Paper 2509.22186
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing
- Published
- Sep 2025
- Research lab
- Independent
- Citations
- 86
- GitHub
- 77K stars
01 In brief
Summary
MinerU2.5 is a 1.2B-parameter vision-language model for document parsing that achieves state-of-the-art accuracy with high efficiency.
It uses a decoupled, two-stage strategy: first, global layout analysis on a downsampled 1036x1036 image; second, targeted content recognition on native-resolution crops guided by the layout.
The model uses a 675M NaViT vision encoder and a 0.5B Qwen2 language model.
Training includes three stages: modality alignment, pre-training on 6.9M samples, and fine-tuning on 630K samples.
A data engine with iterative mining via inference consistency (IMIC) identifies hard cases for human annotation.
The model introduces a unified tagging system, PageIoU metric, ADR framework for formulas, and OTSL for tables.
On OmniDocBench, it scores 90.67 overall, outperforming models like MonkeyOCR-pro-3B (88.85) and dots.ocr (88.41).
It also achieves SOTA on Ocean-OCR and olmOCR-bench, with throughput of 2.12 pages/s on A100, 4x faster than MonkeyOCR-pro-3B.
02 From the paper
Abstract
We introduce MinerU2.5, a 1.2B-parameter document parsing vision-language model that achieves state-of-the-art recognition accuracy while maintaining exceptional computational efficiency. Our approach employs a coarse-to-fine, two-stage parsing strategy that decouples global layout analysis from local content recognition. In the first stage, the model performs efficient layout analysis on downsampled images to identify structural elements, circumventing the computational overhead of processing high-resolution inputs. In the second stage, guided by the global layout, it performs targeted content recognition on native-resolution crops extracted from the original image, preserving fine-grained details in dense text, complex formulas, and tables. To support this strategy, we developed a comprehensive data engine that generates diverse, large-scale training corpora for both pretraining and fine-tuning. Ultimately, MinerU2.5 demonstrates strong document parsing ability, achieving state-of-the-art performance on multiple benchmarks, surpassing both general-purpose and domain-specific models across various recognition tasks, while maintaining significantly lower computational overhead.