Assign the photo to the person, and the video to the motion
A useful Money Flex prompt separates appearance from action. The photo identifies the person: their face, skin tone, hair and visible clothing. The video provides where that person stands, how the hands move, when the bundles approach the lens and when the bills open into a fan.
EasyVideoAI already makes that assignment in its prepared template. It asks for one consistent person from the first frame through the final face occlusion, without carrying over the source person's hair or black jacket. The portrait background and pose should not become the bedroom performance.
Open the Money Flex AI studio · Follow the one-photo Money Flex tutorial
What the cash-flex scene direction asks to preserve
In plain language, the direction is: replace the foreground person using the photo, follow the original cash gestures and timing, keep the bedroom and props, and maintain the same identity when the money crosses the face. This is an explanation of the existing template, not a second prompt field or another generation mode.
The low camera angle, dresser, bed, curtains, lighting and cash chain belong to the scene. The final fan should still meet the hands naturally. The template asks for coherent fingers and sensible occlusion; it cannot guarantee that every note or finger will remain perfect in fast movement.
| Part | Reference to check |
|---|---|
| Face, hair and outfit | Your single person photo |
| Bundle handling and final fan | The prepared video performance |
| Bedroom and camera angle | The source scene |
| Audio | The matching private source track restored after generation |
Choose a face that remains clear before the bills cover it
Give the model a sharp, well-lit face with the eyes and facial outline visible. Clothing in the photo helps define the outfit. A tightly cropped face leaves more of the clothes for the model to complete, while a very distant person can leave too little facial detail.
The cash fan hides part of the face near the ending. That occlusion is part of the action, so removing it would change the scene. Compare identity before the fan rises, above the bills and during any turn. A different expression is a separate issue from a genuinely different face.
Check the failure before changing another input
A result can preserve the face but distort a hand, or preserve the gesture while losing the identity in a close-up. Name the problem first. Repeatedly uploading the same photo or starting another generation does not guarantee a correction.
If the cash is cropped away, inspect the frame selection. If the face changes mainly behind the bills, compare the visible portions through that occlusion. If the downloaded video is silent, check the actual MP4 and saved request; extra music instructions are not the audio control in this workflow.
Can I paste a different Money Flex prompt?
This studio uses a prepared prompt and does not expose a free-text prompt editor. Choose the matching scene and upload the person you want to cast.
Does a money-spread prompt create a different template?
Not here. Money spread describes the fan at the end of this cash-flex scene. A text-to-video prompt elsewhere can produce a different action and timing.
Does the portrait supply the sound?
No. It supplies appearance. The finished clip receives the source audio, not a voice recorded from the photo.
Review the cost before a new attempt · Check identity and motion problems
Reviewed against the EasyVideoAI studio on October 30, 2026. Check the live studio for current availability and the final credit quote.