Nequip-OAM-L

Version: 0.1 Added: 2025-09-08 Published: 2025-08-28 9.6M parameters Missing preds: 0
Repo Paper Docs DOI Files PR Checkpoint
Predictions
κSRMEGeometry OptimizationDiscovery

Convex hull distance prediction errors projected onto elements

1 H 0.136
2 He 0
3 Li 0.0262
4 Be 0.0246
5 B 0.0593
6 C 0.0533
7 N 0.0735
8 O 0.111
9 F 0.109
10 Ne 0
11 Na 0.0272
12 Mg 0.0352
13 Al 0.0452
14 Si 0.0588
15 P 0.0516
16 S 0.067
17 Cl 0.0865
18 Ar 0
19 K 0.0317
20 Ca 0.0353
21 Sc 0.0301
22 Ti 0.0382
23 V 0.0592
24 Cr 0.0828
25 Mn 0.107
26 Fe 0.0893
27 Co 0.047
28 Ni 0.0416
29 Cu 0.0355
30 Zn 0.0365
31 Ga 0.0446
32 Ge 0.0488
33 As 0.0496
34 Se 0.0866
35 Br 0.0735
36 Kr 0
37 Rb 0.0326
38 Sr 0.0325
39 Y 0.038
40 Zr 0.0385
41 Nb 0.0484
42 Mo 0.0535
43 Tc 0.0393
44 Ru 0.0521
45 Rh 0.0471
46 Pd 0.0439
47 Ag 0.0328
48 Cd 0.0285
49 In 0.0501
50 Sn 0.0427
51 Sb 0.054
52 Te 0.111
53 I 0.0593
54 Xe 0.074
55 Cs 0.0308
56 Ba 0.0335
57 La 0.0321
58 Ce 0.0324
59 Pr 0.0329
60 Nd 0.0301
61 Pm 0.0321
62 Sm 0.0313
63 Eu 0.0597
64 Gd 0.0418
65 Tb 0.0306
66 Dy 0.0332
67 Ho 0.0299
68 Er 0.0302
69 Tm 0.0304
70 Yb 0.0543
71 Lu 0.0313
72 Hf 0.0384
73 Ta 0.07
74 W 0.0448
75 Re 0.0444
76 Os 0.052
77 Ir 0.0583
78 Pt 0.0544
79 Au 0.056
80 Hg 0.0309
81 Tl 0.0322
82 Pb 0.0519
83 Bi 0.0438
84 Po 0
85 At 0
86 Rn 0
87 Fr 0
88 Ra 0
89 Ac 0.0288
90 Th 0.0417
91 Pa 0.0466
92 U 0.0551
93 Np 0.0918
94 Pu 0.198
95 Am 0
96 Cm 0
97 Bk 0
98 Cf 0
99 Es 0
100 Fm 0
101 Md 0
102 No 0
103 Lr 0
104 Rf 0
105 Db 0
106 Sg 0
107 Bh 0
108 Hs 0
109 Mt 0
110 Ds 0
111 Rg 0
112 Cn 0
113 Nh 0
114 Fl 0
115 Mc 0
116 Lv 0
117 Ts 0
118 Og 0
57-71 La-Lu Lanthanides
89-103 Ac-Lr Actinides

Model Authors

  1. Seán R. Kavanagh  Center for the Environment, Harvard University & MIR Group, Harvard University    
  2. Chuin Wei Tan  MIR Group, Harvard University  MIR Group, Harvard University logo
  3. Albert Musaelian  MIR Group, Harvard University & Mirian Technologies
  4. William C. Witt  MIR Group, Harvard University  MIR Group, Harvard University logo
  5. Gabriel de Miranda Nascimento  MIR Group, Harvard University & MIT
  6. Ulrik Unneberg  MIR Group, Harvard University & MIT
  7. Marc L. Descoteaux  MIR Group, Harvard University  MIR Group, Harvard University logo
  8. Boris Kozinsky  MIR Group, Harvard University  MIR Group, Harvard University logo

Trained By

  1. Seán R. Kavanagh (Center for the Environment, Harvard University & MIR Group, Harvard University)

Model Info

  • Model Version 0.1
  • Model Type UIP
  • Targets EFSG
  • Openness OSOD
  • Train Task S2EFS
  • Test Task IS2RE-SR
  • Trained for Benchmark Yes

Training Set

OMat24: 101M structures from 3.23M materials

Subsampled Alexandria: 10.4M structures from 3.23M materials

MPtrj: 1.58M structures from 146k materials

Description

Large NequIP foundation potential; see https://www.nequip.net/models/mir-group/NequIP-OAM-L:0.1 for details and https://arxiv.org/abs/2504.16068 for model/training infrastructure.

Steps

Training performed by: (1) pre-training on OMat24; (2) fine-tuning on MPtrj+sAlex, with a reduced learning rate (1e-4), energy-loss-upweighting (1:1:0.01 instead of 1:5:0.01) and StochasticWeightAveraging (SWA).

Hyperparameters

  • max_force: 0.05
  • max_steps: 500
  • ase_optimizer: "GOQN"
  • cell_filter: "FrechetCellFilter"
  • optimizer: "AdamW"
  • weight_decay: 1e-8
  • graph_construction_radius: 6
  • sph_harmonics_l_max: 3
  • n_layers: 6
  • n_features: "128 (l=0 scalars), 64 (l=1 vectors), 32 (l=2,3 tensors)"
  • parity: false
  • zbl_potential: true
  • type_embed_num_features: 48
  • polynomial_cutoff: 5
  • n_radial_bessel_basis: 8
  • loss: "Huber - delta=0.01 for energy, delta=0.1 for stress, stratified delta (0.01, 0.007, 0.004, 0.001) for force"
  • loss_weights: {"energy":1,"force":5,"stress":0.01}
  • batch_size: 640
  • initial_learning_rate: 0.005
  • gradient_clip_val: 1
  • learning_rate_schedule: "ReduceLROnPlateau - factor=0.1, patience=10, min_lr=1e-6"
  • epochs: 30
  • max_neighbors: null

Dependencies