ByteDance’s Dreamina Seedance 2.0 comes to CapCut
ByteDance is integrating its new audio and video generation model, Dreamina Seedance 2.0, into its CapCut editing platform.
The artificial intelligence model allows users to create, edit, and synchronize video and audio content using prompts, images, or reference videos, marking a direct competitor to similar technologies.
The phased rollout of Dreamina Seedance 2.0 will begin for CapCut users in Brazil, Indonesia, Malaysia, Mexico, the Philippines, Thailand, and Vietnam.
This deployment follows a report indicating a temporary halt to the model’s global release due to intellectual property concerns, which may explain the limited initial market availability.
In China, the model is already accessible through ByteDance’s Jianying application. The video generation model can function without reference images, creating scenes based on text descriptions alone, ByteDance stated.
CapCut can render realistic textures, movement, and lighting, which allows creators to edit, enhance, or correct their own footage. Creators can also use the model to test conceptual ideas before filming final videos.
Dreamina Seedance 2.0 supports various content types, including cooking recipes, fitness tutorials, and product overviews, areas where previous AI video models faced challenges.
At its launch, the model supports video clips up to 15 seconds in length across six aspect ratios.
The model will integrate into CapCut’s AI Video and Video Studio features and will also be available on ByteDance’s AI generation platform, Dreamina, and its marketing platform, Pippit.
ByteDance has implemented safety restrictions to prevent the model from generating videos with real faces from images or videos, and CapCut will block unauthorized intellectual property generation.
Content produced by Dreamina Seedance 2.0 will include an invisible watermark to identify AI-generated material when shared off-platform, aiding in intellectual property enforcement.
ByteDance plans to collaborate with experts and creative communities to refine the model’s capabilities during its rollout.