Data to be translated: Shanghai Artificial Intelligence Laboratory has open-sourced a video generation model named LaVie. LaVie employs a cascade latent diffusion model, capable of producing high-quality, realistic video content. Users can create unique video scenes through various text prompts, achieving a personalized video generation experience. LaVie also offers pre-trained models and a rich set of configuration options, allowing users to make flexible adjustments according to their needs. This model not only generates videos of single scenes but also supports interpolation of multiple videos and super-resolution processing to enhance video quality.