16 with prompts
3D 与手工 lists 16 AI image prompts with the source image and author credit. Images that have no prompt yet are listed after them.
通用 / 未注明2将上传照片转化为一张盲压浮雕感图像，保持原图画幅、取景、主体数量、身份特征、姿态和相对位置。先把照片概括为清楚的外轮廓、主要体块与必要结构线，再把这些形状压印在一整张【纸色】厚棉纸上。主体与背景完全同色，不保留照片原来的色彩；整张纸面连续，图像由浅浅的压凸和压凹构成，凸起顶面平缓，边缘柔润，纸纤维细腻。将景深压缩为浅层起伏，背景简化为安静的纸面和少量必要轮廓，不做深洞或独立立体摆件。按主体自身结构保留辨识特征，简化细碎纹理，避免密集毛发、织物纹路和装饰抢走轮廓。左上方低角度柔和侧光掠过纸面，产生明亮边缘和右下方窄影，暗部仍可见纸纹。整体呈现精致、克制、有触感的无墨压印纸样摄影效果。仅输出转换后的单张完整图像，不做原图对比、分屏或拼贴，不加文字、边框、标志、水印，不出现剪纸叠片、石膏、陶瓷、金属、塑料或发光效果。
GPT Image 24Create a premium 4:5 vertical architectural travel artwork for [COUNTRY / CITY / LOCATION / LANDMARK]. TEXT-TO-IMAGE ONLY — generate the entire scene creatively from [COUNTRY / CITY / LOCATION / LANDMARK] alone. No reference image required. CORE CONCEPT Transform the destination into a charming handcrafted miniature world made entirely from knitted yarn, soft felt, embroidery, and plush textile materials. Reimagine its architecture, streets, nature, transportation, and local atmosphere as a sophisticated collectible textile diorama. LANDMARK & COMPOSITION Automatically select ONE unmistakable landmark representing [COUNTRY / CITY / LOCATION / LANDMARK] and make it the central hero. Build a complete miniature environment around it using authentic local architecture, streets, gardens, trees, shops, landscape, and a few tiny generic yarn characters. Keep the landmark’s recognizable silhouette and defining architectural features while translating everything into rounded handmade textile forms. MATERIAL LANGUAGE Every element should visibly feel handmade: - chunky knitted yarn - soft felt - embroidered details - woven fabric - stitched edges - wool fibers - plush textile surfaces - tiny handcrafted imperfections Architecture should look carefully constructed from layered fabric and yarn rather than plastic, clay, or CGI. ATMOSPHERE Create a warm, playful, cozy travel atmosphere with soft natural lighting, gentle shadows, tactile fibers, rounded forms, and subtle depth. Add small destination-specific details such as local food, transportation, plants, signs, or cultural objects where they naturally enhance the scene. Keep the background simple and softly colored so the miniature world remains the focus. STYLE Architectural travel poster × knitted miniature diorama × handmade textile art × plush yarn world × sophisticated editorial illustration. The result should feel like an elaborate handcrafted collectible discovered in a miniature museum — whimsical but visually refined and architecturally recognizable. COLOR Use a harmonious palette inspired by the destination itself, combining warm neutrals with a few distinctive local accent colors. Keep the colors soft, rich, and tactile rather than overly bright. Avoid photorealism, plastic materials, smooth CGI, generic architecture, unrelated landmarks, hard geometric edges, excessive clutter, realistic humans, logos, watermarks, and flat vector illustration. FORMAT 4:5 vertical, one cohesive miniature world, central landmark, rich textile detail, soft depth of field, premium handcrafted finish, highly detailed yarn and felt textures.
