PROMPT RADAR / IMAGE平面设计

个人资料 / 头像 - Mona-lisa-1 Cyberpunk Profile Avatar

Generates a circular cyberpunk profile portrait with a clean white title header, useful for social media comparison graphics or model showcase posts.

@Rooko_ai 0 热度 0 浏览
个人资料 / 头像 - Mona-lisa-1 Cyberpunk Profile AvatarORIGINAL VISUAL / 来源见文末
编辑先看

它为什么值得进入灵感库

先别急着照抄整段 Prompt:真正有用的是主体、空间和镜头之间的约束关系。 先规定信息层级和文字安全区,再谈视觉风格;无法落版的漂亮画面不是完整设计。

完整视觉 Prompt

Create a clean social-media profile style image on a vertical 3:4 canvas with a white header band at the top and a large circular avatar image below. In the header, center the large black sans-serif title text {argument name="title text" default="Mona-lisa-1"}, using a modern rounded font, with generous white space. Below it, place one perfectly circular image crop against a black surrounding background, like a Twitter/X profile picture preview. Inside the circle, show a cinematic photorealistic cyberpunk portrait of {argument name="character description" default="a handsome young man with wavy dark brown hair, light stubble, sharp cheekbones, and a calm confident expression"}, framed from chest to head, slightly turned toward the viewer, wearing {argument name="outfit" default="a black futuristic high-collar leather jacket with subtle tactical straps"}. The background inside the circle is a dense rainy neon megacity at night with tall skyscrapers, blue and red lighting, glowing windows, atmospheric haze, and a small hovering futuristic vehicle on the right. Include exactly 3 visible text elements: 1) the top header title “Mona-lisa-1”, 2) a red vertical building sign on the left reading {argument name="left sign text" default="BUILD BETTER WORLDS TOGETHER //"}, and 3) a blue sign on the right reading {argument name="right sign text" default="TECH PEOPLE IMPACT //"}. Use high-detail realistic lighting, dramatic rim light, red highlights on the left side of the jacket, cool blue light on the right side of the face, shallow depth of field, and a premium sci-fi movie-poster aesthetic. Keep the circle crop obvious and accurate, with the portrait and city fully contained inside the circular mask; avoid extra logos, watermarks, or additional text.
查看原作者内容

不要一次改完所有变量

用 GPT Image 2 复刻时,先固定主体和构图,再逐次替换光线、色调或材质中的一个变量,这样更容易判断哪项描述真正起作用。

  1. 主体:Generates a circular cyberpunk profile portrait with a clean white title header, useful for social media comparison graphics or model showcase posts.
  2. 视觉方向:社区实测摄影提示词;按原提示词的光线描述执行
  3. 镜头与构图:按原提示词的镜头与景别执行;按原提示词的构图与画幅执行
  4. 交付约束:auto,遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。
负向约束

遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。

模型建议

GPT Image 2 / Grok Imagine / Midjourney / Flux。先用原始 Prompt 建立基准结果,再根据模型对自然语言、镜头运动或文字排版的能力做局部改写。

来源与编辑原则

保留作者,也保留判断。

本页收录公开案例并提供中文索引、结构化拆解和复刻建议,不冒充原作者作品,也不替代原帖上下文。

作者:Rooko.ai · 收录于 2026年8月14日

访问原始来源

沿着这条线继续看