← Back to prompt library
GPT Image 2 · Image
Japanese Gpt Image2 Thumbnail
A reusable GPT Image 2 image prompt for creative generation in AutoAGC workflows.
842Views
36Likes
14Copies
ImageFormat
Generation prompt
{"type":"日文 YouTube 缩略图拼贴","style":"大胆的动漫科技宣传缩略图，采用简洁的数字插画、高对比度、高饱和度色彩、粗轮廓字体、充满活力的杂志风格构图","canvas":{"aspect_ratio":"16:9","resolution":"1200x630"},"background":{"frame":"薄荷绿外边框，顶部和底部边缘带有青色装饰条","main_scene":"温馨的现代卧室室内，配有木质床头、米色床上用品、奶油色窗帘、床头架和一个小闹钟"},"layout":{"sections":[{"title":"数据制作!","position":"top-left","count":8,"labels":["背面视图","四分之三背面视图","正面视图","全身背面站立视图","侧面轮廓","后侧面轮廓","四分之三正面视图","交叉双臂站立肖像"]},{"title":"局部修正!","position":"bottom-left","count":1,"labels":["角色编辑应用截图，展示身穿白色夹克内搭蓝色衬衫的动漫女性"]},{"title":"素材制作!","position":"lower-mid-left","count":1,"labels":["智能手机素材设计图，展示一只手拿着紫色屏幕手机，屏幕上有巨大的眼睛符号和一列 5 个小缩略图"]},{"title":"合成!","position":"upper-center-right","count":2,"labels":["动漫女性剪影肖像","卧室背景底板"]}],"connectors":{"count":2,"style":"带有白色轮廓的粉色曲线箭头，从剪影肖像和卧室底板指向右侧的最终合成图"},"text_overlays":{"count":7,"items":["数据制作!","局部修正!","素材制作!","合成!","「」彻底验证!","从去噪方法到 ","©"]}},"subjects":{"main":{"type":"动漫女性","count":1,"appearance":{"age":"年轻成年人","build":"苗条曲线美","hair":{"color":"","style":"非常长的直发，带有柔和光泽，中分，发丝垂在肩上"},"face":"面部故意模糊/遮挡矩形","outfit":["白色修身长袖圆领上衣","可见黑色吊带","黑色长裤"]},"pose":"坐在床边，躯干微向左倾，双手放在膝盖附近，看向镜头"},"inset_character":{"type":"相同的动漫女性参考图","count":10,"notes":"所有插图人物均使用相同的角色设计，多个面板中面部模糊处理"}},"typography":{"primary":"底部非常大的日文标题，白色字体配黑色轮廓，GPT Image2 周围有亮粉色强调","secondary":"副标题采用橙色、白色、黑色和亮蓝色混合强调","annotation":"每个插图部分附近有手写风格的活力粉色标签，带有白色描边"},"color_palette":{"count":8,"colors":["亮粉色","白色","黑色","青色","薄荷绿","橙色","米色","深棕色"]},"composition":"右半部分由最终的动漫卧室插画占据，左半部分填充了插图过程示例和截图，超大的底部标题横跨几乎整个宽度，旨在为关于 GPT Image2 工作流的深度验证文章做广告，涵盖数据制作、局部编辑、素材生成、合成、去噪和 LoRA 训练"}
青虎AI workflow
Turn it into a reusable workflow
- Copy the prompt text and keep the role, task, model, style, and output constraints.
- Paste it into GPT Image 2 or a comparable model with your reference material or workflow brief.
- Replace the final constraints for output format, brand voice, language, and execution objective.
Asset
Resource preview
- Asset type
- Image
- Model
- GPT Image 2
- Scene
- 广告
More prompts
More GPT Image 2 system prompts
Browse workflows