# LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR arXiv: https://arxiv.org/abs/2601.14251 · HTML: https://mlsanity.com/paper/2601.14251/ ## Key facts - Cited by 23 papers in the map (data through 14 Oct 2026). - Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models. - Builds on 15 of its 27 references that are in the map. - First posted on arXiv 20 Jan 2026. ## Placed in - Multimodal LLMs / vision-language models / [Document understanding and multimodal RAG](https://mlsanity.com/mllm/document-understanding-and-multimodal-rag/) ## Builds on (most cited first, 15 of 15) - [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024) - [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024) - [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023) - [Editing Models with Task Arithmetic](https://mlsanity.com/paper/2212.04089/) (2022) - [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024) - [An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://mlsanity.com/paper/1507.05717/) (2015) - [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://mlsanity.com/paper/2109.10282/) (2021) - [MinerU: An Open-Source Solution for Precise Document Content Extraction](https://mlsanity.com/paper/2409.18839/) (2024) - [Nougat: Neural Optical Understanding for Academic Documents](https://mlsanity.com/paper/2308.13418/) (2023) - [PP-OCR: A Practical Ultra Lightweight OCR System](https://mlsanity.com/paper/2009.09941/) (2020) - [olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models](https://mlsanity.com/paper/2502.18443/) (2025) - [PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model](https://mlsanity.com/paper/2510.14528/) (2025) - [olmOCR 2: Unit Test Rewards for Document OCR](https://mlsanity.com/paper/2510.19817/) (2025) - [dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model](https://mlsanity.com/paper/2512.02498/) (2025) - ['Eclair -- Extracting Content and Layout with Integrated Reading Order for Documents](https://mlsanity.com/paper/2502.04223/) (2025) ## Built on by (most cited first, 23 of 23) - [Structure-Aware Text Recognition for Ancient Greek Critical Editions](https://mlsanity.com/paper/2603.02803/) (2026) - [Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions](https://mlsanity.com/paper/2605.27750/) (2026) - [RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference](https://mlsanity.com/paper/2605.00392/) (2026) - [Infinity-Parser2 Technical Report](https://mlsanity.com/paper/2607.07836/) (2026) - [Beyond String Matching: Semantic Evaluation of PDF Table Extraction](https://mlsanity.com/paper/2603.18652/) (2026) - [GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scripts](https://mlsanity.com/paper/2604.12978/) (2026) - [P-MTP: Efficient Document Parsing via Multi-Token Prediction with Progressive Depth Scaling](https://mlsanity.com/paper/2606.24447/) (2026) - [StrucTab: A Structured Optimization Framework for Table Parsing](https://mlsanity.com/paper/2606.29905/) (2026) - [Benchmarking Vision-Language Models for French PDF-to-Markdown Conversion](https://mlsanity.com/paper/2602.11960/) (2026) - [METATR: A Multilingual, Evolving Benchmark for Automatic Text Recognition](https://mlsanity.com/paper/2605.26712/) (2026) - [End-to-End Text Line Detection and Ordering](https://mlsanity.com/paper/2606.04166/) (2026) - [Freeing the Law with LOCUS: A Local Ordinance Corpus for the United States](https://mlsanity.com/paper/2606.19334/) (2026) - [Cross-Temporal Sinhala OCR: Page-Level Adaptation and Diachronic Analysis](https://mlsanity.com/paper/2606.29378/) (2026) - [ClinOCR-Bench: A Comprehensive Clinical Scanned Document Dataset for Optical Character Recognition Model Evaluation](https://mlsanity.com/paper/2607.03650/) (2026) - [HPD-Parsing: Hierarchical Parallel Document Parsing](https://mlsanity.com/paper/2607.18839/) (2026) - [PUMA: A Polish Benchmark for Culturally Grounded Multimodal Understanding](https://mlsanity.com/paper/2608.21853/) (2026) - [OCR-Based Field Extraction for Archaeological Pottery Metadata: The CENTURIA Dataset](https://mlsanity.com/paper/2608.30616/) (2026) - [Jina-OCR-v1: Efficient Document Parsing with Speculative Decoding and Dense Verifiable Rewards](https://mlsanity.com/paper/2609.03181/) (2026) - [ExpertHTR: Unified Handwritten Text Recognition with Multi-Task Learning and Sparse Mixture-of-Experts](https://mlsanity.com/paper/2609.12705/) (2026) - [A Free Lunch? Adapting PP-OCRv6 for Historical Text Recognition](https://mlsanity.com/paper/2609.20064/) (2026) - [The Spoken Wikipedia Presentation Corpus](https://mlsanity.com/paper/2609.21676/) (2026) - [Xiaomi-OCR-0 Technical Report](https://mlsanity.com/paper/2609.36136/) (2026) - [The Last Mile of Coding Agents for Long-Document Tasks: An Empirical Study on Clinical Trial Protocol Digitization](https://mlsanity.com/paper/2610.18507/) (2026) ## Abstract (arXiv) We present LightOnOCR-2-1B, a 1B-parameter end-to-end multilingual vision--language model that converts document images (e.g., PDFs) into clean, naturally ordered text without brittle OCR pipelines. Trained on a large-scale, high-quality distillation mix with strong coverage of scans, French documents, and scientific PDFs, LightOnOCR-2 achieves state-of-the-art results on OlmOCR-Bench while being 9× smaller and substantially faster than prior best-performing models. We further extend the output format to predict normalized bounding boxes for embedded images, introducing localization during pretraining via a resume strategy and refining it with RLVR using IoU-based rewards. Finally, we improve robustness with checkpoint averaging and task-arithmetic merging. We release model checkpoints under Apache 2.0, and publicly release the dataset and LightOnOCR-bbox-bench evaluation under their respective licenses.