Full GPT Image2 Prompt Example

Use this section when you need the complete GPT Image2 prompt wording, not just the summary. Review the structure, copy the full GPT Image2 prompt as-is, and then replace the subject, brand, style, or use case to fit your own workflow. This full GPT Image2 prompt block is the best place to study how the prompt is assembled before you test your own version. Treat this GPT Image2 prompt section as the source text you refine step by step.

A trendy young woman sitting at an outdoor café table, holding a hot coffee cup near her lips. She has short curly black hair and wears oversized tinted sunglasses, a cropped beige sweater, loose brown high-waisted pants, and chunky white sneakers. A tote bag with minimal coffee-themed illustrations hangs on her shoulder. The scene blends realistic photography with playful 2D cartoon overlays. Floating around her are animated coffee elements: a smiling cappuccino cup with heart-shaped latte art, cute coffee beans with tiny faces, and stylized splashes of caramel forming comic-style shapes. Soft doodle steam rises into hearts and stars. On the table: a croissant, a notebook with sketch drawings, and a second coffee cup. Background: a realistic cozy café street with brick walls, plants, warm golden hour lighting, and soft shadows. Slight depth of field, subject in sharp focus. Style: mix of photorealism and vibrant cartoon illustration, pop-art aesthetic, high saturation but warm tones, clean composition, Instagram-worthy, soft glow, 4k detail.

Refine This Prompt

Why this prompt works

It defines the core subject and headline, specifies typography style, lighting, texture, colors, and composition to guide the model toward a cinematic, high-contrast poster with strong readability.

What you can change

  • Title text and tagline
  • Color palette and lighting mood
  • Typography style and weight
  • Background environment or texture
  • Aspect ratio and composition

Reusable template

Create a premium editorial poster for the headline "[TITLE]" with [TYPOGRAPHY STYLE], [LIGHTING STYLE], [COLOR PALETTE], [TEXTURE], [COMPOSITION], and [MOOD/ATMOSPHERE].

How to adapt this GPT Image2 prompt

Use this GPT Image2 prompt as a working reference, not just a one-off sample. Keep the structure that already defines the output clearly, then replace the subject, typography, environment, campaign context, or aspect ratio to match your own goal. If you need more examples before rewriting the GPT Image2 prompt, go back to the Realistic People category page or continue into the related prompts below. This GPT Image2 prompt workflow works best when you keep the original structure visible while you rewrite each key variable. A stable GPT Image2 prompt baseline usually makes revision work much faster.

FAQ about this prompt example

How should I use the 咖啡馆写实照片与 2D 涂鸦叠加 prompt example?

Start by copying the full GPT Image2 prompt, then replace the subject, style language, and output goal with your own brief. Keep the visual structure if you want results in the same direction, and reuse the GPT Image2 prompt skeleton before you add brand-specific details. In practice, this GPT Image2 prompt works best when you change a small number of variables per iteration.

What can I change without losing the original structure?

You can change the headline, product, mood, lighting, composition, typography, and end use while keeping the core GPT Image2 prompt logic that makes the example useful. In most cases, the best result comes from editing one GPT Image2 prompt variable at a time instead of rewriting everything at once. That keeps the GPT Image2 prompt structure stable while you test new creative directions.

Continue browsing

Related Realistic People prompt examples

Explore more from the realistic people collection to compare GPT Image2 prompt structures, visual directions, and reusable ideas that stay close to this example. Open another GPT Image2 prompt example if you want a nearby variation before rewriting your final version. That extra GPT Image2 prompt comparison step often helps you refine the final wording with less guesswork.

红跑道低角度夏日人像
Realistic People

红跑道低角度夏日人像

Use the uploaded portrait photo as the appearance reference for the person. Create a bright photorealistic outdoor portrait of a young woman lying on a red running track on a modern white arch pedestrian bridge. Ultra-wide low-angle selfie perspective, her arm reaching toward the camera in the foreground, relaxed pose, wired headphones around her neck, white sleeveless top, loose gray pants, black hair spread on the ground. Clean blue sky with soft white clouds, strong midday sunlight, crisp shadows, high clarity, fresh youthful mood, architectural symmetry, realistic skin texture, cinematic composition, 3:4 vertical image Negative Prompt: watermark, logo, text, caption, signature, AI label, extra fingers, deformed hands, distorted face, wrong identity, duplicate person, blurry face, low resolution, over-smoothed skin, plastic skin, unnatural anatomy, bad perspective, messy background, harsh artifacts, overexposure, underexposure

