apax-hub/apax

A flexible and performant framework for training machine learning potentials.

58
/ 100
Established

This project helps computational chemists and materials scientists train and use machine learning models to predict how atoms and molecules interact. You provide atomic structure data, and it generates a specialized machine learning potential that can then be used to simulate molecular dynamics, offering a faster way to explore material properties or chemical reactions. It is designed for researchers working on atomistic simulations.

Available on PyPI.

Use this if you need to quickly train and deploy high-performance machine learning models to simulate the behavior of atoms and molecules.

Not ideal if you are looking for a general-purpose machine learning framework outside of atomistic simulations or need to implement a custom potential that is not neural network-based.

computational-chemistry materials-science molecular-dynamics atomistic-simulation machine-learning-potentials
Maintenance 10 / 25
Adoption 7 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

36

Forks

7

Language

Python

License

MIT

Last pushed

Mar 13, 2026

Commits (30d)

0

Dependencies

16

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/apax-hub/apax"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.