Skip to content

Lighton · OCR

LightOnOCR-2-1B

LightonOCRLightOnOCR familyOpen weights. The weights are publicly available in a Hugging Face repo.
From Eurouter's listing:
LightOnOCR-2-1B is LightOn's compact 1B-parameter end-to-end multilingual vision-language OCR model. It converts document images such as scans and photos into clean, naturally ordered Markdown, handling tables, receipts, forms, multi-column layouts and LaTeX math. It accepts image input only; PDFs must be converted to page images first.

Evidence and sources

Source: EURouter dataPrimary data from EURouter
Tags from its sources:open-sourcevisionocrmultilinguallightweight

Input & output

from the model's type
Images
Files
Text

Listings

We don't have any priced listings for this model yet.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

OCR

Other

Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works