dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model

Open paper

Benchmark	Model	Rank	Results
optical-character-recognition-ocr-on-omnidocbench-v1-0	dots.ocr	#2	Edit Distance: 0.125
optical-character-recognition-ocr-on-olmocr-bench	dots.ocr	#9	Accuracy: 79.1