-range electrostatics in neural network potentials. MIT C++ electrostatics single-paper
- gkx: Green-Kubo Method in JAX (π₯5 Β· β 8 Β· π) - Green-Kubo + JAX + MLPs = Anharmonic Thermal Conductivities Done Fast. MIT transport-phenomena
- Per-site PAiNN (π₯5 Β· β 2 Β· π) - Fork of PaiNN for PerovskiteOrderingGCNNs. MIT probabilistic pretrained single-paper
- Per-Site CGCNN (π₯5 Β· β 1 Β· π) - Crystal graph convolutional neural networks for predicting material properties. MIT pretrained single-paper
- Graph Transport Network (π₯4 Β· β 15 Β· π) - Graph transport network (GTN), as proposed in Scalable Optimal Transport in High Dimensions for Graph Distances,.. Custom transport-phenomena
- atom_by_atom (π₯4 Β· β 12 Β· π) - Atom-by-atom design of metal oxide catalysts for the oxygen evolution reaction with Machine Learning. Unlicensed surface-science single-paper
- EGraFFBench (π₯4 Β· β 11 Β· π) - Unlicensed single-paper benchmarking ML-IAP
- Element encoder (π₯3 Β· β 6 Β· π) - Autoencoder neural network to compress properties of atomic species into a vector representation. GPL-3.0 single-paper
- Point Edge Transformer (π₯2) - Smooth, exact rotational symmetrization for deep learning on point clouds. CC-BY-4.0
- SphericalNet (π₯1 Β· β 3 Β· π) - Implementation of Clebsch-Gordan Networks (CGnet: https://arxiv.org/pdf/1806.09231.pdf) by GElib & cnine libraries in.. Unlicensed
Machine-learned interatomic potentials (ML-IAP) that have been trained on large, chemically and structural diverse datasets. For materials, this means e.g. datasets that include a majority of the periodic table.
π TeaNet - Universal neural network interatomic potential inspired by iterative electronic relaxations.. ML-IAP
π PreFerred Potential (PFP) - Universal neural network potential for material discovery https://doi.org/10.1038/s41467-022-30687-9. ML-IAP proprietary
FAIRChem EquiformerV2 models (π₯32 Β· β 2K Β· π) - FAIRChem implementation of Equiformer V2 (eqV2) models. MIT pretrained UIP rep-learn catalysis
- [GitHub](https://github.com/facebookresearch/fairchem) (π¨βπ» 69 Β· π 450 Β· π 570 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/FAIR-Chem/fairchem
pip install fairchem-core
FAIRChem eSEN models (π₯32 Β· β 2K Β· π) - FAIRChem implementation of Smooth Energy Network (eSEN) models arXiv:2502.12147. MIT pretrained UIP rep-learn catalysis
- [GitHub](https://github.com/facebookresearch/fairchem) (π¨βπ» 69 Β· π 450 Β· π 570 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/FAIR-Chem/fairchem
pip install fairchem-core
DPA-2 (π₯30 Β· β 1.9K) - A large atomic model as a multi-task learner https://arxiv.org/abs/2312.15492. LGPL-3.0 ML-IAP pretrained workflows datasets
- [GitHub](https://github.com/deepmodeling/deepmd-kit) (π¨βπ» 84 Β· π 600 Β· π₯ 69K Β· π¦ 47 Β· π 1K - 13% open Β· β±οΈ 08.04.2026):
git clone https://github.com/deepmodeling/deepmd-kit
conda install -c conda-forge deepmd-kit
docker pull deepmodeling/deepmd-kit
DeePMD-DPA3 (π₯30 Β· β 1.9K) - Successor of DPA-2. LGPL-3.0 ML-IAP pretrained workflows datasets
- [GitHub](https://github.com/deepmodeling/deepmd-kit) (π¨βπ» 84 Β· π 600 Β· π₯ 69K Β· π¦ 47 Β· π 1K - 13% open Β· β±οΈ 08.04.2026):
git clone https://github.com/deepmodeling/deepmd-kit
conda install -c conda-forge deepmd-kit
docker pull deepmodeling/deepmd-kit
