LongCat-Video-Avatar 1.5
Open-source audio-driven avatar video generation with stronger stability, lip-sync accuracy, identity consistency, and fast 8-step inference.
LongCat-Video-Avatar 1.5 is an upgraded open-source audio-driven avatar video generation framework from the Meituan LongCat team. It focuses on production-ready talking-avatar and performance-avatar generation with stronger lip synchronization, long-video temporal stability, identity consistency, stylized-domain generalization, and practical 8-step inference.
The technical report presents a broader avatar generation system built for expressive speech-driven video, singing, animation, and multi-person scenarios.
Links: Project page · arXiv · GitHub