# QM9: Quantum Chemistry of 134k Small Organic Molecules (GDB-9), ColabFit Conversion

## Facts

- Provider: Raghunathan Ramakrishnan, Pavlo O. Dral, Matthias Rupp, O. Anatole von Lilienfeld; ColabFit conversion
- License: CC0-1.0 (https://api.figshare.com/v2/articles/1057646)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/colabfit/GDB_9_nature_2014
- Access: public, readable without an account
- DOI: https://doi.org/10.60732/b82731e4
- Category: Chemistry & Materials
- Source: https://doi.org/10.1038/sdata.2014.22
- Page: https://scigantic.com/archive/bcb2f004-0504-5bae-84c1-d8ba55d4977b
- Keywords: Chemistry & Materials, quantum chemistry, graph learning, benchmark, Parquet, Hugging Face

## About

QM9, the benchmark of Ramakrishnan, Dral, Rupp and von Lilienfeld (Scientific Data 2014): equilibrium geometries and properties of the stable small organic molecules with up to nine heavy atoms (C, N, O, F) from the GDB-17 chemical universe, computed at the B3LYP/6-31G(2df,p) level. The original release includes energies, dipole moments, polarizabilities, enthalpies and orbital energies, and QM9 is the most widely used regression benchmark for molecular property prediction and graph neural networks (SchNet, DimeNet, PaiNN and successors). ColabFit's card counts 133,877 configurations with 2.4 million atoms; the repository is about 0.26 GB (measured 2026-11-14). 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 original figshare release is CC0; cite Ramakrishnan et al. 2014.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("colabfit/GDB_9_nature_2014")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `co/co_0.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/bcb2f004-0504-5bae-84c1-d8ba55d4977b
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/bcb2f004-0504-5bae-84c1-d8ba55d4977b
- Over MCP: `get_schema_card` with id `bcb2f004-0504-5bae-84c1-d8ba55d4977b` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Benchmark datasets](https://scigantic.com/topics/benchmark)
- [Chemistry & Materials datasets](https://scigantic.com/topics/chemistry-and-materials)
- [Quantum chemistry datasets](https://scigantic.com/topics/quantum-chemistry)
- [Graph learning datasets](https://scigantic.com/topics/graph-learning)
