LightOnOCR-3 Launches: Tops FRBench, Adds Layout Structure and Chart Extraction
IgorCarron · x · 2026-11-02
French company LightOn released LightOnOCR-3, an end-to-end document understanding model family (predecessor downloaded 4.5M+ times on Hugging Face) that goes beyond transcription:
- Benchmarks: #1 on FRBench-pdf2md (notably strong on handwriting), #2 on OlmOCR-Bench, and #1 among open-weight models on ParseBench
- Layout structure: groups content into logical paragraphs, headings and tables with labels and bounding boxes for chunking, search and page anchoring
- Charts & images: describes images and extracts structured data from charts, curves and scientific figures
A commenter claims it can read 16th-century Florentine, Ancient Greek and Arabic without task-specific training (third-party anecdote, unverified).
Related event: LightOn Releases LightOnOCR-3 for End-to-End Document Understanding(2 posts)→
More from Models
- OpenAI merges ChatGPT's chat and Work tabs; Codex now shares the same usage pool — TechCrunch AI · 2026-11-02
- Reddit users report Gemini Argon underperforming Gemini 3.1 Pro on instructions and audio — PreparationOdd430 · 2026-11-02
- Dev warning: avoid yearly prepaid LLM plans as labs quietly cut quotas — HankYeomans · 2026-11-02
- KVARK Sparkle 1.0 open-weights: Qwen3.8-27B derivative cuts thinking tokens by up to 67% — dzulajla98 · 2026-11-02
- 129 Cyber Incidents in 10 Days: Open Models Already Causing "Billions per Day" in Damages — Afinetheorem · 2026-11-02
- Rumor: Qwen 4 will be Alibaba's last open-weight model release — Chemical_Hawk_6307 · 2026-11-02