Promote Sign in Sign up

Multimodal model

dots.ocr

dots.ocr is a 1.7B parameter vision-language model for multilingual document layout parsing, unifying layout detection, OCR and reading order in one model, and achieving state-of-the-art results on OmniDocBench.

Rednote HiLab Multimodal Open weights mit
Open model card ↗
Type
Multimodal
Weights
Open
Downloads
1M
Likes
1.3k
Pipeline
image text to text
License
mit

Model hub

dots-studio/dots.ocr

Repo
dots-studio/dots.ocr
Library
dots_ocr
Access
Public
Technical tags and languages
en zh dots_ocr safetensors text-generation image-to-text ocr document-parse layout table formula transformers custom_code image-text-to-text
Open on Hugging Face

More from Rednote HiLab

4

About Rednote HiLab

1

rednote (Xiaohongshu) is a Chinese social commerce platform where users share lifestyle content through photos, videos, and live streams, discover products, and make purchases. It operates as a hybrid social network and e-commerce hub serving over 350 million monthly active users, primarily in China.