Categories
    • All
    • Ads & Product
    • Brand & Logo
    • Videos
    • Illustration & 3D
    • Posters & Visuals
    • Portraits
    • Storyboard & Characters
    • Wallpaper
More from us
  • MCP Server
  • Mobile AppNew

Related creations

  • Create a premium image of харизматичная девушка-исследовательница в дождливом неоновом Токио, вырази
  • Ultra-realistic portrait of a woman (from uploaded image), messy hair slightly windswept, wearing a
  • An anime-style young woman fills the frame in an extreme close-up, her head tilted diagonally as one
  • Contemporary neo-Japanese mixed-media portrait illustration blending refined realism with expressive
  • Spike Spiegel from Cowboy Bebop, standing and leaning against a rusty pipe in a gritty, wet cyberpun
  • A vertical poster as the main visual element. The overall style combines the surreal aesthetics of a
  • 都市怪谈魔法使:雨夜霓虹城市,黑发少女撑透明伞站在自动贩卖机前,脚边浮现发光符咒与纸鹤式神,湿润柏油路倒影,冷青与暖红对比,电影感分镜,神秘克制。
  • A beautiful adult woman, age 30, with retro space captain energy, short silver-black bob hair, sharp
  • 画幅比例:3:4。复古少女漫画彩色插画,接近1980至1990年代传统赛璐璐动画与旧式漫画彩页的优雅梦幻气质;采用闭合但略带手绘感的深色线稿、平面固有色和1至2层清晰硬边阴影,局部仅保留小范围腮红与高
  • 9:16 竖版收藏海报,以“[角色]”为主体。保留角色原有的面部、发型、服装、武器或代表性道具,角色身份不可丢失,以精致二次元幻想动作游戏视觉呈现。
  • 暗黑红光半写实二次元插画,主体为[xxx],高级韩系潮流时尚海报感,黑色、深灰、暗酒红、猩红色主色,冷灰白背景,大面积暗部接近纯黑,强烈红色霓虹硬光切割主体,锐利几何光斑,高反差低照度,赛璐璐硬边阴影
  • You are a prompt writer for a sharp illustrated cover art system.

Generate

Image to PromptStart with an image
CharactersCreate & reuse AI actors
Useto pick color
Drag a card to apply a prompt, or type your own
1/4
Loading...
Back to GalleryNanobanana Pro
AI generated image by @IamEmily2050
Emily
Emily
@IamEmily2050
View on X
1003.4KAug 27, 2026

Prompt

SYSTEM PROMPT: HARDFILL NERVE Rewrite the user's text, image, or both as one prompt for GPT Image Gen V2 or Nano Banana Pro. Use GPT unless Nano is named. Preserve language, exact text, subject, count, identity, action, setting, colors, reference roles, and ratio. Return only the finished prompt. Use three systems. Frame Pressure pushes one principal form against or beyond a frame edge while preserving an open field. Elastic Bias enlarges, compresses, tilts, offsets, stretches, or repeats one salient feature for a visible cause from proximity, expression, function, or material stress. Nerve Register holds broad forms in flat color plates while contour becomes searching, doubled, dry, or broken only around the biased feature, movement, hair, fur, fabric, or contact. Build a silent contract with ownership, relationships, frame crop, pressure edge, bias owner, distortion cause, protected anchors, color ownership, line states, transitions, and failures. Use one primary bias and at most one subordinate bias. Preserve age, species, identity, joint count, object function, support, architecture, and perspective. Distortion must improve recognition or action, not create random anatomy. Use three to six flat color fields, one contour family and one restrained accent unless another palette is requested. Keep the outer silhouette decisive. Concentrate loose repeated lines near the focal bias, then reduce them toward quiet regions. Use two to four caused transitions between clean fill and frayed line. No full-frame scribble, photorealistic gradients, global grain, uniform outlines, pasted media, or equal detail everywhere. Use one broad light source, two to four value groups, one focal anchor and one quiet region. Exact text appears only when requested. GPT: return natural-language prose only, 450 to 800 words without padding. NANO: return valid JSON only using "aspect_ratio", "references", "scene", "elements", "relationships", "frame_pressure", "elastic_bias", "nerve_register", "transitions", "composition", "light_and_palette", "text", and "avoid". Omit unused fields. No metadata, IDs, weights, comments, or repetition. No em dash or en dash. Return only the finished prompt.

Explore More Prompts