All Folmsbee Results

The Folmsbee conformers benchmark (GitHub) is a high-accuracy molecular benchmark that evaluates the accuracy of predicting and ranking conformer relative energies. The benchmark contains 708 subsets of 10 conformers each, 632 of which have energies computed at the DLPNO-CCSD(T) level of theory. DLPNO-CCSD(T) calculations were not finished for 76 subsets, so every level of theory will show at least 76 incomplete subsets. Mean absolute error (MAE) and root mean square error (RMSE) are shown in kcal/mol. See all Folmsbee results.

Exclude non-neutral subsets?

Name						Benchmarked By
OMol25's eSEN Conserving Small	0.21	0.33	0.85	0.86	76
UMA Small (OMol task)	0.21	0.34	0.85	0.85	76
ωB97X-D/def2-TZVP	0.24	0.39	0.83	0.85	80	link
B3LYP/def2-TZVP	0.25	0.41	0.83	0.84	76	link
B97-3c	0.30	0.53	0.80	0.82	77	link
Egret-1e	0.31	0.50	0.78	0.81	115
Egret-1	0.31	0.53	0.78	0.81	115
MACE-ANI-CC	0.32	0.55	0.77	0.79	437
Egret-1t	0.34	0.59	0.76	0.79	115
AIMNet2 (ωB97M-D3, new)	0.54	0.93	0.64	0.65	76
AIMNet2 (ωB97M-D3, old)	0.54	0.93	0.64	0.65	76
Prescient's StrainRelief MACE	0.57	1.02	0.60	0.63	115
GFN2-xTB	0.72	1.32	0.57	0.59	76	link
eSEN-OAM	0.84	1.71	0.56	0.55	115
Orb-v3 (Conservative Inf. OMat)	0.88	1.71	0.51	0.50	115
MACE-MP-0b2(Large)-D3BJ	0.81	1.45	0.49	0.47	115
Orb-v3 (Conservative 20 OMat)	1.05	2.11	0.49	0.44	115
MACE-MP-0(Large)-D3BJ	0.88	1.59	0.49	0.45	115
MACE-MP-0b2(Medium)	0.97	1.83	0.47	0.44	115
OMat24 eqV2-L	0.87	1.44	0.47	0.50	115
MACE-MP-0b2(Small)	1.03	2.01	0.46	0.41	115
MACE-MP-0b2(Large)	1.08	2.06	0.46	0.37	115
MACE-MP-0b(Medium)	1.02	1.89	0.46	0.43	115
MACE-MP-0b(Small)	1.25	2.58	0.45	0.40	115
MACE-MP-0(Large)	1.16	2.22	0.45	0.34	115
MACE-MP-0(Medium)	1.08	1.96	0.44	0.37	115
Orb-v3 (Direct Inf. OMat)	1.15	2.06	0.44	0.41	115
Orb-d3-v2	0.94	1.56	0.43	0.41	115
MACE-MP-0(Small)	1.25	2.42	0.43	0.34	115
Orb-v2	1.17	2.15	0.42	0.36	115
Orb-v3 (Direct 20 OMat)	1.30	2.33	0.40	0.36	115
SO3LR	2.65	4.85	0.28	0.18	149