GPT Image 2AI 提示词
Anime Character Design Sheet — Customizable Prompt
Create a professional anime character reference sheet with full-body turnaround, expression variants, and detailed outfit breakdowns.

提示词 · EN
type
anime character reference sheetcharacter
hair
{argument name="hair color" default="blonde"}eyes
{argument name="eye color" default="blue"}outfit
{argument name="uniform style" default="military-style dress with white trim"}main_color
{argument name="uniform main color" default="blue"}layout
top_section
description
Full body turnaroundcount
3views
item 1
frontitem 2
sideitem 3
backbottom_left_top
description
Facial expressionscount
6expressions
item 1
neutralitem 2
winking smileitem 3
open mouth smileitem 4
slight smileitem 5
shyitem 6
gentle smilebottom_left_bottom
description
Accessories and detailscount
5items
item 1
beretitem 2
leg with black tights and blue heel with fin attachmentitem 3
blue and white ribbon bowitem 4
upper uniform torsoitem 5
pair of glovesbottom_right_top
description
Outfit breakdowncount
3items
item 1
front coat and ruffled skirtitem 2
back of jacketitem 3
back of skirtbottom_right_middle
description
Additional details and color paletteitems
item 1
pair of blue heelsitem 2
single glovecolor_palette_count
8bottom_right_bottom
description
Text boxlabel
SETTING & NOTESstyle
clean anime art style, white background with subtle horizontal lines, decorative vintage borders关于这个提示词
Create a professional anime character reference sheet with full-body turnaround, expression variants, and detailed outfit breakdowns. 它更适合作为 GPT Image 2 的概念艺术起点:先保留画面结构、主体关系和风格约束,再替换成你的品牌、人物或场景。
这条提示词使用结构化 JSON 组织信息,包含 type、character、layout、style 等字段。保留这种层级能让模型更清楚地区分画面主题、布局、界面元素和细节约束。
使用时建议先小幅修改主体、场景、镜头和色调,再生成多个版本对比构图与细节。这样页面内容对用户有实际帮助,也避免把模型名或标签机械堆在正文里。
如何使用这个提示词
- Copy the prompt template from the page.
- Choose your preferred AI model (e.g., Sora 2, Midjourney).
- Customize variables like hair color, eye color, and uniform style to match your vision.
- Paste the prompt into your AI tool and generate.
- Refine the output by tweaking variables or regenerating for different results.



