UMA (FAIRChem)¶
UMA checkpoints from Meta FAIR are loaded through the upstream fairchem-core
stack. uma and fairchem (the upstream package name) are the same runtime, so
either works as --model-type.
What the model is¶
- Architecture. UMA is an eSEN-type equivariant network with a 6 Å cutoff whose linear layers are a mixture of linear experts: the expert weights are chosen once per system from its composition, charge, spin and task, so a large total parameter count costs only a small active one at inference (Wood et al., 2025).
- Training data. About 500 million structures from OMat24 (bulk inorganic, PBE / PBE+U), OMol25 (molecules, ωB97M-V), OC20 (catalysis surfaces), ODAC25 (metal-organic frameworks) and OMC25 (molecular crystals). The datasets use different DFT settings, which is why you choose a task: each task reproduces its own dataset's reference level, and energies from different tasks must not be compared.
- Forces. The UMA paper reports UMA-S and UMA-M as conservative (forces and
stress by differentiation, after a direct-force pretraining stage). mliport does
not rely on that statement: NEB and the configuration search read the
predictor's
direct_forcesflag and measure energy–force consistency at run time (see below). - What this means for you. UMA is the one example backend trained on surfaces,
molecules and porous materials as well as bulk, so for an adsorbate on a slab,
compare the
omatandoc20tasks against your own DFT before choosing. Its license is non-commercial research only.
Install¶
| Environment | Python | Runtime stack |
|---|---|---|
.venv |
3.11, 3.12, 3.13 | fairchem-core==2.25.0 (torch 2.13.0) |
Pinned profiles¶
| Profile | Identity | dtype | task | head | model SHA-256 |
|---|---|---|---|---|---|
uma_omat |
UMA-s 1.2 (OMat task head) | upstream/model-defined | omat | None | ba5c0d912efa... |
Run it¶
Get the checkpoint from the upstream fairchem/UMA release channels and verify its SHA-256 before use. mliport never redistributes model weights, and UMA's license is non-commercial research only; see Citations.
Things to know¶
--taskpicks the task family (omatfor periodic bulk,omolfor molecules,oc20for catalysis). UMA has no separate head field. The families are not interchangeable: run a molecule with anomoltask and a periodic crystal with anomattask.- Inference modes.
--inference-modeacceptsdefaultorturbo. Turbo (TF32 plustorch.compile) is opt-in and MD does not use it by default, so precision-sensitive runs stay in the default mode.--activation-checkpointingtrades speed for GPU memory. - Precision is defined by the model; mliport reports the framework version rather than pretending to control the weights.
- Runtime pin.
fairchem-core==2.25.0on torch 2.13.0. Volta (V100) stays on thecu126wheels, Turing and newer usecu130, because the CUDA 13 builds carry no kernel below Turing and need an NVIDIA driver of 580 or newer; thecu126wheels work with older drivers. 2.25.0 is the pinned release that loadsuma-s-1p2p1(UMA-S 1.2.1). The beta scientific validation ofuma_omatran onfairchem-core==2.21.0with torch 2.8.0; it describes that runtime and is not carried over to this one, which has its own callable check. - No Nitro. fairchem-core 2.22 and later made UMA Nitro the upstream
defaultpreset and removedturbo; Nitro needs Ampere or newer and runs at reduced precision. mliport keeps its own two names and maps them onto upstream presets that carry the numbers it has always documented:defaultruns upstreambatch(tf32 off, no compile, no merged MOLE) andturboruns upstreammultigpu(upstream calls this "turbo's former settings": tf32, compile, merged MOLE). A preset that would enable Nitro is refused, and the preset that really ran, its tf32 flag and its Nitro flag are written to the run record. - NEB and configuration search need forces that are the gradient of the
energy. mliport decides this at run time, not from this page: a predictor that
reports
direct_forces = trueis refused, otherwise a finite-difference pre-flight measures the consistency on your structures (see NEB). - Stress is reported only if the checkpoint exposes it.
mliport is MIT-licensed. The upstream software and model must be cited separately.