Broadening access to Skala creates a faster path to predictive DFT
AI Summary: Microsoft Research has released Skala 1.1, an updated deep-learning density functional theory (DFT) approach that demonstrates improved accuracy across key molecular simulation challenges, including thermochemistry, reaction kinetics, and molecular structure prediction. Trained on 2.5 times more data than its predecessor, Skala 1.1 outperforms previous functionals, ranking first in 32 of 55 categories of the GMTKN55 benchmark. Skala 1.1 is now available in CP2K and is being integrated into several other electronic-structure software packages, including Psi4, FHI-aims, ORCA, and VASP. A new living benchmark has also been introduced to track the computational performance of successive Skala releases.