video generation

At CVPR 2025, Kuaishou’s Kling AI division presented advancements in video generation and world model research. Their work focuses on improving model architectures, enhancing user control, establishing robust evaluation methods, and developing multimodal understanding. Key innovations include efficient scaling laws, novel Mixture of Experts architectures, a unified framework for spatiotemporal control, and frameworks for interactive and controllable video creation. The team’s research, highlighted across seven papers, aims to significantly advance video creation capabilities.

Alibaba open-sourced its modular Wan2.1-VACE video generation model (1.3B/14B parameters), featuring multimodal input support (text, images, video clips) and Lego-like customizable modules. The lightweight 1.3B version runs on consumer GPUs, democratizing AI video creation. Released on GitHub/Hugging Face, it has gained 330K+ downloads and 11K+ stars, becoming a leading open-source video generation framework.