---
library_name: transformers
license: mit
datasets:
- CodeGoat24/UniGenBench-Eval-Images
base_model:
- CodeGoat24/UnifiedReward-2.0-qwen-72b
---


# UniGenBench-EvalModel-qwen-72b-v1

This model is tailored for offline T2I model evaluation on [UniGenBench](https://github.com/CodeGoat24/UniGenBench), which achieves an average accuracy of 94% compared to evaluations by Gemini 2.5 Pro.

Feel free to use this model to assess and compare the performance of your models.

![image](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F654c6845bac6e6e49895a5b5%2FgsE1RuzJq2lAlPh8bLSWf.png)
![image](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F654c6845bac6e6e49895a5b5%2Fk9ZzcWY6zGM0mUmml8lnS.png)


For further details, please refer to the following resources:
- 📰 Paper: https://arxiv.org/pdf/2508.20751
- 🪐 Project Page: https://codegoat24.github.io/UnifiedReward/Pref-GRPO
- 🤗 UniGenBench: https://github.com/CodeGoat24/UniGenBench
- 🤗 Leaderboard: https://huggingface.co/spaces/CodeGoat24/UniGenBench_Leaderboard
- 👋 Point of Contact: [Yibin Wang](https://codegoat24.github.io)


## Citation

```bibtex
@article{UniGenBench++,
  title={UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation},
  author={Wang, Yibin and Li, Zhimin and Zang, Yuhang and Bu, Jiazi and Zhou, Yujie and Xin, Yi and He, Junjun and Wang, Chunyu and Lu, Qinglin and Jin, Cheng and others},
  journal={arXiv preprint arXiv:2510.18701},
  year={2025}
}

@article{UniGenBench,
  title={Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning},
  author={Wang, Yibin and Li, Zhimin and Zang, Yuhang and Zhou, Yujie and Bu, Jiazi and Wang, Chunyu and Lu, Qinglin, and Jin, Cheng and Wang, Jiaqi},
  journal={arXiv preprint arXiv:2508.20751},
  year={2025}
}
```