Performance of universal machine learning potentials in global optimization of inorganic crystal structures

Abstract

Rapid development of universal machine learning potentials (uMLPs) and expansion of training data sets are reshaping the state of the art in atomistic simulation, highlighting the need for concurrent systematic benchmarking of their capabilities. Global optimization is among the most demanding uMLP applications because unconstrained exploration includes probing motifs not present in reference sets. We examined twelve pretrained uMLPs in unconstrained evolutionary searches to assess whether these models can consistently predict complex nonmagnetic crystal structure ground states under ambient pressure across diverse inorganic systems. Our findings demonstrate that the considered M3GNet, MACE-MATPES, MACE-mh-1, SevenNet, EquiformerV2, EquiformerV3, MatterSim, GRACE, eSEN, Orb-v3, PET-MAD, and PET-OAM models span a wide performance range, from near {\it ab initio} to essentially non-predictive, in their ability to resolve competing phases within low-energy basins. Additional tests on hcp-Zn, MB${}_4$ (M = Cr, Mn, and Fe), and LiB${}_y$ ($y\approx 0.9$) ground states reveal that several uMLPs capture fine energy differences arising from subtle electronic structure features.