Extract text from documents and images with Datalab Marker and OCR

Turn whole documents into markdown or grab line-level polygons with two new models from Datalab.