Tencent Open-Sources 5B Omni-Modal Parsing Model

Tencent open-sourced Youtu-Parsing-Omni, a 5B-parameter omni-modal model on Hugging Face that outputs structured parsing from a single input of documents, images, charts, geometry, audio or video, unifying OCR and AV parsing.

2026-10-09 ~ 2026-10-09 · 2 related posts

• Tencent releases Youtu-Parsing-Omni, an open multimodal model for document parsing, OCR, audio and video — tencent · 2026-10-09
• Tencent open-sources Youtu-Parsing-Omni, a 5B omni-modal parsing model — jacek2023 · 2026-10-09