
Original document
GPT Image 2 · en
{
"type": "Anime Character Design Collection",
"character": {
"name_japanese": "{argument name=\"character name\" default=\"Rei Ayanami\"}",
"name_english": "{argument name=\"english name\" default=\"REI AYANAMI\"}",
"hair": "{argument name=\"hair color\" default=\"light blue\"} shortwave head",
"eyes": "{argument name=\"eye color\" default=\"red\"}",
"outfit": "{argument name=\"outfit\" default=\"Blue vest dress, white short-sleeved shirt, red bow tie, black over-the-knee socks, white flat shoes\"}"
},
"layout": {
"header": {
"position": "Top left corner",
"elements": [
"Japanese name",
"English name",
"A Japanese descriptive text"
]
},
"turnaround": {
"position": "right side",
"count": 3,
"poses": [
"front",
"side",
"back"
]
},
"sections": [
{
"title": "Facial expressions",
"position": "Middle left",
"count": 6,
"description": "Avatars with different expressions: neutral, slight smile, surprised, blinking, open mouth, head down."
},
{
"title": "Detailed display",
"position": "Lower left",
"count": 4,
"items": [
"Red bow",
"Uniform upper body",
"Blue shoulder strap",
"small blue bag"
]
},
{
"title": "Color scheme",
"position": "bottom left corner",
"count": 5,
"items": [
"Red bow",
"blue vest dress",
"Mini ID Card",
"Black socks",
"Four circular color blocks (blue, light blue, white, and black)."
]
},
{
"title": "NERV ID Card",
"position": "bottom right corner",
"count": 3,
"items": [
"White flats",
"keychain",
"NERV identification tag on the lanyard"
]
}
]
},
"style": "Official anime concept art, cel-shaded style, clean lines, white background with light blue geometric decorations."
}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.





