ByteDance is rolling out its new audio and video generation model, Dreamina Seedance 2.0, to its editing platform, CapCut.
This deployment positions ByteDance as a key player in the AI video generation market, following OpenAI’s decision to scale back its Sora app efforts.
The model enables creators to draft, edit, and sync video and audio content using prompts, images, or reference videos, according to ByteDance.
The rollout will initially target CapCut users in Brazil, Indonesia, Malaysia, Mexico, the Philippines, Thailand, and Vietnam. ByteDance stated that additional markets will be added progressively.
The company previously faced intellectual property concerns regarding the model, which prompted a reported pause in its global rollout and criticism from Hollywood over alleged copyright infringement. This may explain the limited initial market availability.
In China, Dreamina Seedance 2.0 is available within ByteDance’s Jianying app.
The video generation model can function without reference images, allowing creators to describe scenes with a few words, ByteDance announced.
CapCut can render realistic textures, movement, and lighting across various visual perspectives and angles for the model. The company noted this capability could enhance or correct existing creator footage.
Users can also test content ideas based on early concepts or sketches before filming, the company said.
Dreamina Seedance 2.0 supports content types such as cooking recipes, fitness tutorials, business overviews, and action-focused videos, areas where AI video models previously encountered challenges, ByteDance explained.
At launch, the model supports video clips up to 15 seconds long across six aspect ratios.
Integration into CapCut will include features like AI Video and generation tools such as Video Studio. The model will also extend to ByteDance’s AI generation platform, Dreamina, and its marketing platform, Pippit.
ByteDance has implemented safety restrictions to prevent the model from generating videos from images or videos containing real faces due to its realistic content generation capabilities. CapCut will also block the unauthorized generation of intellectual property.
Content produced by Dreamina Seedance 2.0 will contain an invisible watermark to identify AI-generated material when shared off-platform. This feature could assist in addressing takedown requests from rights holders, ByteDance added.
ByteDance plans to collaborate with experts and creative communities during the rollout to iteratively improve the model’s capabilities.
