Seedance 2.5 is live in FOTOhub

Summary

Seedance 2.5 has moved from "coming soon" to live. It is the first model in the Seedance family to generate thirty seconds of footage in a single pass with native audio, plan a full shot before rendering a frame of it, and take fifty separate references into one request. It sits in the video generator today alongside the full Seedance 2.0 lineup and every other model in FOTOhub's catalogue, ready to select like any other option.

What actually changed from Seedance 2.0

Seedance 2.0, still live and still the cheaper option per second of footage, tops out at fifteen seconds per generation and accepts nine images, three video clips, and three audio samples as references, fifteen in total. Seedance 2.5 doubles the clip length to thirty seconds and more than triples the reference budget to fifty: thirty images, ten video clips, and ten audio samples in one call. A cast of characters, a product, a location, and a sound palette can now be locked into a single request without anything drifting halfway through the take.

The bigger change is how the model gets there. Seedance 2.5 plans camera movement, lighting, and pacing across the entire thirty seconds before it renders anything, then executes in one pass, rather than stitching short clips together and hoping the seams hold. Multi-round extension carries that same continuity forward across several minutes of footage.

Two capabilities have no equivalent in the 2.0 generation. Audio is generated in the same pass as the picture, not layered on afterward, so sound and motion land in sync by construction. And timestamp-level editing lets you name the second that needs to change, describe the fix, and get back a revised clip, sound included, without re-rendering everything around it. That second capability changes how a production day gets planned when a client note comes in late.

Seedance 2.5 also introduces staging on untextured 3D blockouts: set camera position, character poses, and motion paths on plain geometry, then let the model dress the scene in a finished shot.

Who this is for

Marketing and creative teams producing ad spots that need a consistent product and cast from open to close. E-commerce brands generating product video where the item cannot change appearance between the wide shot and the close-up. Filmmakers and directors who want timestamp control instead of a random-output generator. Anyone currently spending more time stitching and re-prompting than shipping finished work.

Pricing and availability

Seedance 2.5 is available today across FOTOhub's paid plans through the same credit-based video generator every other model uses. Per-second credit pricing shows directly in the generator before a job is submitted, consistent with how the rest of the catalogue is priced, and we will publish the exact figure separately once usage data settles rather than lock in a number on day one.

Seedance 2.0 Pro, Fast, and Mini remain in the catalogue unchanged, with fifteen-second generations, native audio, multi-reference input, and 4K masters on the Pro tier. Nothing about them changes now that 2.5 is live, and 2.0 stays the lower-cost option per second of footage for anyone who does not need the extra length or reference count.

What comes next

FOTOhub will keep tracking Seedance's roadmap the same way it has tracked every model in the catalogue: add it the day access opens, confirm the real numbers once they are live, and leave the older tier running for anyone who prefers it. For now, Seedance 2.5 is one more option in the video generator you already know.

About FOTOhub. FOTOhub is a creative AI platform combining a multi-provider model catalogue across image, video, audio, text, and three-dimensional generation with a brand system, a workflow engine, and a documented API, exposed through official SDKs and a machine-callable tool layer.

Press contact. FOTOhub Press Team, FOTOhub.app

Learn more: Seedance 2.5 model page · plans and pricing · full model catalogue