Nequip-MP-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 0.255
2 He 0
3 Li 0.0512
4 Be 0.109
5 B 0.135
6 C 0.116
7 N 0.169
8 O 0.16
9 F 0.206
10 Ne 0
11 Na 0.0606
12 Mg 0.0668
13 Al 0.104
14 Si 0.113
15 P 0.11
16 S 0.135
17 Cl 0.178
18 Ar 0
19 K 0.0692
20 Ca 0.076
21 Sc 0.0752
22 Ti 0.0843
23 V 0.108
24 Cr 0.127
25 Mn 0.15
26 Fe 0.144
27 Co 0.0896
28 Ni 0.0811
29 Cu 0.0736
30 Zn 0.08
31 Ga 0.0924
32 Ge 0.101
33 As 0.0999
34 Se 0.147
35 Br 0.157
36 Kr 0
37 Rb 0.0735
38 Sr 0.0702
39 Y 0.0782
40 Zr 0.0911
41 Nb 0.109
42 Mo 0.1
43 Tc 0.0801
44 Ru 0.121
45 Rh 0.1
46 Pd 0.0846
47 Ag 0.06
48 Cd 0.0614
49 In 0.0961
50 Sn 0.0933
51 Sb 0.0895
52 Te 0.157
53 I 0.139
54 Xe 0.002
55 Cs 0.0781
56 Ba 0.0694
57 La 0.0672
58 Ce 0.0733
59 Pr 0.067
60 Nd 0.0659
61 Pm 0.0612
62 Sm 0.0673
63 Eu 0.0844
64 Gd 0.0706
65 Tb 0.069
66 Dy 0.0729
67 Ho 0.0694
68 Er 0.0711
69 Tm 0.0743
70 Yb 0.0774
71 Lu 0.0742
72 Hf 0.0908
73 Ta 0.14
74 W 0.112
75 Re 0.0947
76 Os 0.118
77 Ir 0.122
78 Pt 0.102
79 Au 0.109
80 Hg 0.0613
81 Tl 0.0663
82 Pb 0.0938
83 Bi 0.0744
84 Po 0
85 At 0
86 Rn 0
87 Fr 0
88 Ra 0
89 Ac 0.0645
90 Th 0.0962
91 Pa 0.115
92 U 0.111
93 Np 0.156
94 Pu 0.316
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

MPtrj: 1.58M structures from 146k materials

Description

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

Steps

Single training run on MPtrj with specified parameters.

Hyperparameters

  • max_force: 0.05
  • max_steps: 500
  • ase_optimizer: "GOQN"
  • cell_filter: "FrechetCellFilter"
  • optimizer: "AdamW"
  • weight_decay: 0.001
  • 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: 0.02
  • gradient_clip_val: 0.01
  • learning_rate_schedule: "ReduceLROnPlateau - factor=0.1, patience=100, min_lr=1e-6"
  • epochs: 300
  • max_neighbors: null

Dependencies