Datalab releases OmniParseBench and ranks 17 open OCR models
VikParuchuri · x · 2026-10-25
Datalab founder Vik Paruchuri announces that 17 open OCR models have been benchmarked on OmniParseBench, Datalab's new OCR benchmark building on AllenAI's olmOCR-bench. The leaderboard chart shows Chandra OCR 2 at 90.1, dots.mocr 87.9, MinerU2.5-Pro 87.1, Infinity-Parser2-Pro 87.0, with HunyuanOCR last at 64.6. Details in the blog and follow-up posts.
Related event: Datalab Launches OmniParseBench, Chandra OCR 2 Tops Open-Source Ranking(2 posts)→
More from Research
- EMNLP Paper: Repetition Mismatch Explains Why Data Mixture Experiments Don't Scale — MarekRei · 2026-10-25
- Imperial NLP Brings 6 Main Papers and Agent-LM Workshop to EMNLP 2026 — MarekRei · 2026-10-25
- Google's Jon Barron shares his full AI paper-writing workflow with Gemini — lmoroney · 2026-10-25
- Only 30% of suspected rare-disease newborns get diagnosed; Gamow Labs launches 122-case RareBench for agentic AI — Sauers_ · 2026-10-25
- MMGR benchmark: top video models solve only ~3% of mazes despite stunning visuals — zhaofeng_wu · 2026-10-25
- Torino: reference-free isoform discovery learns RNA processing from coverage — anshulkundaje · 2026-10-25