ByteDance brings Dreamina Seedance 2.0 AI video model to CapCut

While OpenAI appears to be scaling back its push in the video generation space following the shutdown of its Sora app, ByteDance confirmed on Thursday that its latest audio and video model, Dreamina Seedance 2.0, is now being introduced within its editing platform, CapCut.

According to ByteDance, the new model enables creators to generate, edit, and synchronize video and audio content using prompts, images, or reference clips.

The rollout will happen in phases, beginning with users in Brazil, Indonesia, Malaysia, Mexico, the Philippines, Thailand, and Vietnam, with additional markets expected to be included over time.

The introduction of Dreamina Seedance 2.0 into CapCut follows a recent report suggesting that the model’s broader global release had been paused while the company addressed intellectual property concerns. These concerns stemmed from criticism by Hollywood stakeholders regarding potential copyright issues. This likely explains why the model is initially being made available only in select regions.

Within China, the model is already accessible through ByteDance’s Jianying app.

ByteDance stated that creators can describe a scene using just a few words, and the model will generate corresponding visuals. It is also capable of rendering detailed textures, realistic movement, and accurate lighting across various angles and perspectives. These capabilities can be used to edit, refine, or enhance existing footage. Another practical application of the model is allowing creators to experiment with ideas at an early stage. Users can generate sample visuals from concepts or sketches before proceeding with full production.

Dreamina Seedance 2.0 is designed to support a wide range of content types, including cooking demonstrations, fitness tutorials, product showcases, and action-oriented videos. In this area, AI video models have traditionally faced limitations.

At launch, the model supports video clips up to 15 seconds long and works across six aspect ratios.

Within CapCut, the model will be integrated into multiple features, including AI Video editing tools and generation functions such as Video Studio. It will also be available across ByteDance’s broader ecosystem, including its AI creation platform, Dreamina, and its marketing platform, Pippit.

Given the model’s ability to produce highly realistic content, ByteDance has implemented safety measures. The system is restricted from generating videos using images or footage that contain real human faces, and it blocks unauthorized use of intellectual property. However, the limited rollout suggests that further refinements may still be underway before wider availability, including in markets like the United States.

Content created with Dreamina Seedance 2.0 will also include an invisible watermark, designed to help identify AI-generated material when it is shared outside the platform. This feature may assist rights holders in tracking content and requesting its removal if copyright issues arise.

ByteDance stated that it will continue to collaborate with experts and creative communities throughout the rollout to refine and improve the model’s performance and capabilities.