HunyuanOCR-1.5：讓輕量級OCR視覺語言模型更快更好

摘要

我們提出HunyuanOCR-1.5，一款輕量級的端到端OCR專用視覺語言模型。HunyuanOCR將文檔解析、文字定位、資訊提取、文字-圖像翻譯以及多圖像文檔理解，統一在單一的端到端VLM中。基於HunyuanOCR-1.0的輕量級架構，HunyuanOCR-1.5並未重新設計主幹網路，而是系統性地提升了效率與能力。在效率方面，我們將DFlash適配至OCR解碼，顯著降低了密集文檔、表格與公式等長結構化輸出的延遲，同時保持輸出分佈不變。藉由DFlash，HunyuanOCR-1.5在Transformer推理上實現了6.37倍加速，在vLLM下實現2.14倍加速，成為輕量級OCR VLM中推理速度最快的模型。在能力方面，我們提出了Agentic Data Flow，這是一個基於智能體（agent）的數據構建系統，能將模型弱點轉化為可執行的數據需求，並自主進行資料搜索、品質驗證與流程開發。該系統顯著提升了古文字OCR、細粒度圖表與表格解析、多圖像文本中心問答、低資源多語言解析以及文檔幻覺評估等長尾能力。HunyuanOCR-1.5在OmniDocBench v1.6上位居頂尖端到端OCR解決方案之列，同時在這些長尾任務上創下新的性能里程碑。結合升級的預訓練與後訓練方案，HunyuanOCR-1.5進一步擴展了其在高解析度、長上下文以及多任務場景下的能力。實驗證明了更快的推理速度、更廣泛的OCR能力覆蓋，以及輕量級端到端模型的部署優勢。我們將釋出模型權重與訓練程式碼，以支援未來的研究與實際OCR應用。

English

We present HunyuanOCR-1.5, a lightweight end-to-end OCR-specialized vision-language model. HunyuanOCR unifies document parsing, text spotting, information extraction, text-image translation, and multi-image document understanding within a single end-to-end VLM. Building upon the lightweight architecture of HunyuanOCR-1.0, HunyuanOCR-1.5 does not redesign the backbone, but systematically improves both efficiency and capability. For efficiency, we adapt DFlash to OCR decoding, significantly reducing the latency of long structured outputs such as dense documents, tables, and formulas while preserving output distribution. Powered by DFlash, HunyuanOCR-1.5 achieves a 6.37x Transformer inference speedup and a 2.14x speedup under vLLM, delivering the fastest inference among lightweight OCR VLMs. For capability, we propose Agentic Data Flow, an agent-driven data construction system that transforms model weaknesses into executable data requirements and autonomously performs material search, quality verification, and pipeline development. It substantially improves long-tail capabilities in ancient-script OCR, fine-grained chart and table parsing, multi-image text-centric QA, low-resource multilingual parsing, and document hallucination evaluation. HunyuanOCR-1.5 ranks among the top-tier end-to-end OCR solutions on OmniDocBench v1.6 while achieving new performance milestones across these long-tail tasks. Combined with an upgraded pretraining and post-training recipe, HunyuanOCR-1.5 further extends its capability in high-resolution, long-context, and multi-task scenarios. Experiments demonstrate faster inference, broader OCR capability coverage, and the deployment advantages of a lightweight end-to-end model. We will release the model weights and training code to support future research and real-world OCR applications.