Does the jisoo addicted ai prompt work?
A Jisoo Addicted AI prompt has to keep two real faces, two separate bodies, a fixed hallway and a six-beat routine all correct at once, and a pasted prompt rarely does. On Jisoo Addicted AI it is already written and tested. Make the video from your photos.
What the prompt has to solve
- Keep each person looking like their own photo.
- Keep the two people distinct while they lean and swap sides.
- Land each move on the music: six beats in 15 seconds.
- Hold one hallway for the whole clip.
Which model runs it
Jisoo Addicted AI renders with ByteDance's Seedance video models, in four tiers: Seedance 2.0 Mini, Seedance 2.0 Fast, Seedance 2.0, Seedance 2.5. The same built-in setup runs on every tier, so you never choose or edit model settings.
Do you need to write one?
No. Each look carries its own tested prompt and motion reference, so you only upload photos and choose a format. A prompt pasted from elsewhere usually loses the faces or the timing because it lacks that reference, which is covered in why a pasted prompt fails. We do not publish ours.
Which tier to pick
Standard at 720p (10 credits) is the usual starting point. If a face drifts, render again on Pro or Ultra before changing anything else. Every render follows the same shot list, which you can read on the dance page, and the trend sound is mixed in after rendering. Full costs are on the pricing page.
Independent fan-made tool. Not affiliated with, endorsed by or sponsored by JISOO, ROSÉ, BLACKPINK, their agencies or record labels, or TikTok.