LightOnOCR-3 released: 0.8B/1B/4B open models for OCR, layout, charts under Apache 2.0
antoine_chaffin · x · 2026-10-09
LightOn released LightOnOCR-3, going beyond OCR with a single model handling text transcription, document layout analysis, image descriptions, and chart extraction. Three sizes are available—0.8B, 1B, and 4B—all under Apache 2.0. Researcher antoinechaffin endorsed the release, noting parsing is now built in.
Related event: LightOn open-sources LightOnOCR-3, beating Mistral OCR on benchmarks(3 posts)→
More from Models
- MathArena gave 4 agents $500 each to write blog posts; only Opus-5.5 delivered — ChrSzegedy · 2026-10-09
- Goodfire builds cybersecurity monitors for Kimi K3 and GLM 5.3, 50x faster and cheaper — CatAstro_Piyush · 2026-10-09
- Anthropic bans sustained abuse toward its models in extreme cases — Angaisb_ · 2026-10-09
- Dev complaint: models are still horrible at code reuse — snowclipsed · 2026-10-09
- Live: Liquid AI's Maxime Labonne on d1, decision models that never write a word — altryne · 2026-10-09
- Perplexity Decider tops DecisionBench as the lowest-cost decision model — victormustar · 2026-10-09