OpenCodePapers

optical-character-recognition-ocr-on-olmocr-bench

Optical Character Recognition (OCR)
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeAccuracyModelNameReleaseDate
Infinity-Parser2 Technical Report✓ Link87.6Infinity-Parser2-Pro2026-07-08
LightOnOCR-3: High-Performance OCR and Layout Extraction in One Model✓ Link86.3LightOnOCR-3-4B2026-10-08
Chandra OCR 2✓ Link85.8Chandra 22026-03-16
Mistral OCR 4✓ Link85.20Mistral OCR 42026-06-23
Jina-OCR-v1: Efficient Document Parsing with Speculative Decoding and Dense Verifiable Rewards83.4Jina-OCR-v12026-09-02
Surya OCR 2✓ Link83.3Surya OCR 22026-05-27
LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR83.2LightOnOCR-2-1B (1B)2026-01-20
Chandra83.1Chandra2025-10-21
Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing✓ Link82.5Infinity-Parser-7B2025-06-01
olmOCR 2: Unit Test Rewards for Document OCR82.4olmOCR-2-7B-1025 (7B)2025-10-22
DARL: Efficient Document-to-Markup Generation via Look-Ahead Diffusion Trajectory Sampling81.7DARL (extra data)2026-09-08
PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vision-Language Model✓ Link80.0PaddleOCR-VL2025-10-16
Qianfan-OCR: A Unified End-to-End Model for Document Intelligence✓ Link79.8Qianfan-OCR2026-03-11
dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model✓ Link79.1dots.ocr2025-12-02
olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models✓ Link75.5olmOCR2025-02-25
Check out how to contribute new results. Then edit this file on GitLab or GitHub.