Open Molecules 2025 (OMol25) Neutral Training Set, ColabFit Conversion

Provider
Meta FAIR Chemistry, Lawrence Berkeley National Laboratory and collaborators (Levine et al.); ColabFit conversion
License
CC-BY-4.0 (source)
Size
347 files
Format
Parquet (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/colabfit/OMol25_train_neutral)
Access
Free and public. No account with the data provider is needed.
DOI
10.60732/3c2ddc75
Category
Chemistry & Materials
Source
https://doi.org/10.48550/arXiv.2505.08762

About Open Molecules 2025 (OMol25) Neutral Training Set, ColabFit Conversion

The neutral training split of Open Molecules 2025 (OMol25), the molecular DFT dataset from Meta FAIR Chemistry and Lawrence Berkeley National Laboratory computed at wB97M-V/def2-TZVPD in ORCA. It keeps only the charge-neutral systems, over main-group elements, which makes it the natural training set for potentials that take no charge or spin input, such as MACE-OFF style models, and for comparisons against SPICE or ANI data. ColabFit counts 34,335,828 configurations and about 930 million atoms over 17 elements (H, B, C, N, O, F, Si, P, S, Cl, Br, I, Li, Na, K, Mg, Ca), each with total energy and atomic forces. This is the ColabFit Exchange conversion: 340-odd Parquet files under co/ hold one configuration per row (positions, species, computed properties and metadata) and ds.parquet holds dataset-level information; read them with pyarrow, polars or Hugging Face datasets following the ColabFit schema at materials.colabfit.org/docs. For the original ASE-LMDB shards that fairchem reads, use Meta's neutral_train archive instead. Meta licenses OMol25 under CC-BY-4.0; cite Levine et al., arXiv:2505.08762 (2025), and the ColabFit record doi:10.60732/3c2ddc75.

How to load Open Molecules 2025 (OMol25) Neutral Training Set, ColabFit Conversion in Python

from datasets import load_dataset
ds = load_dataset("colabfit/OMol25_train_neutral")
print(ds)

Data structure

File types in sampled files: Parquet

Files

Example file: co/co_0.parquet

Analyze this dataset without downloading it

On Scigantic, Open Molecules 2025 (OMol25) Neutral Training Set, ColabFit Conversion opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Chemistry & Materials datasets