thorben-frank/mlff
Build neural networks for machine learning force fields with JAX
This tool helps computational chemists and materials scientists build and use neural network models to predict atomic forces and energies. You provide atomic structures and their known energies/forces from quantum mechanics calculations, and it trains a model. The output is a trained 'machine learned force field' that can quickly simulate molecular behavior, replacing slower first-principles calculations.
133 stars. No commits in the last 6 months.
Use this if you need to perform molecular dynamics simulations faster and more efficiently than traditional quantum mechanics methods allow, especially for large systems or long timescales.
Not ideal if you need to run molecular dynamics simulations immediately without training a custom force field, as the MD functionality is currently under active re-development.
Stars
133
Forks
37
Language
Python
License
MIT
Category
Last pushed
Jun 02, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/thorben-frank/mlff"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
explosion/thinc
🔮 A refreshing functional take on deep learning, compatible with your favorite libraries
google-deepmind/optax
Optax is a gradient processing and optimization library for JAX.
patrick-kidger/diffrax
Numerical differential equation solvers in JAX. Autodifferentiable and GPU-capable....
google/grain
Library for reading and processing ML training data.
patrick-kidger/equinox
Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/