MLIP Training Datasets

MLIP Training Datasets

This document consolidates publicly available datasets for training and fine-tuning machine-learned interatomic potentials. Please feel free to modify, update, or add any datasets you think are missing, as well as any additional details you believe would be helpful.

Inorganic Materials

Dataset
Dataset Size
Level of Theory
Dataset Link
Reference
OMat24
100M / 1M / 5M
PBE / PBE+U
Icon for https://huggingface.co/datasets/facebook/OMAT24
facebook/OMAT24 · Datasets at Hugging Face
Icon for https://arxiv.org/abs/2410.12771
Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models
sAlex
10M
PBE / PBE+U
Icon for https://huggingface.co/datasets/facebook/OMAT24#salex-dataset
facebook/OMAT24 · Datasets at Hugging Face
Icon for https://arxiv.org/abs/2410.12771
Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models
sAlex25
1.3M
PBE
Icon for https://alexandria.icams.rub.de/
Alexandria Materials Database
Icon for https://iopscience.iop.org/article/10.1088/2515-7639/ae6620
AI-Driven expansion and application of the Alexandria database
HighP Alex
32M
PBE
Icon for https://alexandria.icams.rub.de/
Alexandria Materials Database
Icon for https://arxiv.org/html/2508.17792v1#S5
Universal Machine Learning Potentials under Pressure
WBM
215K
PBE / PBE+U
Icon for https://github.com/janosh/matbench-discovery/tree/main/data/wbm
matbench-discovery/data/wbm at main · janosh/matbench-discovery
Icon for https://www.nature.com/articles/s41524-020-00481-6
Predicting stable crystalline compounds using chemical similarity
MPtrj
1.58M
PBE/ PBE+U
Icon for https://doi.org/10.6084/m9.figshare.23713842
https://doi.org/10.6084/m9.figshare.23713842
Icon for https://www.nature.com/articles/s42256-023-00716-3
CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling
MATPES
434K
PBE
Icon for https://matpes.ai/
MatPES
Icon for https://arxiv.org/abs/2503.04070
A Foundational Potential Energy Surface Dataset for Materials
MATPES
434K
r2SCAN
Icon for https://matpes.ai/
MatPES
Icon for https://arxiv.org/abs/2503.04070
A Foundational Potential Energy Surface Dataset for Materials
MP-ALOE
1M
r2SCAN
Icon for https://doi.org/10.6084/m9.figshare.29452190
https://doi.org/10.6084/m9.figshare.29452190
Icon for https://www.nature.com/articles/s41524-025-01834-9
https://www.nature.com/articles/s41524-025-01834-9
MAD
217K
r2SCAN
Icon for https://archive.materialscloud.org/records/18tke-tt476
High-quality, high-information datasets for universal atomistic machine learning
Icon for https://arxiv.org/abs/2603.02089
High-quality, high-information datasets for universal atomistic...
SMAX
1.69M
PBE
Icon for https://arxiv.org/html/2602.23489v1
Exploring the extremes: atomic basis for multi-elemental materials science under complex thermodynamic conditions

Metal-Organic Frameworks

Dataset
Dataset Size
Level of Theory
Dataset Link
Reference
ODAC25
70M
PBE+D3(BJ)
Icon for https://huggingface.co/facebook/ODAC25
facebook/ODAC25 · Hugging Face
Icon for https://arxiv.org/abs/2508.03162v1
The Open DAC 2025 Dataset for Sorbent Discovery in Direct Air Capture
MACE-MP-MOF-0
4.7k
PBE+D3(BJ)
Icon for https://github.com/ddmms/data/tree/main/mace-mof-0
data/mace-mof-0 at main · ddmms/data
Icon for https://www.nature.com/articles/s41524-025-01611-8
Machine learned potential for high-throughput phonon calculations of metal—organic frameworks
MOF-OFF
TBA
PBE+D3(BJ)
MOF-OFF
TBA
r2SCAN+D4

Molecules

Molecular Crystals

Surfaces and Catalysts

Dataset
Dataset Size
Level of Theory
Dataset Link
Reference
OC20/22/25
A lot
Mixed
Icon for https://fair-chem.github.io/summary-3/
Datasets - FAIR Chemistry Documentation
Icon for https://fair-chem.github.io/summary-3/
Datasets - FAIR Chemistry Documentation
AQCat25
13.5M
RPBE
Icon for https://huggingface.co/datasets/SandboxAQ/aqcat25-dataset
SandboxAQ/aqcat25-dataset · Datasets at Hugging Face
Icon for https://www.nature.com/articles/s41524-026-02099-6
AQCat25: unlocking spin-aware, high-fidelity machine learning potentials for heterogeneous catalysis