
Original document
GPT Image 2 · en
{
"type": "Japanese movie trailer visual collage",
"typography": {
"title": "{argument name=\"movie title\" default=\"Tail of the Evening Calm\"}",
"subtitle": "—To you from that day—",
"tagline": "{argument name=\"tagline text\" default=\"What's lost can never be recovered. Yet, by connecting with someone on an emotional level, people can look forward again.\"}"
},
"characters": {
"girl": "Young Japanese woman, dark hair, wearing a sailor-style school uniform.",
"man": "Middle-aged Japanese man with stubble, wearing a dark blue shirt.",
"animal": "{argument name=\"dog breed\" default=\"Shiba Inu\"}"
},
"aesthetics": {
"setting": "{argument name=\"setting\" default=\"Seaside town at sunset\"}",
"style": "{argument name=\"color grading\" default=\"Nostalgic movie feel\"}"
},
"layout": {
"panel_count": 8,
"sections": [
{
"position": "Larger image on the left",
"description": "A girl pets a dog; an old house; text superimposed."
},
{
"position": "Top right corner",
"description": "A man's profile as he gazes at the sea."
},
{
"position": "Top right corner",
"description": "The girl rested on the table."
},
{
"position": "Middle right",
"description": "The girl and the man sat at the table, with a dog between them, and the ocean visible through the window."
},
{
"position": "bottom left corner",
"description": "A man walks his dog on the beach."
},
{
"position": "bottom left corner",
"description": "Close-up of a girl hugging a dog"
},
{
"position": "bottom right corner",
"description": "The girl and dog are facing away from the camera, looking towards the sunset."
},
{
"position": "bottom right corner",
"description": "A dog walking on a country street"
}
]
}
}How to use
- Copy the complete original prompt, including any parameters.
- Replace the subject, text and composition with your own requirements.
- Use a compatible external tool; check its current access conditions and review the result.
Before you start
- Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
- Keep names, figures and visible text accurate; results can vary between model versions.