SevenNet (π₯24 Β· β 240) - SevenNet - a graph neural network interatomic potential package supporting efficient multi-GPU parallel molecular.. GPL-3.0 ML-IAP MD pretrained
- [GitHub](https://github.com/MDIL-SNU/SevenNet) (π¨βπ» 20 Β· π 52 Β· π₯ 14K Β· π 90 - 17% open Β· β±οΈ 07.04.2026):
git clone https://github.com/MDIL-SNU/SevenNet
MACE-FOUNDATION models (π₯23 Β· β 1.1K) - MACE foundation models (MP, OMAT, mh-1). MIT ML-IAP pretrained rep-learn MD
- [GitHub](https://github.com/ACEsuit/mace-foundations) (π¨βπ» 3 Β· π 380 Β· π₯ 360K Β· π 36 - 13% open Β· β±οΈ 19.11.2025):
git clone https://github.com/ACEsuit/mace-foundations
MatterSim (π₯22 Β· β 530) - MatterSim: A deep learning atomistic model across elements, temperatures and pressures. MIT ML-IAP active-learning multimodal phase-transition pretrained
- [GitHub](https://github.com/microsoft/mattersim) (π¨βπ» 20 Β· π 77 Β· π₯ 36 Β· π 44 - 40% open Β· β±οΈ 18.03.2026):
git clone https://github.com/microsoft/mattersim
Orb Models (π₯21 Β· β 560) - ORB forcefield models from Orbital Materials. Custom ML-IAP pretrained
- [GitHub](https://github.com/orbital-materials/orb-models) (π¨βπ» 16 Β· π 78 Β· π¦ 35 Β· π 75 - 4% open Β· β±οΈ 18.03.2026):
git clone https://github.com/orbital-materials/orb-models
CHGNet (π₯20 Β· β 380) - Pretrained universal neural network potential for charge-informed atomistic modeling https://chgnet.lbl.gov. Custom ML-IAP MD pretrained electrostatics magnetism structure-relaxation
- [GitHub](https://github.com/CederGroupHub/chgnet) (π¨βπ» 13 Β· π 96 Β· π¦ 69 Β· π 77 - 5% open Β· β±οΈ 19.02.2026):
git clone https://github.com/CederGroupHub/chgnet
PET-MAD (π₯20 Β· β 200) - Universal interatomic potentials for advanced materials modeling. BSD-3 ML-IAP MD rep-learn transformer
- [GitHub](https://github.com/lab-cosmo/upet) (π¨βπ» 15 Β· π 17 Β· π₯ 46 Β· π¦ 3 Β· β±οΈ 31.03.2026):
git clone https://github.com/lab-cosmo/pet-mad
conda install -c conda-forge pet-mad
M3GNet (π₯18 Β· β 320 Β· π€) - Materials graph network with 3-body interactions featuring a DFT surrogate crystal relaxer and a state-of-the-art.. BSD-3 ML-IAP pretrained
- [GitHub](https://github.com/materialyzeai/m3gnet) (π¨βπ» 16 Β· π 74 Β· π 35 - 42% open Β· β±οΈ 07.04.2025):
git clone https://github.com/materialsvirtuallab/m3gnet
MLIP Arena Leaderboard (π₯13 Β· β 95) - [NeurIPS 25 Spotlight] Fair and transparent benchmark of machine learning interatomic potentials (MLIPs), beyond basic.. Apache-2 ML-IAP benchmarking
- [GitHub](https://github.com/atomind-ai/mlip-arena) (π¨βπ» 3 Β· π 8 Β· π¦ 2 Β· π 23 - 56% open Β· β±οΈ 27.03.2026):
git clone https://github.com/atomind-ai/mlip-arena
GRACE (π₯13 Β· β 89) - GRACE models and gracemaker (as implemented in TensorPotential package). Custom ML-IAP pretrained MD rep-learn rep-eng
- [GitHub](https://github.com/ICAMS/grace-tensorpotential) (π¨βπ» 4 Β· π 11 Β· π¦ 10 Β· π 18 - 66% open Β· β±οΈ 06.03.2026):
git clone https://github.com/ICAMS/grace-tensorpotential
EScAIP (π₯7 Β· β 60 Β· π€) - [NeurIPS 2024] Official implementation of the Efficiently Scaled Attention Interatomic Potential. MIT ML-IAP rep-learn transformer single-paper
- [GitHub](https://github.com/ASK-Berkeley/EScAIP) (π¨βπ» 2 Β· π 6 Β· π₯ 15 Β· π 9 - 22% open Β· β±οΈ 26.09.2025):
git clone https://github.com/ASK-Berkeley/EScAIP
Show 3 hidden projects...
