Two new OCR models land on Hugging Face: WeVisDoc-4B scores 95.38 on OmniDocBench

NielsRogge · x · 2026-09-18

Two OCR models landed on Hugging Face today:

Which to pick: on the OmniDocBench v1.6 benchmark (Papers with Code), WeVisDoc-4B leads with an overall score of 95.38, though the current SOTA remains NaviDC-OCR.

Related event: Tencent WeVisDoc and Jina's jina-ocr-v1 open-source OCR models launch same day(5 posts)→

Original post →

More from Models

Models channel →