Garden's Almanac of Matter Models

Model availability by cluster

These machine-learned interatomic potentials are pre-installed on shared HPC clusters. Rootstock lets you run them all through the same drop-in ASE calculator.

Polaris
ALCF
GPU types
NVIDIA A100 40GB
0 verified 40 installed // 22 not installed
Sophia
ALCF
GPU types
NVIDIA A100 40GB
37 verified 6 installed // 19 not installed
Perlmutter
NERSC
GPU types
NVIDIA A100 40GB
39 verified 20 installed // 3 not installed
Delta
NCSA
GPU types
NVIDIA A100 40GB
47 verified 4 installed // 11 not installed
Frontier
OLCF
GPU types
AMD MI250X
15 verified 0 installed // 47 not installed
Della
Princeton
GPU types
NVIDIA A100 40GB
28 verified 1 installed // 33 not installed
verified — ran in last 30 days installed — not recently verified not installed — not on this cluster
Model checkpoint Polaris · ALCF
NVIDIA A100 40GB
Sophia · ALCF
NVIDIA A100 40GB
Perlmutter · NERSC
NVIDIA A100 40GB
Delta · NCSA
NVIDIA A100 40GB
Frontier · OLCF
AMD MI250X
Della · Princeton
NVIDIA A100 40GB
allscaip-md-conserving-all-omol
allscaip-md-direct-all-omol
allscaip:custom
ANI
ani-2x
ani-1ccx
ani-1x
chgnet-default
chgnet:custom
dimenet-plus-plus-s2ef-oc20-all
dimenet-plus-plus-s2ef-oc20-20m
dimenet-plus-plus-s2ef-oc20-2m
dimenet-plus-plus-s2ef-oc20-200k
equiformer-v2-153m-s2ef-oc20-all-md
equiformer-v2-31m-s2ef-oc20-all-md
equiformer-v2-83m-s2ef-oc20-2m
eqv2-31m-mp
eqv2-31m-omat
eqv2-86m-omat
eqv2-153m-omat
eqv2-31m-omat-mp-salex
eqv2-86m-omat-mp-salex
eqv2-153m-omat-mp-salex
eqv2-dens-31m-mp
eqv2-dens-86m-mp
eqv2-dens-153m-mp
escn-l6-m2-lay12-s2ef-oc20-all-md
escn-l6-m3-lay20-s2ef-oc20-all-md
escn-l6-m2-lay12-s2ef-oc20-2m
escn-l4-m2-lay12-s2ef-oc20-2m
esen-30m-mptrj
esen-30m-oam
esen-30m-omat
esen-md-direct-all-omol
esen-sm-conserving-all-omol
esen-sm-direct-all-omol
esen:custom
gemnet-oc-large-s2ef-oc20-all-md
gemnet-oc-s2ef-oc20-all-md
gemnet-oc-s2ef-oc20-all
gemnet-dt-s2ef-oc20-all
grace-2l-smax-omat-large
grace-3l-omat-large-ft-am
mace-matpes-r2scan-0
mace:custom
mace-mh-1
mace:custom
mace-mp-0-small
mace-mp-0-medium
mace-mp-0-large
mace:custom
mace-mpa-0-medium
mace:custom
mace-off23-small
mace-off23-medium
mace-off23-large
mace:custom
mace-omol-0-extra-large
mace:custom
mace-polar-1-s
mace-polar-1-m
mace-polar-1-l
mace-polar:custom
mattersim-v1-0-0-5m
mattersim-v1-0-0-1m
mattersim:custom
orb-v2
orb-d3-v2
orb-mptraj-only-v2
orb-v3-conservative-20-omat
orb-v3-conservative-20-mpa
orb-v3-conservative-inf-omat
orb-v3-conservative-inf-mpa
orb-v3-direct-20-omat
orb-v3-direct-20-mpa
orb-v3-direct-inf-omat
orb-v3-direct-inf-mpa
orb-v3-conservative-omol
orb-v3-direct-omol
painn-s2ef-oc20-all
PET
pet-oam-xl
pet-omatpes-l
schnet-s2ef-oc20-all
schnet-s2ef-oc20-20m
schnet-s2ef-oc20-2m
schnet-s2ef-oc20-200k
SCN
scn-s2ef-oc20-all-md
scn-t4-b2-s2ef-oc20-2m
scn-s2ef-oc20-2m
tece-oam-rra-1.0
tensornet-matpes-pbe-2025-2
torchmdnet-et-oc20
UMA task heads omat ·omol ·oc20 ·odac ·omc
uma-s-1p1
uma-s-1p2
uma-m-1p1
uma:custom

Running a model

Cluster admins pre-install each MLIP with its own dependencies. You just install the lightweight rootstock package in your working environment and pick a supported checkpoint. Behind the ASE calculator, Rootstock starts the model as a subprocess in the model’s pre-installed environment. That lets you switch MLIPs without rebuilding your project dependencies.

# In a Perlmutter job/session
from ase.build import bulk
from rootstock import RootstockCalculator

atoms = bulk("Cu", "fcc", a=3.6) * (5, 5, 5)

with RootstockCalculator(
    cluster="perlmutter",
    checkpoint="mace-mp-0-medium",
    device="cuda",
) as calc:
    atoms.calc = calc
    print(atoms.get_potential_energy())