TikTok expands Symphony video generation with Dreamina Seedance updates
TikTok has expanded the video-generation capabilities available through Symphony, its suite of AI-powered ad creation tools. The updates include access to newer Dreamina Seedance models, more room for reference materials, longer generated video clips, an updated API and wider access to BytePlus tokens.
The changes give advertisers additional ways to shape AI-generated creative within Symphony Agent, TikTok’s tool for developing ad content. In particular, the platform now supports as many as 50 multimodal references per prompt, up from nine. Those inputs can include images, video and audio, allowing advertisers to supply a broader set of examples when directing generation.
Generated videos can now run for up to 30 seconds per prompt, twice the previous maximum of 15 seconds. TikTok also said the latest Dreamina Seedance model can produce sharper, more realistic imagery, with improved lighting and greater consistency in characters across frames. These are TikTok’s descriptions of the model’s capabilities; the practical results will depend on the creative inputs and generated output.
More references and longer clips
Expanded reference limits could give teams more latitude to communicate a desired look or provide examples for a campaign. The longer duration may also allow more of a concept to be developed in a single generated clip. Neither change removes the need for advertisers to assess whether generated material is suitable for a particular placement, brand or campaign objective.
TikTok’s announcement positions the updates as additional creative options rather than a guarantee that every output will be usable. Advertisers considering the tools can test how different references and prompts affect the resulting videos, then judge the output against their own requirements. The source material does not provide performance results or indicate how adoption has changed following the updates.
API and token access changes
TikTok has also introduced an updated Symphony API. It is intended to let advertisers generate content using Symphony’s tools from within their own creative management tools. That integration could make the generation capabilities available in workflows advertisers already use, although TikTok’s announcement does not specify which external systems are supported or provide further implementation details.
A separate change broadens access to BytePlus tokens, which can be used to access Symphony’s visual-generation tools. Starting Nov. 9, 2026, advertisers outside North America can purchase tokens to access Dreamina Seedance models through Symphony, TikTok said. The company associated the token access with faster video generation, higher-quality output and more formats.
BytePlus is the enterprise technology arm of ByteDance, TikTok’s parent company, and manages the company’s AI tools and systems. The expanded token availability therefore adds a purchasing route for eligible advertisers beyond the model and workflow updates themselves.
Together, the changes extend the range of inputs, duration and integration options available for AI-assisted ad creation on TikTok. For marketers, the immediate question is how well the additional controls translate into creative that meets campaign needs. Testing the tools may help teams evaluate that fit, but the announcement does not establish that AI-generated video will be appropriate for every advertiser or use case.
