Tencent releases WeMM-Video technical report on open-source omni-modal video understanding
The WeMM-Video family advances open-source video models with omni-modal capabilities, according to a new Tencent technical report.
Watch on YouTube · Watch the Short
The report introduces the WeMM-Video family, a set of models designed for omni-modal video understanding, combining visual and textual information.
Authors include Changli Tang, Jie Yang, Tianyi Wang, Lei Li, Daoze Zhang, Zijie Xin, Zhan Liu, and Li Li, among others.
The paper is hosted on Hugging Face and is linked to an interactive ResearchStudio Reel that provides a visual poster, video, and editable blog files.
The release highlights Tencent's commitment to open-source AI research and aims to foster community development around video understanding technologies.
What it means for you
The WeMM-Video report gives developers free access to advanced video models, supporting projects that need omni-modal capabilities without licensing costs. For users seeking local or free AI solutions, the open-source nature of WeMM-Video offers a new resource to build and experiment with video understanding applications.
Sources
- Tencent paper: WeMM-Video: Pushing the Frontiers of Open-Source Omni-Modal Video Understanding, Hugging Face Papers (official), 18 Nov 2026
Written automatically from the sources above, then checked by our code: every number and name must appear in them, and no sentence may copy a source. Spotted a mistake? Report it.
Which AI can you use for free today?
Every free AI model, ranked by blind human votes, with the memory and speed it needs on your own GPU.