Elamur Research⌕
☰
home / papers / 2025 / HunyuanOCR Technical Report

HunyuanOCR Technical Report

Paper · 2025 · Hunyuan Vision Team, Pengyuan Lyu, Xingyu Wan et al.

Summary

HunyuanOCR is Tencent's open-source OCR model. It addresses a trade-off in current systems. Cascaded pipelines such as PaddleOCR chain several expert modules, which compounds errors and adds maintenance cost. Large general VLMs are flexible but expensive to run. Many OCR-specialised VLMs still depend on a separate layout-analysis step.

The authors built a compact end-to-end VLM with about 1B parameters. A 0.4B native-resolution ViT feeds a 0.5B Hunyuan LLM through a learnable pooling MLP adapter. A single instruction-driven inference handles text spotting, document parsing, information extraction, text-centric VQA and text image translation.

Training has two stages:

  • Pre-training in four steps on roughly 200M samples, mixing open-source, synthetic and application-oriented data.
  • Post-training with online RL (GRPO) using task-specific rewards.

The model beats MinerU2.5 and PaddleOCR-VL on OmniDocBench. It also outperforms much larger VLMs on extraction and translation. The authors stress that application-aligned data and RL drive these gains. The model and a vLLM deployment recipe are released publicly.

Key points

  • Information extraction (Table 5): HunyuanOCR scores 92.29 on cards, 92.53 on receipts and 92.87 on video subtitles. Gemini-2.5-Pro, the strongest larger VLM in the comparison, reaches 80.59 and 80.66 on cards and receipts.
  • Translation: HunyuanOCR scores 83.48 COMET on DoTA (en2zh), above Qwen3-VL-8B (79.86) but below Gemini-2.5-Flash (85.60). It placed first in the ICDAR 2025 DIMT Track 2.2 (OCR-free Small Model).
  • OCRBench: HunyuanOCR is reported as state of the art among VLMs under 3B parameters. It is far ahead of DeepSeek-OCR (430) and roughly on par with Qwen3-VL-2B-Instruct.
  • Benchmarks: The authors introduce an in-house multilingual benchmark, DocML, with English and Chinese reference translations, plus a "Wild" version of OmniDocBench. Both releases are promised but still pending.
  • Limitations: The authors admit that translation lags behind the model's perception skills because the LLM is small. For higher translation quality, they suggest pairing it with Hunyuan-MT-7B.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksComputer Vision and Pattern Recognition · 2,869 works

Cite

Cite as Hunyuan Vision Team, Pengyuan Lyu, Xingyu Wan, Gengluo Li, Shangpin Peng, Weinong Wang, Liang Wu, Huawen Shen, Yu Zhou, Canhui Tang, Qi Yang, Qiming Peng, Bin Luo, Hower Yang, Xinsong Zhang, Jinnian Zhang, Houwen Peng, Hongming Yang, Senhao Xie, Longsha Zhou, Ge Pei, Binghong Wu, Rui Yan, Kan Wu, Jieneng Yang, Bochao Wang, Kai Liu, Jianchen Zhu, Jie Jiang, Linus, Han Hu and Chengquan Zhang. 2025. HunyuanOCR Technical Report. arXiv:2511.19575.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2511.19575 · arXiv 2511.19575

@misc{team2025hunyuanocr,
  title = {{HunyuanOCR Technical Report}},
  author = {Hunyuan Vision Team, Pengyuan Lyu, Xingyu Wan, Gengluo Li, Shangpin Peng, Weinong Wang, Liang Wu, Huawen Shen, Yu Zhou, Canhui Tang, Qi Yang, Qiming Peng, Bin Luo, Hower Yang, Xinsong Zhang, Jinnian Zhang, Houwen Peng, Hongming Yang, Senhao Xie, Longsha Zhou, Ge Pei, Binghong Wu, Rui Yan, Kan Wu, Jieneng Yang, Bochao Wang, Kai Liu, Jianchen Zhu, Jie Jiang, Linus, Han Hu and Chengquan Zhang},
  year = {2025},
  doi = {10.48550/arxiv.2511.19575},
  eprint = {2511.19575},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2511.19575}
}