- ffonons (π₯7 Β· β 23 Β· π) - Phonons from ML force fields. MIT benchmarking density-of-states
- CHIPS-FF (π₯6 Β· β 9 Β· π) - Evaluation of universal machine learning force-fields https://doi.org/10.1021/acsmaterialslett.5c00093. Custom benchmarking structure-optimization MD materials-discovery transport-phenomena
- Joint Multidomain Pre-Training (JMP) (π₯5 Β· β 62 Β· π) - Code for From Molecules to Materials Pre-training Large Generalizable Models for Atomic Property Prediction. CC-BY-NC-4.0 pretrained ML-IAP general-tool
Projects that focus on unsupervised, semi- or self-supervised learning for atomistic ML, such as dimensionality reduction, clustering, contrastive learning, etc.
DADApy (π₯16 Β· β 140 Β· π) - Distance-based Analysis of DAta-manifolds in python. Apache-2
- [GitHub](https://github.com/sissa-data-science/DADApy) (π¨βπ» 21 Β· π 23 Β· π 42 - 28% open Β· β±οΈ 12.02.2026):
git clone https://github.com/sissa-data-science/DADApy
Show 9 hidden projects...
- mat_discover (π₯13 Β· β 46 Β· π) - A materials discovery algorithm geared towards exploring high-performance candidates in new chemical spaces. MIT materials-discovery rep-eng HTC
- ASAP (π₯12 Β· β 150 Β· π) - ASAP is a package that can quickly analyze and visualize datasets of crystal or molecular structures. MIT
- pumml (π₯10 Β· β 37 Β· π) - Positive and Unlabeled Materials Machine Learning (pumml) is a code that uses semi-supervised machine learning to.. MIT materials-discovery
- Sketchmap (π₯8 Β· β 48 Β· π) - Suite of programs to perform non-linear dimensionality reduction -- sketch-map in particular. GPL-3.0 C++
- paper-ml-robustness-material-property (π₯5 Β· β 4 Β· π) - A critical examination of robustness and generalizability of machine learning prediction of materials properties. BSD-3 datasets single-paper
- 3D-EMGP (π₯4 Β· β 33 Β· π) - [AAAI 2023] The implementation for the paper Energy-Motivated Equivariant Pretraining for 3D Molecular Graphs. MIT pretrained rep-learn single-paper
- Coarse-Graining-Auto-encoders (π₯4 Β· β 21 Β· π) - Implementation of coarse-graining Autoencoders. Unlicensed single-paper
- KmdPlus (π₯4 Β· β 8 Β· π) - This module contains a class for treating kernel mean descriptor (KMD), and a function for generating descriptors with.. MIT
- Descriptor Embedding and Clustering for Atomisitic-environment Framework (DECAF) ( β 2) - Provides a workflow to obtain clustering of local environments in dataset of structures. Unlicensed
Projects that focus on visualization (viz.) for atomistic ML.
