UMA
MetaUMA is Meta FAIR's family of Universal Models for Atoms, trained on roughly half a billion DFT structures compiled from five chemical-domain datasets and routed at inference by a task embedding covering molecules, inorganic materials, catalysis, MOFs, and molecular crystals. Weights are distributed through a gated Hugging Face repository and run through the fairchem library's ASE calculator and multi-GPU simulation interfaces.
Access is gated on manual approval and requires registering a full legal name and organization, which is a conduct and identity condition rather than a commercial bound. Verified 2026-09-10 via the Hugging Face API, the fairchem README, LICENSE and UMA documentation, and the UMA paper.
Openness
4 high confidence- weights
- open(manually gated but downloadable uma-s and uma-m checkpoint files)
- data
- open(OMol25, OMat24, OC20, ODAC23, and OMC25 training datasets published alongside the model)
- code
- open(configs/uma/training_release ships uma_sm_direct_pretrain.yaml and uma_md_direct_pretrain.yaml with their backbone, dataset, tasks and element_refs configs, driven by the fairchem CLI, alongside the MIT inference and calculator library)
- license
- FAIR-Chemistry-License-v1(royalty-free commercial use with an acceptable-use policy and gated registration)
Gating and an acceptable-use policy are identity and conduct conditions rather than a commercial ceiling, so the licence does not push UMA below the open-weights centre of gravity. The pretraining configs in configs/uma/training_release and the five published training datasets are what carry it above the open-weights centre of gravity.
- https://api.github.com/repos/facebookresearch/fairchem/contents/configs/uma/training_release recorded 2026-09-10
Lists uma_sm_direct_pretrain.yaml and uma_md_direct_pretrain.yaml beside uma_sm_conserve_finetune.yaml and uma_md_conserve_finetune.yaml, with backbone, dataset, tasks, element_refs and cluster config directories - the published pretraining recipe, not just inference code.
- https://huggingface.co/api/models/facebook/UMA recorded 2026-09-10
gated: manual; private: false; license: other under the FAIR Chemistry License v1 with a full-legal-name registration prompt; checkpoints/uma-s-1p2.pt, uma-s-1p2p1.pt, and uma-m-1p1.pt present.
- https://raw.githubusercontent.com/facebookresearch/fairchem/main/docs/core/uma.md recorded 2026-09-10
Names the five DFT training datasets behind the tasks — OMol25, OMC25, OMat24, OC20, and ODAC23, plus OC25 and OC22 in UMA-1.2 — each with a published reference.
- https://raw.githubusercontent.com/facebookresearch/fairchem/main/LICENSE.md recorded 2026-09-10
MIT License body for the fairchem library, read directly because the license endpoint reports NOASSERTION; the README adds that model and checkpoint licenses vary by application area.
- https://raw.githubusercontent.com/facebookresearch/fairchem/main/README.md recorded 2026-09-10
Documents installation, pretrained-model instantiation, ASE calculators, molecular dynamics, and multi-GPU and LAMMPS interfaces, with no pretraining or fine-tuning recipe.
Adoption
1 medium confidence83 downloads in the trailing 30 days for the declared Hub repository, with manual gating suppressing the count. fairchem-core is also declared and draws far more, but it is not banded here for a reason rather than by omission: the package is the library that serves every fairchem model, so its download population is not UMA's population, and docs/reference/adoption.md requires a package to be the product's own primary channel before it may band the head product. The Hub repository is UMA's own channel; the gate is what suppresses it.
- https://huggingface.co/api/models/facebook/UMA recorded 2026-09-10
downloads: 83; likes: 374; gated: manual.
Capability
3 high confidenceTwo attributes. The five tasks are separate data systems with different functionals and pseudopotentials rather than variables of one, which is what makes this multi-instrument. Adaptation is documented directly - docs/core/common_tasks/fine_tuning.md is a first-party UMA fine-tuning page shipping create_uma_finetune_dataset.py and a templated fine-tuning config, and configs/uma/training_release carries the conserve_finetune YAMLs beside it. The publisher's headline claim is that fine-tuning is usually unnecessary, which is a claim about convenience rather than about whether a pathway exists; the pathway is documented and is counted.
- https://raw.githubusercontent.com/facebookresearch/fairchem/main/docs/core/uma.md recorded 2026-09-10
States UMA is trained on five different DFT datasets with different levels of theory, learns a per-task embedding, and requires the user to select a task at inference; tabulates the functional and pseudopotentials for each.
- https://arxiv.org/abs/2506.23971 recorded 2026-09-10
Describes training on half a billion 3D atomic structures compiled across multiple chemical domains and reports that a single model without any fine-tuning matches or beats specialized models.
Verified 2026-09-10