Free tool · Official formula · Reference syntax built in
Seedance prompt generator
Built on the structure ByteDance's own prompt guide teaches, with the multi-modal reference syntax one tap away. See 24 official examples with their output videos, or plan a whole sequence with the Shot Planner.
Multi-modal reference syntax (tap to insert)
Seedance accepts up to 9 images, 3 videos and 3 audio files as references — number them in the prompt exactly as uploaded.
Your prompt
Fill the blocks above — they combine into one prompt.
Frequently asked questions
- What's the three-block Seedance prompt formula?
- ByteDance's official guide structures prompts as: subject & action first, then environment/lighting/style, then camera movement & audio — written as one flowing paragraph. Stacking information in that order gives the model the cleanest signal. This builder enforces the order and merges your blocks automatically.
- How does the reference syntax work?
- Seedance's signature capability is multi-modal referencing: up to 9 images, 3 videos and 3 audio files per generation. In the prompt you point at them by number — 'the character from Image 1', 'the camera movement from Video 1'. The tap-to-insert chips add the exact phrasing.
- Will these prompts work on Seedance 2.5?
- Yes — 2.5 keeps the same prompt conventions and raises the ceiling to 30-second single takes. Track the release on our Seedance 2.5 page, and when it goes live you can run your prompt directly in our generator (first clip free).