References

60 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mistral OCR: Free online ai ocr tool to extract text. https://www.mistralocr.com/, 2025. Accessed: 2025-07-30. ↗
  2. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023.
  3. Ibrahim Adeshola and Adeola Praise Adepoju. The opportunities and challenges of chatgpt in education. Interactive Learning Environments, 32(10):6159–6172, 2024. find
  4. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, et al. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923, 2025.
  5. Baidu. BaiduOCRAPI, 2025. URL https://ai.baidu.com/tech/ocr/general. ↗
  6. Lukas Blecher, Guillem Cucurull, Thomas Scialom, and Robert Stojnic. Nougat: Neural optical understanding for academic documents. arXiv preprint arXiv:2308.13418, 2023. find
  7. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020.
  8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, et al. Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities. arXiv preprint a…
  9. Corinna Cortes and Vladimir Vapnik. Support-vector networks. Machine learning, 20(3):273–297, 1995.
  10. Cheng Cui, Ting Sun, Suyin Liang, Tingquan Gao, Zelun Zhang, Jiaxuan Liu, Xueqing Wang, Changda Zhou, Hongen Liu, Manhui Lin, et al. Paddleocr-vl: Boosting multilingual document parsing via a 0.9 b ultra-compact vision-language model. arXiv preprint arXiv:2510.14528, 2025a. — [arXiv:2510.14528](https://arxiv.org/abs/2510.14528) find
  11. Cheng Cui, Ting Sun, Manhui Lin, Tingquan Gao, Yubo Zhang, Jiaxuan Liu, Xueqing Wang, Zelun Zhang, Changda Zhou, Hongen Liu, et al. Paddleocr 3.0 technical report. arXiv preprint arXiv:2507.05595, 2025b. — [arXiv:2507.05595](https://arxiv.org/abs/2507.05595) find
  12. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volum…
  13. Sean R Eddy. Hidden markov models. Current opinion in structural biology, 6(3):361–365, 1996. find
  14. Hao Feng, Shu Wei, Xiang Fei, Wei Shi, Yingdong Han, Lei Liao, Jinghui Lu, Binghong Wu, Qi Liu, Chunhui Lin, et al. Dolphin: Document image parsing via heterogeneous anchor prompting. arXiv preprint arXiv:2505.14059, 2025. find
  15. JaidedAI. Easyocr, 2020. URL https://github.com/JaidedAI/EasyOCR. ↗
  16. Zhanghui Kuang, Hongbin Sun, Zhizhong Li, Xiaoyu Yue, Tsui Hin Lin, Jianyong Chen, Huaqiang Wei, Yiqin Zhu, Tong Gao, Wenwei Zhang, et al. Mmocr: a comprehensive toolbox for text detection, recognition and understanding. In ACM Multimedia, pp. 3791–3794, 2021. find
  17. Minghao Li, Tengchao Lv, Jingye Chen, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, and Furu Wei. Trocr: Transformer-based optical character recognition with pre-trained models. In Proceedings of the AAAI conference on artificial intelligence, volume 37, pp. 13094–13102, 2023. find
  18. Yulin Li, Yuxi Qian, Yuechen Yu, Xiameng Qin, Chengquan Zhang, Yan Liu, Kun Yao, Junyu Han, Jingtuo Liu, and Errui Ding. Structext: Structured text understanding with multi-modal transformers. In Proceedings of the 29th ACM international conference on multimedia, pp. 1912–1920, 2021. find
  19. Zhang Li, Yuliang Liu, Qiang Liu, Zhiyin Ma, Ziyang Zhang, Shuo Zhang, Zidun Guo, Jiarui Zhang, Xinyu Wang, and Xiang Bai. Monkeyocr: Document parsing with a structure-recognition-relation triplet paradigm. arXiv preprint arXiv:2506.05218, 2025. find
  20. Yupu Liang, Yaping Zhang, Cong Ma, Zhiyang Zhang, Yang Zhao, Lu Xiang, Chengqing Zong, and Yu Zhou. Document image machine translation with dynamic multi-pre-trained models assembling. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistic… find
  21. Minghui Liao, Baoguang Shi, Xiang Bai, Xinggang Wang, and Wenyu Liu. Textboxes: A fast text detector with a single deep neural network. In Proceedings of the AAAI conference on artificial intelligence, volume 31, 2017. find
  22. Minghui Liao, Zhisheng Zou, Zhaoyi Wan, Cong Yao, and Xiang Bai. Real-time scene text detection with differentiable binarization and adaptive scale fusion. IEEE transactions on pattern analysis and machine intelligence, 45(1):919–931, 2022. find
