MatterSim 1M
MatterSim 1M parameter foundation model for atomistic simulations. A deep learning interatomic potential trained across elements, temperatures, and pressures. Lightweight and fast for rapid screening applications.
| Version | 1.0.0 |
|---|---|
| Creator | Microsoft Research |
| Delivery | executable |
| Requirements | ram: 4GB; disk: 300MB |
Keywords
materials-scienceatomistic-simulationinteratomic-potentialmachine-learningfoundation-modelfast-inferencelightweight
Citation
Yang, H., et al. (2024). MatterSim: A Deep Learning Atomistic Model Across Elements, Temperatures and Pressures. arXiv:2405.04967