PaddlePaddle/PaddleOCR-VL-1.6
Image-Text-to-Text • 1.0B • Updated • 64.4k • 391
OCR
A bunch of VL Models/OCR Engines for OCR Tasks
-
dots-studio/dots.mocr
Image-Text-to-Text • 3B • Updated • 209k • 156Note Best performing OCR Engine
-
dots-studio/dots.ocr
Image-Text-to-Text • 3B • Updated • 384k • 1.32k
google/gemma-4-12B-it
Any-to-Any • 12B • Updated • 3.03M • 1.39kNote Best performing Vision-Language Model. could be used with a OCR Engine as a judge.