Shinning1010Portrait
Scrapbook 真人图与迷你分身
Realistic People

Scrapbook 真人图与迷你分身

Transform the provided reference image into a cozy aesthetic scrapbook-style composition while strictly preserving the original subject, identity, pose, lighting, and background. Add multiple small “mini version” characters of the same person (chibi / doll-like style), placed naturally around the scene (on objects, table, shoulder, etc.). These mini figures must match the subject’s face, hairstyle, outfit, and vibe consistently, styled as cute 3D collectible figurines. Show them doing different activities (reading, posing, taking photos, relaxing). Overlay handwritten-style doodles and annotations across the image: arrows, hearts, stars, sparkles, icons, and playful captions connected to elements in the scene. Use a soft pastel color palette (white base with pink, peach, blue accents). Keep the frame visually rich and filled but balanced and clean. Style: warm, cozy lighting, dreamy Instagram scrapbook aesthetic, soft depth of field, highly detailed, polished but playful. The final result must look like the SAME original image enhanced with mini alter-egos and aesthetic annotations — not a recreated or different scene.

Kashberg\_0Portrait
韩国便利店粉色 Hoodie 人像
Realistic People

韩国便利店粉色 Hoodie 人像

Ultra-realistic cozy Korean convenience store portrait of a beautiful Korean woman standing in front of glowing refrigerator aisles at night, wearing an oversized fluffy pastel pink hoodie with the hood up. She holds a bottle of strawberry milk in one hand and a tiny strawberry cake in the other while shyly looking toward the camera. Long soft black hair, glossy eyes, natural skin texture, subtle blushy makeup, gentle expression, youthful Korean beauty aesthetic. Warm fluorescent convenience store lighting mixed with realistic iPhone flash photography. Tiny reflections on drink bottles and glass refrigerator doors, dreamy romantic atmosphere, soft pink and cream color palette, slice-of-life anime realism blended with cinematic photography, highly detailed, cozy late-night Seoul convenience store vibe, shallow depth of field, realistic Korean snack packaging, candid aesthetic, soft glow, ultra photorealistic.

john_my07Portrait
运动轨迹舞者光绘海报
Theme Poster

运动轨迹舞者光绘海报

<instructions> input = 3 legendary dancer/choreographer input = ai inferred signature movement sequence function render_kinesphere ($ dancer, $ movement) anchor: [figure composed entirely of motion trails] :: [the $ movement sequence traced as continuous luminous bands that coalesce into a portrait of $ dancer]::5 morphology: the arabesque trails weave the skull, a grand jeté arc forms the cheekbone, the arms' sweep creates the jaw, stillness is implied by negative spaces::4 material physics: blackened silver gelatin print, faint chemical residue, slight motion blur, glowing filament traces like a Marey wheel photograph::3 illumination: complete darkness with bioluminescent-like light trails, no external light sources, high contrast::2 render stack: chronophotography, fine art dance study, 8k, kinetic sculpture concept::1 negative: [still body parts, studio lights, solid silhouettes]::-1 </instructions>

