Choose a category
3 models available
4 models available
3 models available
2 models available
4 models available
8 models available
Seedance 2.5 Image to Video
One still, up to 30 seconds of motion.
Creative Vision
Describe the motion, not the picture — the model can already see it.
Video Parameters
Priced per second — 30 seconds costs 7.5× what 4 does.
720p costs 2.25× what 480p does. Draft at 480p.
Adaptive keeps your image's shape — usually what you want.
AI audio synced to the video. Does not change the price.
Cinema Preview
Generation History0 records
Seedance 2.5 Image to Video
Upload one still and Seedance 2.5 moves it — up to 30 seconds from a single frame, at 480p or 720p, with audio generated in the same pass. The output keeps your image's shape by default, so nothing gets cropped to fit a template you did not choose.
Describe the motion, not the picture — the model can already see the picture
Why Choose Seedance 2.5 for Image to Video
One image and a sentence is very little direction for half a minute of video. What makes it work here is that the model plans the whole clip in one pass instead of drifting a few seconds at a time.
Up to 30 Seconds From One Frame
4 to 30 seconds, one second at a time. Long enough for a full camera move rather than the usual five-second nudge, and the subject stays recognisable as your original image throughout.
Your Image Sets the Shape
The aspect ratio defaults to Adaptive, which tells the model to match what you uploaded. A vertical photo produces a vertical clip. Override it only when a delivery format demands something else.
Audio Included, Free
Sound is generated alongside the picture in one pass and does not change the price. Ambience, effects and speech land on the action rather than being added over the top of it.
Reference Video Steers the Motion
Under Advanced, attach a clip of up to 30 seconds and the generation follows its movement — useful when you know exactly how the camera should behave. It also lowers the price by about 40%.
Reference Audio Drives the Performance
Attach up to 30 seconds of speech, singing or rhythm and the model builds the delivery around it. Combined with a portrait as the input frame, that is a specific person saying a specific line.
Failed Runs Are Refunded
Credits are held at submission and returned automatically if the generation fails or the provider never answers. A clip that never arrives costs you nothing.
One Frame In, a Whole Shot Out
If you also know how the shot should end, the First to Last Frame page takes both images and plans the path between them.
How to Use Seedance 2.5 Image to Video
The image decides what the shot looks like. Everything you type should be about what changes — movement, camera, light, sound.
Upload the Opening Frame
JPEG, PNG or WebP up to 30MB, one image. Use the composition you actually want in the video, because the output takes its shape from this file rather than from a ratio setting.
Write the Motion
"The camera drifts left as the steam rises and the neon flickers" beats re-describing what is already in the frame. Dialogue in quotation marks gets spoken. The box holds 20,000 characters.
Test at 480p, 4 Seconds
30 credits per attempt. This is where you learn whether the model reads your image the way you do — whether it thinks the subject is the person or the room behind them.
Deliver Longer and Larger
Raise the duration and switch to 720p for the final take. Generation takes a few minutes and you can leave the page; the finished clip appears in your history.
Chain Clips Together
Take the last frame of one generation, upload it as the opening frame of the next, and you can build a sequence a shot at a time without ever leaving this page.
Seedance 2.5 Image to Video
FAQ
Image requirements, aspect ratio, clip length, pricing and how this page differs from the other Seedance 2.5 modes.
It is ByteDance's Seedance 2.5 model driven by a single still: you upload one image, describe the motion, and it generates up to 30 seconds from that starting point with audio. FlowVeo3 is an independent platform and is not affiliated with ByteDance.
JPEG, PNG and WebP, up to 30MB. One image per run — this page is the single-frame mode. Higher-resolution stills give the model more detail to hold on to across a long clip.
Between 30 and 430 credits, set by resolution and length: 30 credits for 4 seconds at 480p, 75 for 5 seconds at 720p, and 430 for a full 30 seconds at 720p. Attaching a reference video under Advanced lowers all of those by about 40%. The exact cost is shown before you submit.
Your image's, by default. The ratio control is set to Adaptive, which tells the model to match the input instead of fitting it into a fixed shape. Pick 16:9, 9:16, 1:1, 4:3, 3:4 or 21:9 explicitly only when the delivery format requires it.
Not on this page — use Seedance 2.5 First to Last Frame, which takes a start image and an end image and plans the movement between them. It is the same model at the same price; the difference is how much of the shot you are pinning down.
Not here. This page animates one frame. For as many as 30 reference images called out as @Image1, @Image2 and so on in the prompt — a character from one, a set from another — use Seedance 2.5 Reference to Video.
Usually a few minutes, and longer clips take longer. You can leave the page while it runs — the result lands in your history and the credits are only kept if it succeeds.
You are not charged. Credits are held when you submit and released back to your balance if the model fails or the provider never returns a result. Generated videos stay available for seven days, so download anything you want to keep.
Try Seedance 2.5 Image to Video
One still, thirty seconds, with sound. Draft it for 30 credits.