Matlantis is an atomic-level simulator based on the cutting-edge Neural Network Potential (NNP), incorporating deep learning technology. It addresses the computational cost challenges of traditional physics-based atomic simulations by directly learning the energies and forces related to atomic behavior using deep learning. Currently supporting 72 elements, it is capable of simulating various atomic combinations, such as molecules and crystals, and has a speed more than 10,000 times faster than conventional methods using DFT. Additionally, its browser-based operation allows users to effortlessly utilize pre-trained models and libraries, simplifying material exploration simulations.


Official site
Openness ☆☆☆

Commercial License

Core Developers

Preferred Computational Chemistry