microsoft/mattersim

MatterSim: A deep learning atomistic model across elements, temperatures and pressures.

71
/ 100
Verified

MatterSim helps materials scientists and researchers simulate the behavior of atomic structures under varying conditions. You input an atomic structure, and it predicts properties like energy, forces, and stress across different elements, temperatures, and pressures. This is useful for computational materials scientists, chemists, and physicists who need to understand and predict material characteristics without extensive lab experiments.

520 stars. Used by 3 other packages. Actively maintained with 2 commits in the last 30 days. Available on PyPI.

Use this if you need to quickly and accurately simulate the atomic-level behavior of bulk materials, like silicon crystals, to predict their physical properties.

Not ideal if your work involves surfaces, interfaces, or properties heavily influenced by long-range interactions, as the model's accuracy is reduced for these scenarios.

materials-science atomistic-simulation computational-chemistry solid-state-physics material-design
Maintenance 13 / 25
Adoption 13 / 25
Maturity 25 / 25
Community 20 / 25

How are scores calculated?

Stars

520

Forks

74

Language

Jupyter Notebook

License

MIT

Last pushed

Feb 25, 2026

Commits (30d)

2

Dependencies

24

Reverse dependents

3

Get this data via API

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

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