Alex-MP-20 Training Set for Crystal Generative Models (MatterGen), ColabFit Conversion
- Provider
- Microsoft Research AI for Science (Zeni et al.), from Alexandria and Materials Project; ColabFit conversion
- License
- CC-BY-4.0 (source)
- Size
- 5 files
- Format
- Parquet (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/colabfit/Alex_MP-20_train)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Chemistry & Materials
- Source
- https://doi.org/10.1038/s41586-025-08628-5
About this dataset
The training split of Alex-MP-20, the crystal structure dataset Microsoft Research used to train the MatterGen generative model for inorganic materials (Zeni et al., Nature 2025). It merges structures from the Alexandria database and MP-20 (Materials Project), relaxed with PBE DFT for consistent energies; structures containing Tc, Pm or elements with atomic number 84 or above were excluded, and the training set keeps only structures with at most 20 atoms per unit cell and energy above hull up to 0.1 eV/atom. Labels include energy above hull and band gap. ColabFit's card counts 540,162 configurations; the repository is about 0.9 GB (measured 2026-11-14). The validation and test splits are separate ColabFit repositories. This is the ColabFit Exchange conversion: configurations are Parquet files under co/ (structure, positions, cell, computed properties and metadata per row) with dataset-level information in ds.parquet; read them with pyarrow, pandas, polars or Hugging Face datasets, following the ColabFit configuration schema at materials.colabfit.org/docs. The MatterGen data release states that Alexandria and MP-20 are both CC-BY-4.0.
How to load this dataset in Python
from datasets import load_dataset
ds = load_dataset("colabfit/Alex_MP-20_train")
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, Alex-MP-20 Training Set for Crystal Generative Models (MatterGen), 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
- MSR-ACC/TAE25: 73k CCSD(T)/CBS Total Atomization Energies
- LOTUS natural product occurrences, Wikidata frozen export 2026-04-13: 674,454 structure, organism and reference triples
- Li, Na and K Intercalation in Disordered and Hard Carbon Anodes: DFT Structures (Huang et al. 2019, Cambridge Apollo), ColabFit Conversion
- Spectrophotometry instrument output with ASM: 28 exports from Thermo Fisher GENESYS 30, Thermo Fisher GENESYS On-Board, Thermo Fisher NanoDrop 8000, Thermo Fisher NanoDrop Eight and 4 more
- Chig-AIMD Scaffold Split (Train): DFT Energies and Forces for Chignolin Protein Conformations, ColabFit Conversion
- Cd- and Zn-Doped CsPbI3 Perovskite Energetics: DFT Training Set for GNN Stability Prediction (Eremin et al., AIRI), ColabFit Conversion
- CO2 Hydrogenation to Methanol on In2O3 and Pt-Doped In2O3: ML Force Field Training Data (Schaaf et al. 2023), ColabFit Conversion
- Vector-QM24 (VQM24): DFT Data for 836k Small Molecules, ColabFit Conversion