  23. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, et al. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437, 2024a. find
  24. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. Advances in neural information processing systems, 36:34892–34916, 2023.
  25. Yuliang Liu, Zhang Li, Mingxin Huang, Biao Yang, Wenwen Yu, Chunyuan Li, Xu-Cheng Yin, Cheng-Lin Liu, Lianwen Jin, and Xiang Bai. Ocrbench: on the hidden mystery of ocr in large multimodal models. Science China Information Sciences, 67(12), 2024b.
  26. Yuliang Liu, Zhang Li, Mingxin Huang, Biao Yang, Wenwen Yu, Chunyuan Li, Xu-Cheng Yin, Cheng-Lin Liu, Lianwen Jin, and Xiang Bai. Ocrbench: on the hidden mystery of ocr in large multimodal models. Science China Information Sciences, 67(12):220102, 2024c.
  27. Yuqi Liu, Bohao Peng, Zhisheng Zhong, Zihao Yue, Fanbin Lu, Bei Yu, and Jiaya Jia. Seg-zero: Reasoning-chain guided segmentation via cognitive reinforcement. arXiv preprint arXiv:2503.06520, 2025. — [arXiv:2503.06520](https://arxiv.org/abs/2503.06520) find
  28. Shangbang Long, Xin He, and Cong Yao. Scene text detection and recognition: The deep learning era. International Journal of Computer Vision, 129(1):161–184, 2021. find
  29. Pengyuan Lyu, Minghui Liao, Cong Yao, Wenhao Wu, and Xiang Bai. Mask textspotter: An end-to-end trainable neural network for spotting text with arbitrary shapes. In Proceedings of the European conference on computer vision (ECCV), pp. 67–83, 2018. find
  30. Pengyuan Lyu, Yulin Li, Hao Zhou, Weihong Ma, Xingyu Wan, Qunyi Xie, Liang Wu, Chengquan Zhang, Kun Yao, Errui Ding, et al. Structextv3: An efficient vision-language model for text-rich image perception, comprehension, and beyond. arXiv preprint arXiv:2405.21013, 2024a. — [arXiv:2405.21013](https://arxiv.org/abs/2405.21013) find
  31. Pengyuan Lyu, Chengquan Zhang, Shanshan Liu, Meina Qiao, Yangliu Xu, Liang Wu, Kun Yao, Junyu Han, Errui Ding, and Jingdong Wang. Maskocr: Scene text recognition with masked vision-language pre-training. Transactions on Machine Learning Research, 2024b. find
  32. Junbo Niu, Zheng Liu, Zhuangcheng Gu, Bin Wang, Linke Ouyang, Zhiyuan Zhao, Tao Chu, Tianyao He, Fan Wu, Qintong Zhang, et al. Mineru2. 5: A decoupled vision-language model for efficient high-resolution document parsing. arXiv preprint arXiv:2509.22186, 2025.
  33. Linke Ouyang, Yuan Qu, Hongbin Zhou, Jiawei Zhu, et al. Omnidocbench: Benchmarking diverse pdf document parsing with comprehensive annotations, 2024. URL https://arxiv.org/abs/2412.07626. — [arXiv:2412.07626](https://arxiv.org/abs/2412.07626) ↗
  34. PaddleOCR. Pp-chatocr, 2025. URL https://github.com/PaddlePaddle/PaddleOCR. ↗
  35. Shangpin Peng, Weinong Wang, Zhuotao Tian, Senqiao Yang, Xing Wu, Haotian Xu, Chengquan Zhang, Takashi Isobe, Baotian Hu, and Min Zhang. Omni-dpo: A dual-perspective paradigm for dynamic preference learning of llms. arXiv preprint arXiv:2506.10054, 2025a. — [arXiv:2506.10054](https://arxiv.org/abs/2506.10054) find
  36. Shangpin Peng, Senqiao Yang, Li Jiang, and Zhuotao Tian. Mitigating object hallucinations via sentence-level early intervention. arXiv preprint arXiv:2507.12455, 2025b. — [arXiv:2507.12455](https://arxiv.org/abs/2507.12455) find
  37. Qwen. Qwen3-vl, 2025. URL https://github.com/QwenLM/Qwen3-VL. ↗
  38. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019. find
  39. Ricardo Rei, José G. C. de Souza, Duarte Alves, Chrysoula Zerva, Ana C Farinha, Taisiya Glushkova, Alon Lavie, Luisa Coheur, and André F. T. Martins. COMET-22: Unbabel-IST 2022 submission for the metrics shared task. In Philipp Koehn, Loı̈c Barrault, Ondřej Bojar, Fethi Bougares, Rajen Chatterjee…
  40. Seed. Seed1.6, 2025. URL https://seed.bytedance.com/en/seed1 6. ↗
  41. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. CoRR, abs/2402.03300, 2024. doi: 10.48550/ARXIV.2402.03300. URL https://doi.org/10.48550/arX iv.2402.03300… ↗
  42. Baoguang Shi, Xiang Bai, and Cong Yao. An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition. IEEE Trans. Pattern Anal. Mach. Intell., 39(11):2298–2304, 2017. find