通用 / 未注明4为【城市地标／汉字／品牌图形／人物角色／食物】创作一张3:4竖版单幅“刺绣自生长”视觉。 一块暖象牙色天然亚麻布被绷紧在圆形木质绣绷中，四分之三俯视微距摄影。画面上方已经完成【核心图案】，中部处于正在缝制的生长边界，下方仍保留大片未经刺绣的空白布面。 一枚真实金属绣花针正从布面穿出，彩色丝线经过清楚可见的针孔与收紧线环，沿连续路径向前延伸。针经过的位置逐针形成轮廓、色块与纹理；尚未经过的位置不得提前出现完整图案。 根据结构使用长短针、缎面绣、锁链绣与法式结粒：轮廓采用锁链绣，主体色块采用方向一致的缎面绣，细小高光使用法式结粒。每根线都必须连接到真实落针点，具有合理张力、粗细、遮挡和接触阴影。 图案仍保持刺绣的平面属性，只允许局部线结形成轻微浮雕，不要让建筑、人物或花朵变成从布上升起的三维模型。 自然侧光照亮丝线光泽、针孔、布料经纬与木绷磨损。色彩控制在四至六种，核心图案最鲜明，生长边界保留少量松弛线圈，空白区域保持克制。 禁止成品刺绣平铺、普通毛线滤镜、无针孔悬浮线、线头凭空出现、多个绣绷、拼贴分屏、塑料材质、错误汉字、随机文字和杂乱线团。
GPT Image 23请根据【主题】创作一张高完成度、高质感、适合系列化发布的「地景档案模型 / Landscape Archive Model」视觉图。 这不是普通地图，也不是普通风景插画或城市效果图，而是一张结合“二维地图图纸 + 三维立体区域模型 + 专业说明系统 + 博物馆展陈感”的高端档案式视觉图。画面需要像一件被真实制作出来的地理研究模型、城市规划沙盘、建筑提案模型或世界观设定展品，具有专业、克制、精密、可触摸的高级质感。 【基础设定】 主题：【主题，例如火山口湖 / 广州 / 重庆山城 / 海岛城市 / 古城遗址 / 末日废墟 / 幻想大陆 / 游戏主城】 主题类型：【自然地貌 / 城市空间 / 混合景观】 核心主体：【核心主体，例如火山口、峡谷、山脉、湖泊、CBD、老城区、地标群、岛屿城市、遗址群、主城区域】 内容重点：【内容重点，例如地形高差 / 城市地标 / 路网结构 / 水系关系 / 自然与城市关系 / 设定集展示】 风格方向：【风格方向，例如地理测绘风、纸雕模型风、建筑提案风、博物馆档案风、城市规划风、游戏设定集风】 主色调：【主色调，例如米白、浅灰、羊皮纸白、沙色、雪白】 辅助色：【辅助色，例如土褐、岩灰、冷灰、淡蓝、苔绿】 点缀色：【点缀色，例如黑色细线、红色路线、蓝色水文线、黄色编号标记】 画幅比例：【画幅比例，例如16:9横版 / 3:4竖版 / 4:5竖版 / 1:1方图】 【画面结构】 画面主体是一块放置在干净桌面或白色布面上的矩形地图模型板。底板具有真实厚度，边缘可见剖面结构，像一块精致的地图切片或模型底座。地图表面覆盖清晰但克制的信息层，包括等高线、道路网络、水文线、分区边界、地理纹理、坐标感、轻微标注和图纸边框。 在地图中央或偏中心区域，从二维地图中立体“生长”出一个三维主体模型，成为整张图的视觉核心。 如果【主题类型】为自然地貌： 主体可由山体、峡谷、火山口、湖泊、盆地、岛屿、冰川、断层、海岸线等构成，强调地形高差、层叠等高线结构、岩壁纹理和自然地貌起伏。 如果【主题类型】为城市空间： 主体可由核心城区、CBD、老城街区、地标建筑群、滨水区、路网节点、桥梁、水系、公园绿地等构成，强调建筑体块、城市识别度、空间层次与规划结构。 如果【主题类型】为混合景观： 需同时整合自然地形与人工建成环境，例如山城、海岛城市、古城遗址、末日废墟、幻想主城、科幻基地等，让自然地貌与城市结构共同构成核心视觉。 【细节要求】 - 模型表面保留地图图纸感与专业信息感 - 地形区域应有等高线、坡面层次、凹陷与起伏关系 - 建筑区域应有体块感、街区关系和清晰层次 - 水域使用低饱和淡蓝色表现 - 绿地、森林、公园、山体可用苔绿色或低饱和绿色表现 - 路网、路径、交通轴线或探索线路可用细线表现，但不要喧宾夺主 - 若有地标建筑或重点结构，应具备明确识别度 - 若有幻想、科幻或游戏设定元素，应保留档案模型和图纸展示感，而不是做成纯场景插画 【说明系统】 地图板边缘需要设计完整而克制的说明模块，包括： - 外框线和内框线 - 比例尺 - 图例区 - 标题区 - 档案编号 - 索引标记 - 注释标签 - 简洁的符号说明 文字不必全部可读，但整体必须呈现真实、清晰、精密、有秩序的档案式排版结构。 【构图与镜头】 采用斜俯视等轴测或沙盘式构图，镜头从上方约30到45度观察整块模型板，让观众同时看到地图平面、立体主体高度、边缘厚度和说明区。视觉重心集中在中央主体模型，阅读路径为：先看核心主体，再看周边地形/路网/水系/街区信息，最后看图例、比例尺和说明区。 【材质与光影】 整体需要具有真实模型摄影质感： - 纸张纹理 - 地图印刷质感 - 纸雕或沙盘模型感 - 建筑模型或地形切片质感 - 柔和自然光 - 细腻真实阴影 - 干净背景 - 低饱和、高级、克制 避免卡通化、廉价游戏UI感、过度杂乱、过度炫技，重点突出“地图底板 + 立体主体 + 专业说明系统”的统一视觉。 最终效果应像一张可以用于地理科普、城市研究、建筑提案、文旅视觉专题、世界观设定集或高端系列封面的地景档案模型图。
通用 / 未注明3折纸艺术，真实纸张质感，几何折痕，细节丰富，线条锐利，对称构图，【大熊猫】，颜色丰富栩栩如生，【森林】背景，柔和自然光，电影级景深效果。
GPT Image 24请生成一张竖版 2:3 高清图片，主题为《建筑微缩蓝图》。 建筑名称：[填写具体建筑] 画面是一张现代建筑设计手稿，上面布满中文平面图、剖面图、立面图、结构细节图、尺寸标注和施工说明。一座高度仿真的中国古建筑微缩模型，从图纸中立体“建”出来。 主建筑要占画面 65%–75%，更大、更高、更有视觉冲击力。使用低角度 3/4 俯视视角，突出模型高度、体积感和真实阴影。建筑要符合所选主题的真实特征，并呈现正在施工中的状态，完成度约 70%–85%。 加入微缩施工细节：施工小人、脚手架、木架、坡道、滑轮、木梁、石块、瓦片、材料堆、小推车、马车、半完成台阶、屋顶施工区等。细节要围绕建筑本身展开，形成一个仿真的微缩建造世界。 整体风格是现代建筑事务所概念提案板 + 博物馆级微缩模型。图纸干净高级，颜色以米白、浅灰、淡蓝、灰黑线稿为主。所有文字必须是中文，主标题写“[建筑名称]”，副标题写“建筑营造过程示意图”。 使用柔和中性光线，画面清晰、现代、克制、有高级感。避免复古黄、羊皮纸、古董感、杂乱工作室、书本、蜡烛、卷轴、尺子、指南针、桌面摆件、现代机械、现代车辆和无关背景。 最终效果：让一座中国古建筑像从现代设计图纸里真实建造出来，兼具建筑学专业感、微缩模型趣味和强烈视觉冲击力。
