ModelRadar
/

Youtu-Parsing-Omni

Tencent🇨🇳 CNMajor lab· Oct 9, 2026

TextVisionopen weights

One model that parses documents, images, charts, geometry, audio and video into structured JSON

From the model card

License
other
Modalities
text, image → text

1 variant

StandardHF

Sources

HuggingFacePaper

Provenance

Detected
Oct 9, 2026
First source
huggingface
Description
From the model card
Editorially reviewed
—