README.md

---
license: apache-2.0
library_name: transformers
pipeline_tag: question-answering
---

# CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction

This repository contains the model described in the paper [CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction](https://huggingface.co/papers/2502.07316).

<p align="left">
    📑 <a href="https://huggingface.co/papers/2502.07316" target="_blank">Paper</a> &nbsp&nbsp | &nbsp&nbsp 🌐 <a href="https://codei-o.github.io/" target="_blank">Project Page</a> &nbsp&nbsp | &nbsp&nbsp 💾 <a href="https://huggingface.co/collections/hkust-nlp/codei-o-67a978e28fd926b56a4f55a2" target="_blank">Released Resources</a> &nbsp&nbsp | &nbsp&nbsp 📦 <a href="https://github.com/hkust-nlp/CodeIO" target="_blank">Repo</a>  

This is the resource page of the CodeI/O collection on Huggingface, we highlight your currect position with a blue block.

**Dataset**
<table>
      <tr>
        <th>Dataset</th>
        <th>Link</th>
    </tr>
      <tr>
        <td>CodeI/O-PythonEdu-Reasoning</td>
        <td style="text-align: center; vertical-align: middle;">
          <a href="https://huggingface.co/datasets/hkust-nlp/CodeIO-PyEdu-Reasoning">🤗</a>
        </td>
    </tr>
</table>

**Models**
<table>
    <tr>
        <th rowspan="2">Base Model / Training</th>
        <th colspan="2">CodeI/O</th>
        <th colspan="2">CodeI/O++</th>
    </tr>
    <tr>
        <th>Stage 1</th>
        <th>Stage 2</th>
        <th>Stage 1</th>
        <th>Stage 2</th>
    </tr>
    <tr>
        <td>Qwen 2.5 7B Coder</td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/qwen2.5-7b-coder_codeio_stage1">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/qwen2.5-7b-coder_codeio">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/qwen2.5-7b-coder_codeio_pp_stage1">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/qwen2.5-7b-coder_codeio_pp">🤗</a></td>
    </tr>
    <tr>
        <td>LLaMA 3.1 8B</td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/llama3.1-8b_codeio_stage1">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/llama3.1-8b_codeio">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/llama3.1-8b_codeio_pp_stage1">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/llama3.1-8b_codeio_pp">🤗</a></td>
    </tr>
    <tr>
        <td>DeepSeek v2 Lite Coder</td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/dsv2-lite-coder_codeio_stage1">🤗</a></td>
        <td style="background-color: #e6f3ff; text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/dsv2-lite-coder_codeio">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/dsv2-lite-coder_codeio_pp_stage1">🤗</a></td>
        <td style="text-align: center; vertical-align: middle;"><a href="https://huggingface.co/hkust-nlp/dsv2-lite-coder_codeio_pp">🤗</a></td>
    </tr>
</table>

## Citation
If you find these resources helpful, please kindly cite as:
```
@article{li2025codeio,
  title={CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction},
  author={Li, Junlong and Guo, Daya and Yang, Dejian and Xu, Runxin and Wu, Yu and He, Junxian},
  journal={arXiv preprint arXiv:2502.07316},
  year={2025}
}
```