Up to 30 seconds per clip
One generation runs from 4 to 30 seconds, long enough for several camera cuts inside a single scene.
ByteDance's Seedance 2.5 turns a prompt, a first and last frame, or a set of reference images, clips and sounds into video of up to 30 seconds with sound. On Vgent you use it from the same composer as every other model and pay per second of output.
Examples Vgent made. Works people publish to the gallery appear here once approved.
One generation runs from 4 to 30 seconds, long enough for several camera cuts inside a single scene.
Omni reference takes up to 30 images, 10 video clips and 10 audio clips, so a character, a product or a movement stays the same from shot to shot.
Ambience, effects and spoken lines are generated with the picture; turn sound off when you plan your own mix.
Continue a clip past its last frame, or edit a video from your asset library in the Editor.
Credits are charged by seconds of output at the verified rate. They are held when you submit, and a failed generation releases them.
Each clip runs 4 to 30 seconds. For anything longer, the Agent plans a film from many clips and joins them into one video.
480p, 720p and 1080p, in 16:9, 9:16, 1:1, 4:3, 3:4 or 21:9, or an automatic ratio that follows your images.
Add photos or designs of the character as omni references. In the Agent, every film first gets reference sheets for its characters, places and props, and each shot is generated from them.
By seconds of output and by resolution. The price example on this page uses the rates in force now, and the composer shows the exact credits before you generate.
Vgent 1.0, Vgent's default model, runs on Seedance 2.0 and makes clips of 4 to 15 seconds. Seedance 2.5 goes up to 30 seconds and takes more reference material.
Sign up with Google, then pick the plan that fits how much you make.