Metadata-Version: 2.4
Name: c8data
Version: 0.1.0b2
Summary: Instruction-driven AutoData synthetic dataset generation.
Project-URL: Homepage, https://github.com/peanutsee/c8data
Project-URL: Documentation, https://github.com/peanutsee/c8data/blob/main/docs/README.md
Project-URL: Repository, https://github.com/peanutsee/c8data
Project-URL: Issues, https://github.com/peanutsee/c8data/issues
Project-URL: Paper, https://arxiv.org/abs/2606.25996
Author: c8data contributors
License-Expression: MIT
License-File: LICENSE
Keywords: agentic,autodata,dataset-generation,llm,synthetic-data
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.11
Requires-Dist: openai>=1.0
Requires-Dist: pydantic>=2.0
Provides-Extra: dev
Requires-Dist: pytest; extra == 'dev'
Requires-Dist: pytest-asyncio; extra == 'dev'
Requires-Dist: ruff; extra == 'dev'
Description-Content-Type: text/markdown

# c8data

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)

`c8data` is an installable Python package for instruction-driven synthetic data
generation inspired by [Meta's AutoData](https://facebookresearch.github.io/RAM/blogs/autodata/)
/ Agentic Self-Instruct loop.

It creates candidate examples, verifies their quality, tests them against weak
and strong solvers, judges the solver outputs with a rubric, and only accepts
examples that create a meaningful weak-vs-strong performance gap.

## Documentation

- **[Setup guide](docs/setup.md)** — install, API keys, model configuration
- **[Usage guide](docs/usage.md)** — API reference, sources, output format, examples

## Quick start

```bash
uv add c8data
export C8DATA_API_KEY="your-api-key"
```

```python
from c8data import AutoDataGenerator

gen = AutoDataGenerator(output_dir="./data/out")
result = gen.run(source="paper.md")

print(result.accepted)
print(result.output_path)
```

See the [setup guide](docs/setup.md) for full configuration options and the
[usage guide](docs/usage.md) for custom instructions, batch runs, and output format.

## License

This project is licensed under the [MIT License](LICENSE).

## Citation

If you use `c8data`, please cite the AutoData paper ([arXiv:2606.25996](https://arxiv.org/abs/2606.25996)):

```bibtex
@article{autodata2026,
  title={Autodata: an agentic data scientist to create high quality data},
  author={Kulikov, Ilia and Whitehouse, Chenxi and Wu, Tianhao and Nie, Yixin and Saha, Swarnadeep and Helenowski, Eryk and Yuan, Weizhe and Golovneva, Olga and Lanchantin, Jack and Bachrach, Yoram and Foerster, Jakob and Li, Xian and Fang, Han and Sukhbaatar, Sainbayar and Weston, Jason},
  journal={arXiv preprint arXiv:2606.25996},
  year={2026}
}
```

A copy is also available in [CITATION.bib](CITATION.bib).
