Seedance 2.5 creates a clip from a written description or from one starting image, with whole-second lengths from 4 to 30 seconds and three quality tiers.
There are two starting points. Text to video builds a shot from your description; image to video animates one uploaded still and keeps its framing as the first frame. Both run in the same workspace with the same length and quality controls.
Length is the practical reason to pick this model. A 30-second ceiling gives a sequence room to play out in one request instead of being stitched together from shorter clips afterwards.
Length, quality tier, shape and sound are set before you generate. When a starting image defines the first frame, the output shape follows that image automatically, so the shape selector is not offered in that mode.
Set any whole number of seconds between 4 and 30, so one continuous action can stay in a single clip.
Write the shot from scratch, or upload one still to define the first frame and let the model move outward from it.
Render silent clips for edits that carry their own audio, or switch sound on so the clip arrives with atmosphere included.
Describe a camera move around a product and let a single long clip carry the whole sequence.
Establish a location with atmosphere and optional sound before cutting to tighter shots made with other models.
Bring an existing still to life while keeping its framing, because the output follows the uploaded image.
Use a button on this page. It opens the right workspace with this model already selected, in the language you are reading now.
Write what should happen in the shot or image. Concrete subjects, framing and action give the model more to work with than short labels.
Choose length or quality tier, shape and sound where offered, then read the credit quote for those exact settings before you generate.
Generation uses credits from your account balance. Video models are charged per generated second at the quality tier you choose, and image models per image. The exact amount is quoted in the workspace before you confirm, and the pricing page lists the current packages.
See credit packagesBetween 4 and 30 seconds, in whole seconds. You set the exact length in the workspace and the quote updates with it.
No. On this platform the model takes one starting image in image-to-video mode. If you need a multi-image reference workflow, use a model that exposes that mode.
No. Sound is optional for this model, so you can render a silent clip or switch sound on before generating.
480p, 720p and 1080p. Higher tiers cost more credits per second; the exact amount for your settings is shown before you confirm.
When a still defines the first frame, the output follows that still automatically. Choose a shape in text-to-video mode instead if you need a specific one.
Other models on this platform that do a similar job, or that pair well with this one.
Wan 3.0 is the standard variant of the Wan 3.0 family on this platform: a wide length range, three quality tiers and shape control in both text and image mode.
Read the guideMiniMax H3 turns a description or a single starting image into a clip of 4 to 15 seconds that carries a generated sound track.
Read the guidePixVerse V6 Max renders clips from a single second to fifteen, across four quality tiers, from a description or from one uploaded still.
Read the guide