Sovereign 8s.is mirror of standardgalactic/SevenNet
  • Python 70.6%
  • Cuda 20.8%
  • C++ 7.8%
  • Shell 0.8%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Yutack Park dd71896c37
Some checks failed
Linting / prek (push) Has been cancelled
bump version (#290)
2026-03-03 14:04:21 +09:00
.github/workflows fea: use assert close rather than equal 2026-02-21 11:49:12 -05:00
docs Remove StateDict from TS API (#289) 2026-03-03 14:01:34 +09:00
example_inputs chore 2025-01-29 23:31:06 +09:00
sevenn Remove StateDict from TS API (#289) 2026-03-03 14:01:34 +09:00
tests patch single-atom configuration error when flashTP used (#285) 2026-02-25 22:33:05 +09:00
.gitattributes chore: add .gitattributes 2024-09-08 15:09:35 +09:00
.gitignore fix (#288) 2026-02-27 17:49:42 +09:00
.gitmessage chore: Reforamt code by pre-commit, add .gitmessage & .pre-commit-config.yaml & setup.cfg 2024-01-05 17:44:31 +09:00
.pre-commit-config.yaml Merge remote-tracking branch 'origin/main' into CompRhys/main 2026-02-24 16:17:59 +09:00
.readthedocs.yaml [pre-commit.ci] auto fixes from pre-commit.com hooks 2025-11-28 02:02:10 +00:00
CHANGELOG.md bump version (#290) 2026-03-03 14:04:21 +09:00
LICENSE license to mit 2025-06-20 15:12:41 +09:00
pyproject.toml bump version (#290) 2026-03-03 14:04:21 +09:00
README.md Update README for documentation links and clarity 2025-12-14 12:03:39 +09:00
setup.cfg refactor: change url to git 2025-03-16 15:15:04 +09:00
SevenNet_logo.png chore: Prepare for arXiv publication 2024-02-06 17:42:37 +09:00

Alt text

SevenNet

SevenNet (Scalable EquiVariance-Enabled Neural Network) is a graph neural network (GNN)-based interatomic potential package that supports parallel molecular dynamics simulations using LAMMPS. Its core model is based on NequIP.

Full documentation, including installation, usage, and pretrained models, is available at documentation.

Features

  • Pretrained GNN interatomic potential and fine-tuning interface
  • Atomic Simulation Environment (ASE) calculator (python)
  • GPU-parallelized molecular dynamics with LAMMPS
  • CUDA-accelerated D3 (van der Waals) dispersion
  • Multi-fidelity training for combining multiple databases with different calculation settings
  • Tensor product accelerators

Pretrained Models

SevenNet provides pretrained models (universal potentials). Please refer to the documentation for available checkpoints, and usage examples: Pretrained models

Installation and user guides

Installation (including LAMMPS and D3) and user guides can be found in our documentation.

The old README (prior to v0.12.0) can be found here.

Citation

If you use this code, please cite:

@article{park_scalable_2024,
	title = {Scalable Parallel Algorithm for Graph Neural Network Interatomic Potentials in Molecular Dynamics Simulations},
	volume = {20},
	doi = {10.1021/acs.jctc.4c00190},
	number = {11},
	journal = {J. Chem. Theory Comput.},
	author = {Park, Yutack and Kim, Jaesun and Hwang, Seungwoo and Han, Seungwu},
	year = {2024},
	pages = {4857--4868},
}

If you utilize the multi-fidelity feature of this code or the pretrained model SevenNet-MF-ompa, please cite the following paper:

@article{kim_sevennet_mf_2024,
	title = {Data-Efficient Multifidelity Training for High-Fidelity Machine Learning Interatomic Potentials},
	volume = {147},
	doi = {10.1021/jacs.4c14455},
	number = {1},
	journal = {J. Am. Chem. Soc.},
	author = {Kim, Jaesun and Kim, Jisu and Kim, Jaehoon and Lee, Jiho and Park, Yutack and Kang, Youngho and Han, Seungwu},
	year = {2024},
	pages = {1042--1054},
}

If you utilize the pretrained model SevenNet-Omni or multi-task training strategies including task-specific regularization and domain-bridging dataset, please cite the following paper:

@article{kim_optimizing_2025,
	title = {Optimizing Cross-Domain Transfer for Universal Machine Learning Interatomic Potentials},
	doi = {10.48550/arxiv.2510.11241},
	journal = {arXiv},
	author = {Kim, Jaesun and You, Jinmu and Park, Yutack and Lim, Yunsung and Kang, Yujin and Kim, Jisu and Jeon, Haekwan and Ju, Suyeon and Hong, Deokgi and Lee, Seung Yul and Choi, Saerom and Kim, Yongdeok and Lee, Jae W and Han, Seungwu},
	year = {2025},
}