Crystal Toolkit (π₯26 Β· β 200) - Crystal Toolkit is a framework for building web apps for materials science and is currently powering the new Materials.. MIT
- [GitHub](https://github.com/materialsproject/crystaltoolkit) (π¨βπ» 38 Β· π 66 Β· π¦ 43 Β· π 150 - 45% open Β· β±οΈ 09.04.2026):
git clone https://github.com/materialsproject/crystaltoolkit
pip install crystal-toolkit
pymatviz (π₯23 Β· β 310) - A toolkit for visualizations in materials informatics. MIT general-tool probabilistic
- [GitHub](https://github.com/janosh/pymatviz) (π¨βπ» 14 Β· π 38 Β· π₯ 3.3K Β· π¦ 32 Β· π 65 - 1% open Β· β±οΈ 09.04.2026):
git clone https://github.com/janosh/pymatviz
Chemiscope (π₯23 Β· β 170) - An interactive structure/property explorer for materials and molecules. BSD-3 JavaScript
- [GitHub](https://github.com/lab-cosmo/chemiscope) (π¨βπ» 27 Β· π 43 Β· π₯ 650 Β· π¦ 6 Β· π 170 - 8% open Β· β±οΈ 30.03.2026):
git clone https://github.com/lab-cosmo/chemiscope
Elementari (π₯21 Β· β 320) - Interactive browser visualizations for materials science: crystal structures/molecules, trajectories, convex hulls,.. MIT JavaScript
- [GitHub](https://github.com/janosh/matterviz) (π¨βπ» 6 Β· π 30 Β· π₯ 4.2K Β· π¦ 5 Β· π 43 - 4% open Β· β±οΈ 04.04.2026):
git clone https://github.com/janosh/elementari
ZnDraw (π₯21 Β· β 49) - A powerful tool for visualizing, modifying, and analysing atomistic systems. EPL-2.0 MD generative JavaScript
- [GitHub](https://github.com/zincware/ZnDraw) (π¨βπ» 16 Β· π 5 Β· π¦ 16 Β· π 380 - 20% open Β· β±οΈ 09.04.2026):
git clone https://github.com/zincware/ZnDraw
Atomvision (π₯11 Β· β 34 Β· π€) - Deep learning framework for atomistic image data. Custom computer-vision experimental-data rep-learn
- [GitHub](https://github.com/usnistgov/atomvision) (π¨βπ» 4 Β· π 17 Β· π¦ 4 Β· π 9 - 55% open Β· β±οΈ 25.08.2025):
git clone https://github.com/usnistgov/atomvision
Projects and models that focus on quantities of wavefunction theory methods, such as Monte Carlo techniques like deep learning variational Monte Carlo (DL-VMC), quantum chemistry methods, etc.
DeepQMC (π₯17 Β· β 410) - Deep learning quantum Monte Carlo for electrons in real space. MIT
- [GitHub](https://github.com/deepqmc/deepqmc) (π¨βπ» 14 Β· π 62 Β· π¦ 3 Β· π 64 - 6% open Β· β±οΈ 08.04.2026):
git clone https://github.com/deepqmc/deepqmc
FermiNet (π₯15 Β· β 820) - An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations. Apache-2 transformer
- [GitHub](https://github.com/google-deepmind/ferminet) (π¨βπ» 24 Β· π 160 Β· π 71 - 4% open Β· β±οΈ 11.03.2026):
git clone https://github.com/google-deepmind/ferminet
DeepErwin (π₯8 Β· β 67 Β· π€) - DeepErwin is a python 3.8+ package that implements and optimizes JAX 2.x wave function models for numerical solutions.. Custom
- [GitHub](https://github.com/mdsunivie/deeperwin) (π¨βπ» 9 Β· π 8 Β· π₯ 18 Β· π¦ 2 Β· β±οΈ 18.04.2025):
git clone https://github.com/mdsunivie/deeperwin
JaQMC (π₯7 Β· β 94) - JAX accelerated Quantum Monte Carlo. Apache-2
- [GitHub](https://github.com/bytedance/jaqmc) (π¨βπ» 5 Β· π 11 Β· β±οΈ 08.04.2026):
git clone https://github.com/bytedance/jaqmc
Show 3 hidden projects...
- ACEpsi.jl (π₯7 Β· β 3 Β· π) - ACE wave function parameterizations. MIT rep-eng Julia
- LapNet (π₯5 Β· β 73 Β· π) - Efficient and Accurate Neural-Network Ansatz for Quantum Monte Carlo. Apache-2
- SchNOrb (π₯5 Β· β 69 Β· π) - Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions. MIT
Contributions are encouraged and always welcome! If you like to add or update projects, choose one of the following ways:
- Open an issue by selecting one of the provided categories from the issue page and fill in the requested information.
- Modify the projects.yaml with your additions or changes, and submit a pull request. This can also be done directly via the Github UI.
If you like to contribute to or share suggestions regarding the project metadata collection or markdown generation, please refer to the best-of-generator repository. If you like to create your own best-of list, we recommend to follow this guide.
For more information on how to add or update projects, please read the contribution guidelines. By participating in this project, you agree to abide by its Code of Conduct.
Credit by: @github.com/JuDFTteam/best-of-atomistic-machine-learning
Jax related project
Flake8 related project
Pytest related project
Sphinx related project
MkDocs related project