About LightOnOCR-2

LightOnOCR-2 is an efficient end-to-end 1B-parameter vision-language model for converting documents (PDFs, scans, images) into clean, naturally ordered text without relying on brittle pipelines. This second version is trained on a larger and higher-quality corpus with stronger French, arXiv, and scan coverage, improved LaTeX handling, and cleaner normalization. LightOnOCR-2 achieves state-of-the-art performance on OlmOCR-Bench while being ~9× smaller and significantly faster than competing approaches.

Highlights

• ⚡ Speed: 3.3× faster than Chandra OCR, 1.7× faster than OlmOCR, 5× faster than dots.ocr, 2× faster than PaddleOCR-VL-0.9B, 1.73× faster than DeepSeekOCR
• 💸 Efficiency: Processes 5.71 pages/s on a single H100 (~493k pages/day) for <$0.01 per 1,000 pages
• 🧠 End-to-End: Fully differentiable, no external OCR pipeline
• 🧾 Versatile: Handles tables, receipts, forms, multi-column layouts, and math notation
• 📍 Image detection: Predicts bounding boxes for embedded images (bbox variants)

📄 Paper | 📝 Blog Post | 🚀 Demo | 📊 Dataset | 📊 BBox Dataset | 📓 Finetuning Notebook

Model Variants

Benchmarks

Usage with Ollama

ollama run Maternion/LightOnOCR-2:1b

Rendering and Preprocessing Tips

• Render PDFs to PNG or JPEG at a target longest dimension of 1540px
• Maintain aspect ratio to preserve text geometry
• Use one image per page; batching supported by vLLM

Fine-tuning

LightOnOCR-2 is fully differentiable and supports:

• LoRA fine-tuning
• Domain adaptation (receipts, scientific articles, forms, etc.)
• Multilingual fine-tuning with task-specific corpora

For fine-tuning, we recommend starting with the LightOnOCR-2-1B-base variant.

License

Apache License 2.0

Citation

@misc{lightonocr2_2026,
  title        = {LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR},
  author       = {Said Taghadouini and Adrien Cavaill\`{e}s and Baptiste Aubertin},
  year         = {2026},
  howpublished = {\url{https://arxiv.org/pdf/2601.14251}}
}