ByteDance’s generative AI creative platform, Dreamina AI, has recently unveiled the integration of its latest advancements: the Seedance 2.0 video generation model and the Seedream 5.0 Lite image generation model. Seedance 2.0 stands out with its support for four input modalities—image, video, audio, and text—providing users with a richer and more versatile set of generation methods. It enables mixed inputs from multiple file types and can produce videos of up to 15 seconds in length. Notably, its 'reference capability' is exceptional, although it currently does not support the use of photos or videos featuring real people as primary references.
Seedream 5.0 Lite, on the other hand, boasts three key features: real-time online retrieval, precise and controllable editing, and intelligent logical reasoning. It facilitates image generation through retrieval functions, significantly enhancing the alignment between generated images and the input text. Additionally, it incorporates a built-in professional knowledge base, further refining its capabilities.