How it works
- 01
Choose one dog photo
Use a sharp, permitted image with one dog, visible eyes and ears, and a simple background.
- 02
Crop around the head and chest
Keep the dog centered and leave space below the face for a counter and one prop.
- 03
Request one bartender action
Describe one simple drink-making gesture and a steady camera, then review the paid-credit quote before submitting.
Use one clear subject
Choose one dog in even light with the face and both ears visible. A head-and-chest frame leaves room for a counter and a single prop without asking the model to separate multiple animals.
Keep the action short and safe
Describe one harmless action such as placing a glass on a counter while the dog stays behind it. Avoid dangerous ingredients, flames or actions that imply the animal handled real food or tools.
Compare your options
| Option | Best for | Keep in mind |
|---|---|---|
| Image to video | Turning one permitted dog photo into a short original action. | Four seconds, 720p, square or 16:9; fur, paws and props can change. |
| Crop | Centering the dog's face and leaving room for a counter before animation. | Cropping prepares the still image but does not animate it. |
Questions, answered
Does this copy the external multi-drink template?
No. PromptToVisual uses one first image and a text motion prompt; it does not copy a reference video or provide the external fifteen-second sequence.
What format does PromptToVisual produce?
The current form produces a four-second MP4 with audio at the 720p preset in square or 16:9 for 40 paid credits.