AdvanceSoft Corporation has released a new product, Advance/OF-DFT. This innovative software implements a groundbreaking method in Density Functional Theory (DFT)—the standard computational approach for electronic structure simulations in materials science. It achieves both higher accuracy than conventional Orbital-Free DFT (OF-DFT) and significantly faster performance that surpasses Kohn-Sham DFT (KS-DFT).
While conventional OF-DFT directly optimizes electron density without using orbitals to enable high-speed simulations at low computational costs, it has long faced a major challenge: the kinetic energy functional required for practical accuracy has remained unknown. AdvanceSoft has resolved this issue with “AdvanceSoft26,” a patented deep learning kinetic energy functional developed exclusively by AdvanceSoft. Equipped with “AdvanceSoft26,” Advance/OF-DFT enables electronic structure simulations that are far more accurate than traditional OF-DFT and overwhelmingly faster than KS-DFT.
By utilizing Advance/OF-DFT, users can simulate electronic structures that cannot be handled even by machine learning force fields, which have seen remarkable development in recent years. Furthermore, because its computational cost scales as O(N), it facilitates the easy analysis of large-scale systems that are beyond the reach of KS-DFT.
Please refer to the links below for product details and documentation.
■ Advance/OF-DFT Product Introduction Page
■ Advance/OF-DFT Online Documentation
