30 upvotes
One Encoder Parses Docs, Images, Audio, Video into Structured Data
What is Youtu-Parsing-Omni: One Encoder, One Schema, All Modalities?
Youtu-Parsing-Omni turns any kind of file, text, picture, sound or video, into a time- and space-indexed record of its contents. The single-encoder design lets downstream agents read, locate and cite evidence across modalities without separate models.
Who it is for
Researchers building multimodal agents or retrieval systems can use it to give their bots a unified perception layer.