Tag
Fable 5.5 Told 40,000 Years of Art History in a 15-Second Video, Even Including a Cat Side Plot, Showcasing the Model's Capabilities in Long-Span Narrative and Creative Generation.
美团开源了13.6B参数的LongCat-Video视频生成模型，原生支持文本到视频、图像到视频和视频续写，可在720p/30fps下生成数分钟长视频且无色彩衰减或质量下降，权重采用MIT许可证并允许商业使用。
Tavus announced Griffin, what it calls the first Human Interaction Model and the first real-time video system to pass a video Turing test, with a self-run study claiming 48% of one-minute call participants believed they spoke to a real person, plus a top score on NVIDIA's VideoFDB perception track.
The paper introduces Code World Model, a framework that combines language models for reasoning and coding with video models for visual rendering to simulate persistent world evolution in an open-ended manner.
MiniMax has released an open-weight video model, though with regional restrictions in the license.
LingBot-Video, a 30B-parameter video model with sparse MoE, designed for embodied intelligence, is open-sourced. It outperforms existing models on RBench, trained on 70K+ hours of embodied data.
LingBot-Video is the first open-source large-scale MoE video generation model for embodied intelligence, featuring efficient MoE architecture, massive embodied data training, and multi-reward system for high aesthetics, physical rationality, and task completion.
ByteDance unveiled Seedance 2.5, an AI video model that can generate 30-second 4K clips from a single prompt or up to 50 reference pieces, rivaling OpenAI and Google's offerings.
LongCat has released an open source video avatar model on Hugging Face that is free to use and capable of impressive feats.