---
configs:
  - config_name: default
    data_files:
      - split: en
        path: data/ultrafineweb_en/*
      - split: zh
        path: data/ultrafineweb_zh/*
    features:
      - name: content
        dtype: string
      - name: score
        dtype: float
      - name: source
        dtype: string
task_categories:
- text-generation
language:
- en
- zh
pretty_name: Ultra-FineWeb
size_categories:
- n>1T
---
# Ultra-FineWeb

- [📜 Technical Report]()
- [💻 Github Repo]()
- [🤗 Classifier Models]()

</div>

## 📚 Introduction

Ultra-FineWeb datasets contains approximately 1 trillion English tokens and 120 billion Chinese tokens.

## 📢 What's New

- **[2025.xx.xx]** **Ultra-FineWeb** technical report is available on [arXiv](). 🔥🔥🔥
- Datasets and models are coming soon... 🔜🚀

## 💡 Highlights

## 📈 Evaluation Results

## ❤️ Acknowledgements

- The ***Ultra-FineWeb classifier*** is built based on [fastText](https://fasttext.cc/). 
- The ***Ultra-FineWeb-en dataset*** is built based on [FineWeb](https://huggingface.co/datasets/HuggingFaceFW/fineweb). 
- The ***Ultra-FineWeb-zh dataset*** is constructed based on [IndustryCorpus2](https://huggingface.co/datasets/BAAI/IndustryCorpus2), [MiChao](https://opendatalab.com/OpenDataLab/MiChao), [WuDao](https://data.baai.ac.cn/details/WuDaoCorporaText), [SkyPile](https://huggingface.co/datasets/Skywork/SkyPile-150B), [WanJuan](https://opendatalab.com/OpenDataLab/WanJuanCC), [ChineseWebText](https://huggingface.co/datasets/CASIA-LM/ChineseWebText2.0), [TeleChat](https://huggingface.co/datasets/Tele-AI/TeleChat-PTD), and [CCI3](https://huggingface.co/datasets/BAAI/CCI3-Data).

Thanks for their awesome work! Open-source contributions make Ultra-FineWeb possible! 🙌

## 🌟 Citation

If you find our work useful, please consider citing:

```bibtex
Coming soon...
```

## 💳 License

This project is released under the [MIT](./LICENSE). Please note that since ***Ultra-FineWeb*** is built using multiple datasets, users should check the **LICENSE of each dataset individually** to ensure proper usage and compliance.