On November 12, Volcano Engine launched the video generation model Seedance 2.5 Flash, positioned as a cost-effective version whose results are currently the closest to the flagship Seedance 2.5. It doubles the generation speed and is aimed at batch content production. For text-to-video and image-to-video at 720P, for example, the generation price is about RMB 0.99 per second, roughly 34% lower than Seedance 2.5.
It supports video generation of up to 30 seconds per clip, accepts up to 10 images, 5 videos, and 5 audio clips in a single input, and offers 480P and 720P resolution options. Starting today, it will gradually roll out on platforms including Dreamina, LibTV, Updream, Creao AI, and TapNow. The API will go live on Volcano Ark this Sunday (November 15).
According to the official announcement, Seedance 2.5 Flash delivers better overall results than the previous-generation flagship Seedance 2.0. It maintains a high standard in instruction following, physical motion, and audio performance, with natural, textured visuals. Its video editing capabilities support producing multiple versions around the same creative idea.
Using the same telescope ad prompt for comparison, the two finished videos are close in quality: Seedance 2.5 presents a more complete close-up of the boy adjusting the telescope, while Flash focuses more on coherently completing the action sequence of "putting his eye to it — lowering the telescope — smiling in response." The original article also includes examples such as a one-take circus video and a 30-second, 14-shot headphone ad, pointing to batch production of film, TV, and advertising assets.
Since the release of Seedance 2.5, the team has introduced a Draft sample mode, 1080P, and an upcoming 4K resolution. This time it has added the Flash member, forming a dual-tier lineup of flagship and Flash: the flagship Seedance 2.5 represents the current highest level of intelligence and, together with the Draft mode's "choose a sample first, then generate the final video" workflow, reduces the high-definition generation cost of repeated trial and error, serving 1080P and 4K production; Flash handles high-frequency creation and batch production with near-flagship results, faster speed, and lower cost.
