TMTPOST – Tencent has officially launched and open-sourced its HunYuan Video Generation Model, a major milestone in its AI capabilities. The model, boasting 13 billion parameters, is capable of generating seamless transitions and camera cuts, producing videos up to 5 seconds long in a single process. HunYuan Video demonstrates notable strengths in generating wide-range and smooth motion videos, particularly excelling in categories such as human-centric scenarios, artificial environments, and multi-subject combinations. It also performs well in creating content involving objects, animals, and microorganisms. According to a representative of Tencent’s HunYuan Model team, this version focuses on text-to-video generation. Future updates may expand functionalities to include image-to-video conversion and automatic video scoring. The model is now available on Tencent’s YuanBao app, where users can apply for trial access under the “AI Video” section of the AI applications menu. This launch marks Tencent's continued innovation in video generation and its commitment to fostering AI accessibility.