GPT Image 2Eye-level straight-on view, looking directly into a hyper-vibrant 3D layered paper cut-out diorama. Sakura blush pink, snow white, deep indigo sky and pale celadon accents, extremely high saturation, PHYSICAL PAPER LAYERS clearly visible with hard cut edges and thick paper cross-sections at every depth plane. Maximalist composition, crowded Mount Fuji cherry blossom scene — Mount Fuji as a monumental white cone spanning all background depth planes, layered cherry blossom tree canopies receding toward the mountain in every mid-ground plane, every branch loaded with dense pink blossom clusters, traditional wooden torii gate threading between tree layers, a wooden bridge over koi pond in mid-ground layer, paper-cut falling petal storm frozen mid-air across all depth planes, cranes in flight between upper layers, stone lanterns and bamboo fence along pathway layers, deep carpet of fallen petals and koi fish reflections crowding the serene foreground. Soft single-direction studio lighting creating deep drop shadows between every physical paper layer, matte paper texture throughout, NO glow, NO soft edges, NO volumetric lighting, octane render, 8k, clean details --ar 3:4
Nano Banana4You are a scientific historian + visual prompt engineer.
INPUT:
{科学发现或发明}
TASK:
Automatically infer and determine, with high confidence:
- The correct scientific discovery name (Chinese + English)
- The primary discoverer(s) full name(s)
- The approximate year or historical period of discovery
- The most iconic origin moment of this discovery
- The historical location where it occurred
- The core scientific idea, expressed in ONE simple Chinese sentence
- Suitable visual scientific symbols (equations, structures, metaphors)
Then OUTPUT a SINGLE, FINAL image-generation prompt in ENGLISH.
CRITICAL RULES:
- The discoverer’s name MUST be explicitly written and clearly visible on the right page of the book
- The discoverer must appear as a realistic 3D human figure in the left-page diorama
- Do NOT include reasoning or explanations
- Do NOT show variables or placeholders in the output
- Output ONLY the final image prompt
FINAL IMAGE PROMPT SPECIFICATION:
Scene:
A hyper-realistic 4:5 macro photograph of an ancient scientific codex opened flat on a dark velvet museum plinth. The book features cracked leather binding, gilded edges, thick handmade parchment, and extreme material realism.
