While OpenAI appears to be scaling back its involvement in the video generation sector with the discontinuation of its Sora app, ByteDance officially announced on Thursday the phased rollout of its new audio and video model, Dreamina Seedance 2.0, within its popular editing platform, CapCut.
ByteDance states that this advanced model empowers creators to conceptualize, modify, and synchronize video and audio content by utilizing textual prompts, still images, or existing reference videos.
The initial phase of this rollout will target CapCut users in Brazil, Indonesia, Malaysia, Mexico, the Philippines, Thailand, and Vietnam, with plans for subsequent expansion into additional global markets over time.
This launch in CapCut follows recent reports indicating a temporary pause in the model's broader global deployment. That decision was made to address intellectual property concerns and criticisms from Hollywood regarding alleged copyright infringement, which likely accounts for the current limited geographic availability of the model within CapCut.
Within China, the model is already accessible to users of ByteDance’s Jianying application.
In its official announcement, ByteDance emphasized the video generation model's ability to operate even without reference images, allowing creators to describe their desired scene with just a few words. The company also highlighted CapCut’s proficiency in rendering realistic textures, nuanced movements, and intricate lighting across various visual perspectives and angles, noting its potential for editing, enhancing, or correcting creators' original footage.
Another compelling use case involves enabling creators to test and refine potential ideas based on early concepts or sketches, prior to commencing actual video production.
Furthermore, Dreamina Seedance 2.0 is designed for a diverse array of content, including cooking recipes, fitness tutorials, business or product overviews, and notably, videos featuring motion or action-focused content—areas where AI video models have historically encountered significant challenges, as explained by the company.
At its launch, the model supports video clips with a maximum duration of 15 seconds and is compatible with six different aspect ratios.
The model's integration within CapCut will extend across various functionalities, encompassing editing features such as AI Video and generation tools like Video Studio. It is also slated for deployment on ByteDance’s dedicated AI generation platform, Dreamina, and its marketing platform, Pippit.
Given its capability to produce highly realistic content, ByteDance has implemented stringent safety restrictions. These measures prevent the model from generating videos from images or existing video content that contains real human faces. Additionally, CapCut will block the unauthorized generation of intellectual property. (However, the model's current unavailability in markets such as the United States suggests that further refinements to these restrictions are likely still in progress.)
ByteDance also confirmed that content produced by Dreamina Seedance 2.0 will incorporate an invisible watermark. This feature is designed to aid in identifying AI-generated content when it is shared off-platform, potentially assisting rights holders with takedown requests should any copyrighted material inadvertently be created by the model.
As the model rolls out, ByteDance has committed to forming partnerships with experts and creative communities to continuously iterate upon and enhance the model’s capabilities.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.