LightOnOCR-3 adds page locations to document extraction

What changed

According to LightOn, its grounding prompt adds labelled bounding boxes, image descriptions and chart values as tables alongside transcription.

Source: LightOn release announcement on Hugging Face

Our analysis

Builder’s view: test whether extracted passages link back to the correct page coordinates on your own PDFs before changing ingestion.

Keep in mind

We have not tested extraction accuracy or verified chart values on our documents.

Source: LightOn release announcement on Hugging Face