Left Page — “The Living Discovery”:
The left page is hollowed into a deep 3D diorama carved into the book.
Inside the page is a photorealistic miniature scene depicting the iconic historical moment of the discovery.
A lifelike, realistic 3D human figure of the discoverer is shown mid-action, wearing period-accurate clothing, with natural facial expression and body language.
The environment precisely matches the historical time and location.
Subtle glowing scientific symbols visually represent the core concept.
A warm amber glow emanates from within the pages.
Right Page — “The Scientific Record”:
A flat aged parchment page with visible paper fibers and raised ink texture.
At the top, clearly write:
- Discovery name (Chinese + English)
- Discoverer name(s) (clearly labeled as “发现者”)
- Discovery year or period
In the center:
Write ONE clear, simple Chinese sentence explaining the scientific discovery, suitable for general audiences.
At the bottom:
Include a small antique-style map or diagram indicating the discovery location, styled like a historical scientific annotation.
All text must be written in legible, elegant Chinese handwriting, restrained and scholarly.
Style & Camera:
Museum specimen photography.
85mm macro lens.
Shallow depth of field.
Cool gallery spotlights combined with warm internal glow.
Extreme realism in leather, paper texture, ink embossing, dust, and miniature details.
Output:
ONE single image.
Aspect ratio 4:5.
Ultra-high resolution.
Photorealistic, cinematic.
No illustration, no cartoon style.
Nano Banana4[角色] = 路飞 木质书桌上摆着一件充满奇趣的手工纸艺作品：将手绘的卡通 [角色] 从纸上剪下，折叠成一个立体的站立小人偶。[角色] [姿势]。在它身旁的纸面上，不仅有那个留下的空白剪影，还手写着一句它的 [经典台词]。 画面比例：1080 x 1080
Midjourney4--sref 4283169172
通用 / 未注明4A paper cut-out collage of a [subject], assembled from layered torn paper with visible fibers and textured edges. Set in a handcrafted diorama made of colored cardstock, stitched borders, and crumpled sky elements, the scene feels playful, raw, and full of analog charm like a page from an imaginative craft book come to life.
GPT Image 24This GPT acts as a digital painter who generates detailed 3D-style weather forecast illustrations for any city input by the user. When a user inputs a city name, the assistant fetches the current live weather data for that city from weather action with an English address(make sure to transform the address to english address, e.g. 北京 -> Beijing, China), and then transforms it into an artistic 3D isometric painting, no need to print any other information. Generate a vertical composition image with a clear top-down 45° isometric view of a miniature model featuring iconic architecture from the city [City Name]. Incorporate the current weather condition ([Weather Condition]) gently and creatively into the scene, demonstrating soft interaction between the weather elements and the urban landscape. Use physically-based rendering (PBR) techniques with realistic lighting and subtle shadows. Employ a minimalistic, solid-color background for clarity and simplicity, with the city model centrally composed to highlight the precise, intricate beauty of the 3D model. Clearly display the city name "[City Name]", current weather condition "[Weather Condition]", and temperature "[Temperature]°C" at the top of the image, alongside a matching weather icon. The assistant always responds in the language used by the user when naming the city, and aims to reflect the city's character in each unique image.
通用 / 未注明生成12张适合幼儿英语启蒙的3D卡通记忆卡片图片，每张展示一个“A开头”（必须遵守）单词及其可爱生动的插图，使用明亮色彩和圆润造型，字体大而清晰，不要出现任何中文，设计符合儿童认知特点。
GPT Image 1 / 4oTurn the subject in the attached image into a kawaii enamel pin. Use glossy metal outlines and vibrant enamel fill. No extra added features. Square mockup format. White background.
GPT Image 1 / 4o4将一个简单平面的矢量图标 [💩] 转化为柔软、立体、毛茸茸的可爱物体。整体造型被浓密的毛发完全覆盖，毛发质感极其真实，带有柔和的阴影。物体居中悬浮于干净的浅灰色背景中，轻盈漂浮。整体风格超现实，富有触感和现代感，带来舒适和俏皮的视觉感受。采用摄影棚级灯光，高分辨率渲染，比例为1:1。
GPT Image 24微型立体场景呈现，运用移轴摄影的技法，呈现出Q版【孙悟空大闹天宫】场景
2Sin prompt
4Sin prompt
Sin prompt
Sin prompt
Sin prompt
4Sin prompt