rankings / model / MACE-MP-0b3

MACE-MP-0b3

MACE-MP-0b3 is a machine-learning interatomic potential — version 0b3 · MIT (weights) license · 15.8M parameters. ScoreMol scores it on 14 benchmark suites covering Atomization energy, Force MAE, Non-covalent interactions, Torsion profile MAE, Reaction barriers, Formation energy, Lattice parameters, Polymorph ranking (ρ), Stable-set discovery (F1), Band gap (GW ref), Band gap (experimental ref), Γ-point frequency error, False imaginary-mode rate, Missed-instability rate, Elastic tensor Cij, Adsorption energy, Ionic conductivity, RDF / structure factor, NVE energy drift, MD stability 1000 K, Liquid density, Self-diffusion, OOD degradation, Reaction energy, each run under the same protocol on held-out reference data with cluster-bootstrap confidence intervals.

Suites: QM9-X, NCI-800, TORS-500, RXN-1k, FormE-40k, Hull-5k, Gap-2k, Phon-1k, Elas-600, Adsorb-3k, MD-Electro-800, Liq-300, OOD-X, MTK-55.