On Oct. 29, SenseTime open-sourced SenseNova 6.8 Code2Video, a model that writes code to produce editable videos, broadening the toolbox for AI-generated video workflows.
The release landed amid a burst of Chinese video-generation activity: Tsinghua and Nvidia published a leaderboard of 38 real-time video models, while Alibaba, Kunlun, Aishi and Shengshu rolled out new video models.
SenseTime's approach emphasizes editability through code generation, a direction distinct from pure text-to-video pipelines and attractive for production-grade content.
The cluster of open releases reflects intensifying competition among Chinese AI firms on generative video, a strategically important and fast-moving front.
Image: AI-related computer chips. Source: Wikimedia Commons (CC BY-SA 4.0).