  43. Baoguang Shi, Mingkun Yang, Xinggang Wang, Pengyuan Lyu, Cong Yao, and Xiang Bai. Aster: An attentional scene text recognizer with flexible rectification. IEEE transactions on pattern analysis and machine intelligence, 41(9):2035–2048, 2018. find
  44. Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, and Yunfeng Liu. Roformer: Enhanced transformer with rotary position embedding. Neurocomputing, 568:127063, 2024.
  45. Ting Sun, Cheng Cui, Yuning Du, and Yi Liu. Pp-doclayout: A unified document layout detection model to accelerate large-scale data construction. arXiv preprint arXiv:2503.17213, 2025. find
  46. Qwen Team. Qwen3 technical report, 2025. URL https://arxiv.org/abs/2505.09388. ↗
  47. Tencent. Hunyuan-0.5b, 2025. URL https://github.com/Tencent-Hunyuan/Hunyuan-0.5B. ↗
  48. Michael Tschannen, Alexey Gritsenko, Xiao Wang, Muhammad Ferjad Naeem, Ibrahim Alabdulmohsin, Nikhil Parthasarathy, Talfan Evans, Lucas Beyer, Ye Xia, Basil Mustafa, et al. Siglip 2: Multilingual vision-language encoders with improved semantic understanding, localization, and dense features. arXiv p… find
  49. Bin Wang, Chao Xu, Xiaomeng Zhao, Linke Ouyang, Fan Wu, Zhiyuan Zhao, Rui Xu, Kaiwen Liu, Yuan Qu, Fukai Shang, Bo Zhang, Liqun Wei, Zhihao Sui, Wei Li, Botian Shi, Yu Qiao, Dahua Lin, and Conghui He. Mineru: An open-source solution for precise document content extraction, 2024. URL https://arxiv.or… ↗
  50. Shansong Wang, Mingzhe Hu, Qiang Li, Mojtaba Safari, and Xiaofeng Yang. Capabilities of gpt-5 on multimodal medical reasoning. arXiv preprint arXiv:2508.08224, 2025a. — [arXiv:2508.08224](https://arxiv.org/abs/2508.08224) find
  51. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, et al. Internvl3. 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265, 2025b.
  52. Haoran Wei, Yaofeng Sun, and Yukun Li. Deepseek-ocr: Contexts optical compression. arXiv preprint arXiv:2510.18234, 2025.
  53. Xumeng Wen, Zihan Liu, Shun Zheng, Zhijian Xu, Shengyu Ye, Zhirong Wu, Xiao Liang, Yang Wang, Junjie Li, Ziming Miao, et al. Reinforcement learning with verifiable rewards implicitly incentivizes correct reasoning in base llms. arXiv preprint arXiv:2506.14245, 2025. find
  54. Shukang Yin, Chaoyou Fu, Sirui Zhao, Ke Li, Xing Sun, Tong Xu, and Enhong Chen. A survey on multimodal large language models. National Science Review, 11(12):nwae403, 2024.
  55. Yuechen Yu, Yulin Li, Chengquan Zhang, Xiaoqiang Zhang, Zengyuan Guo, Xiameng Qin, Kun Yao, Junyu Han, Errui Ding, and Jingdong Wang. Structextv2: Masked visual-textual prediction for document image pre-training. In The Eleventh International Conference on Learning Representations. Qintong Zhang, Vi… — [arXiv:2410.21169](https://arxiv.org/abs/2410.21169) find
  56. Qintong Zhang, Bin Wang, Victor Shea-Jay Huang, Junyuan Zhang, Zhengren Wang, Hao Liang, Conghui He, and Wentao Zhang. Document parsing unveiled: Techniques, challenges, and prospects for structured information extraction. arXiv preprint arXiv:2410.21169, 2024b. — [arXiv:2410.21169](https://arxiv.org/abs/2410.21169) find
  57. Yaping Zhang, Yupu Liang, Zhiyang Zhang, Zhiyuan Chen, Lu Xiang, Yang Zhao, Yu Zhou, and Chengqing Zong. Icdar 2025 competition on end-to-end document image machine translation towards complex layouts. In International Conference on Document Analysis and Recognition, pp. 505–522. find
  58. Jiaxing Zhao, Xihan Wei, and Liefeng Bo. R1-omni: Explainable omni-multimodal emotion recognition with reinforcement learning. arXiv preprint arXiv:2503.05379, 2025. — [arXiv:2503.05379](https://arxiv.org/abs/2503.05379) find
  59. Zhiyuan Zhao, Hengrui Kang, Bin Wang, and Conghui He. Doclayout-yolo: Enhancing document layout analysis through diverse synthetic data and global-to-local adaptive perception. arXiv preprint arXiv:2410.12628, 2024. — [arXiv:2410.12628](https://arxiv.org/abs/2410.12628) find
  60. Xinyu Zhou, Cong Yao, He Wen, Yuzhi Wang, Shuchang Zhou, Weiran He, and Jiajun Liang. EAST: an efficient and accurate scene text detector. In CVPR, pp. 2642–2651. IEEE Computer Society, 2017. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/team-2025-vision-hunyuanocr-technical-report/