GdgtifyPortrait
东方神话人物志百科海报
Infographic

东方神话人物志百科海报

请基于【孙悟空】生成一张竖版 A4「东方神话人物志」百科海报。 你是一名东方神话视觉设定师、古籍图谱设计师和中文信息图设计师。请根据【人物名称】在东方神话、民间传说、古籍、戏曲、文学或传统文化中的已知形象，生成一张内容完整、视觉统一、具有东方神话气质的人物档案海报。 最重要原则： 全部内容必须围绕【人物名称】展开，不要套用通用神仙模板。 人物形象、法器、坐骑、神通、典故、象征物、背景场景都必须符合【人物名称】的神话设定。 如果该人物存在多个传说版本，请采用最广为人知的版本；存在异文时，可以在相关板块简短标注“传说版本不一”。 不确定的信息不要强行编造，可写“民间传说中多有异文”或“相关记载较少”。 不要使用北欧符文、维京图腾、哥特边框、西方盔甲、欧式城堡、西幻恶魔角、赛博朋克元素。 整体必须是东方神话审美，而不是“西幻人物换中式衣服”。 整体视觉风格： 东方神话史诗感，融合《山海经》异兽图谱、敦煌壁画、青绿山水、唐代人物画、宋元古画、青铜器纹样、古籍注疏和符箓篆刻风格。画面庄重、神秘、古雅、华丽但克制。 画面构图： 竖版 A4 海报，中轴对称布局。中央为【人物名称】的全身或半身主视觉，姿态庄严，有神性和叙事感。人物服饰、发冠、法器、神态、灵兽、背景都要符合该人物传说。左右两侧为信息卡片，底部为核心总结。四周使用云雷纹、回纹、青铜纹、祥云纹、卷草纹、山海纹、篆刻印章式边框。 主标题：“【人物名称】” 副标题：“东方神话人物志｜根据该人物神话传说生成” 顶部精神题记：请根据【人物名称】的核心精神，自动生成一句古雅、凝练、有哲学意味的题记。题记要体现该人物背后的价值观、生命姿态或精神内核，而不是说明性文字。不要写“非史实考证”“仅供参考”“传说版本”等免责声明。字数控制在 12-28 字，适合放在标题下方。题记应与人物高度绑定，不要套用“守护苍生”“大道无边”等空泛句式。 请生成以下内容板块： 【1. 神格身份】说明【人物名称】在东方神话体系中的身份定位。 【2. 神话源流】说明该人物主要出自哪些神话、古籍、民间传说、小说、戏曲或宗教信仰。 【3. 司职权柄】总结该人物掌管或象征的领域。 【4. 形象特征】描述该人物的典型外貌、服饰、神态、姿态和视觉气质。 【5. 法器神通】展示该人物最具代表性的法器、兵器、符咒、神通或能力。 【6. 灵兽坐骑 / 随身象征】如果【人物名称】有坐骑、灵兽、伴生生灵，请展示其名称、形态和寓意。如果没有明确坐骑，请改为“随身象征物”或“相关意象”。不要强行给每个人物安排龙、凤、仙鹤或麒麟。 【7. 典故出处】列出 2-4 个与【人物名称】相关的经典故事、传说片段或文化典故。不要虚构不存在的典故。若版本不一，请标注“异文较多”。 【8. 象征意象】提炼该人物最核心的视觉符号。符号必须与【人物名称】相关。 【9. 神话谱系】说明该人物与其他神话人物、阵营、族属或体系的关系。如果谱系不明确，请写“谱系传说不一”。 【10. 文化影响】说明该人物在民俗、节日、祭祀、文学、戏曲、绘画、庙宇、影视或现代文化中的影响。 【11. 精神内核】从神话叙事角度总结该人物象征的精神。这里只分析神话形象，不推断真实人物性格。 【12. 核心总结】用一段古雅但易懂的中文总结【人物名称】：“【人物名称】在东方神话中象征……，其形象融合了……，代表着……。” 视觉版式要求： 顶部：大标题 + 副标题 + 精神题记。 中央：人物主视觉，气势庄重，细节丰富。 左侧栏目：神格身份、神话源流、司职权柄、形象特征、法器神通。 右侧栏目：灵兽坐骑/随身象征、典故出处、象征意象、神话谱系、文化影响。 底部：精神内核 + 核心总结。 可加入小型图标、注释线、古籍标签、卷轴卡片、青铜铭牌、朱砂印章、符箓纹样。 信息卡片像古书页、玉简、青铜牌、卷轴或碑铭，不要像现代科技 UI。 材质与纹样： 宣纸肌理、矿物颜料、鎏金线条、朱砂印章、墨色晕染、青铜器铭文、云雷纹、回纹、祥云纹、山水纹、敦煌藻井纹样。 配色规则： 根据【人物名称】的属性自动选择东方配色。火焰 / 战斗 / 护法：朱砂、玄黑、鎏金、赤红。月亮 / 水系 / 清冷神祇：黛青、银白、月白、石蓝。山川 / 木系 / 自然神灵：石绿、青黛、玉白、赭石。帝王 / 天庭 / 尊神：鎏金、玄黑、朱砂、玉白。妖灵 / 山海异兽：墨黑、铜绿、暗金、赭红。不要所有人物都使用同一套配色。 文字要求： 中文必须清晰可读，标题大气，正文简短准确。不要乱码、伪文字、错别字、文字重叠或裁切。如果空间不足，优先压缩正文，但保留所有板块标题。 避免： 北欧符文、维京风、哥特边框、西方盔甲、欧式城堡、魔幻游戏 UI、赛博朋克、现代科技感、过度暗黑、随机龙凤、无关法器、模板化神仙形象、与【人物名称】无关的典故、空泛精神题记。

TanLuAIPortrait