Short answer: Put the subject and non-negotiable continuity rules first, then describe only the actions that change over time. The prompt limit depends on the selected model. On September 28, 2026, this site's text-to-video interface showed 30,000 characters for Seedance 2.5 and 7,000 for Seedance 2.5 Lite. Recheck the live counter before submitting; the limit is not a target prompt length.
What belongs in the prompt
Use four blocks: subject and setting, continuity rules, timed action, and only the exclusions that prevent likely failures. The goal is an unambiguous order of operations, not filling every available character.
A compact prompt formula
Create a [duration]-second [aspect ratio] video of [subject] in [setting].
Keep [three important identity or continuity details] consistent.
[Time range 1]: [opening composition and action].
[Time range 2]: [one visible action and camera movement].
[Time range 3]: [ending action and final hold].
Use [reference label] only for [defined role]. No [likely unwanted additions].
This is an illustrative template, not a tested result. Reattach the same references for each generation and use the labels shown in the interface.
What to cut first when a prompt is too long
Remove repeated adjectives before removing useful instructions. Combine related continuity details, keep one camera movement per beat, and exclude only the most likely unwanted additions. Do not delete the ending; the final frame often determines whether a clip is usable.
Example: a focused product brief
This is illustrative and not a measured generation result:
Create a 10-second 16:9 product video of the travel mug in @Image1 on a light
wooden table beside a window. Keep its shape, lid, handle, color, and visible
markings consistent with @Image1. Use soft morning light.
0–3s: stable medium shot; the mug is centered and still.
3–7s: one slow left-to-right camera move while the mug remains fully visible.
7–10s: ease to a stop, keep the mug centered, and hold the final frame.
Preserve markings already visible in @Image1, but add no new text or logos. No
hands, extra products, zoom, orbit, camera cut, or scene change.
Match detail to settings
Select the model and settings before polishing the wording. Check the live character counter, duration control, aspect ratio, resolution, and credit estimate. A concise prompt helps reviewability but does not guarantee exact timing or motion.
Check before generating
Confirm that each time range has one main action, references have clear roles, exclusions do not contradict requested content, and the final frame is described. If one part fails, revise that part instead of adding more style words. Keep a simple log of prompt version, settings, and the specific failure.
For the workflow, see the Seedance 2.5 model overview, Seedance 2.5 Audio to Video, and Seedance 2.5 prompt examples.
FAQ
Is a longer prompt better?
Not automatically. Repetition and conflicting instructions can make the intended action harder to identify.
Should I put the prompt in one paragraph?
No. Short labeled blocks and time ranges make contradictions easier to catch.
Does the character limit guarantee exact output timing?
No. Use an external editor when the final cut needs precise alignment.
