Awesome AI Art Pics Prompts 🎨
🌐 https://aiart.pics/ | English
🍌 简介
🎉 全网最大的 Nano Banana 提示词库！ 汇聚 900+ 位全球创作者 的 3200+ 精选案例，持续更新中...
欢迎来到 AI Art Prompts 精选集！这是目前最全面的 Gemini Nano Banana / Nano Banana Pro 提示词与生成案例合集。
我们从 X (Twitter)、小红书等全球社交媒体平台精心收集了 3000+ 高质量提示词，涵盖摄影、插画、3D 渲染、UI 设计、创意海报等 20+ 个专业领域，全方位展示 Google 图像生成模型的无限可能。每个案例都经过人工筛选，确保质量和实用性。
💡 为什么选择这个项目？
- 🏆 规模最大 - 3200+ 提示词，900+ 位创作者，持续增长
- ✅ 质量保证 - 所有案例都是真实生成结果，附带原图展示
- 🎯 即用即学 - 每个提示词都可直接复制使用，快速上手
- � 每日更新 - 自动化收集最新优质案例，保持时效性
- � 双语支持 - 中英文双语界面，服务全球用户
- 🔗 完整溯源 - 每个案例都附带原始来源，尊重创作者
⭐ 如果这个项目对你有帮助，请点击右上角 Star 支持我们！你的 Star 是我们持续更新的动力！
📦 数据来源
本项目的提示词主要来源于：
- 🐦 X (Twitter) - 全球 AI 艺术创作者分享
- 📕 小红书 - 中文社区优质内容
- 🌐 其他平台 - Discord、Reddit、Facebook 等社区
[!TIP] 💡 在线体验 Nano Banana Pro：https://gemini.google.com
[!NOTE] 📅 最后更新: 2026-01-01 | 新增 5233 个提示词
[!IMPORTANT] ⭐ 如果这个项目对你有帮助，请点击右上角 Star 支持我们！
[!WARNING] ⚠️ 版权声明: 所有提示词均来自社区，仅供学习交流使用。如有侵权，请提交 Issue，我们将立即删除。
🅰️nanto Mohammad
特斯拉FSD工作原理图解
来源: X
Nano Banana Pro prompt: “Make an infographic that explains how Tesla FSD works”
Only the Tesla AI team can verify its accuracy. https://t.co/RKHJPkp6NS
イーサン・マスク
Nano Banana Pro轻松制作LINE表情包，外包需数万？
作者: @イーサン・マスク
来源: X
Nano Banana ProでLINEスタンプも一瞬で作れた。外注したら数万かかるよね？スタンプ作りたい人はこのプロンプト使ってみて。
プロンプト
添付画像のキャラクターをちびキャラにして。LINEスタンプを作成。表情・ポーズ別・セリフ入で12種類。 https://t.co/BPYBLy5RIB
Framer 🇱🇹
角色周边制作：随机Nano Banana用法
作者: @Framer 🇱🇹
来源: X
One of the most random Nano Banana use cases I’ve found is asking it to create merch for your character.
PROMPT: create merch with this character https://t.co/qbUDjrCH57
INK
旧艺术作品的再次创作
作者: @INK
来源: X
one more rework of my old artwork with Nano Banana pro and Kling 2.6
DuckTales vibes
将新工作流程应用于旧艺术作品
作者: @INK
来源: X
love applying my new workflow to my old artworks
Freepik spaces × Nano Banana pro × Kling 2.6
使用Freepik Spaces和Nano Banana Pro的笔触人物设计
作者: @INK
来源: X
grab a pen
character design using Freepik Spaces and Nano Banana Pro
workflow:
> draw on paper
> apply your 3D style
> build the 3D character
> animate it with Kling 2.6
两斤
盗梦空间
作者: @两斤
来源: X
盗梦空间
Banana Prompt👇
---
{current_weather} in a gravity-defying {city_name}. The asphalt street curves seamlessly upwards, physically bending 90 degrees to become a vertical wall, then arching directly overhead. Not a reflection, but a solid continuous loop of architecture. Iconic buildings and roads protrude perpendicularly from the curved surface all around. {city_name}'s landmarks hang upside down above. Realistic lighting matching the weather, vivid details, photorealistic 8k. --ar 9:16
纸雕立体拼贴
作者: @两斤
来源: X
纸雕何必是平面的呢
---
A four-panel photo split collage arranged in a 2x2 grid, presenting four fully layered 3D paper cut-out dioramas.
GLOBAL STYLE (crucial constraint): Every single element in the scene, from the background hills to the central characters and objects, must be constructed from multiple stacked layers of cut matte cardstock. The edges of these paper layers must be visible on all objects, defining their form and volume through layering, not solid sculpting. Soft, warm directional studio lighting casts realistic drop shadows, revealing the thickness and depth of the stacked paper layers everywhere.
Top Left Panel: A forest scene where oversized mushrooms and tiny foxes are built from stacked paper layers.
Top Right Panel: A terraced village where houses and hills are formed by stepped layers of paper.
Bottom Left Panel: Two small figures and a boat made of layered paper, floating on water made of concentric paper layers with layered lanterns.
Bottom Right Panel: A character and a large castle, both constructed from stacked paper layers, on a layered paper cliff.
COLOR PALETTE: Warm, cozy tones of moss green, amber yellow, terracotta orange, deep teal, and creamy white. Octane render, 8k, palpable layered texture. --ar 1:1 --stylize 300
城市微切片
作者: @两斤
来源: X
感谢 @aestheticsguyy 的灵感。
透过世界的某个街角，将一座城市的温度凝缩进这个精致的微观切片之中。
用户只需填入{city_name}即可
Banana Prompt👇
---
A hyper-realistic vertical tilt-shift photograph of a cross-section of a typical building in {city_name} during {current_time}. The current weather is {current_weather}. The scene looks like an intricate miniature diorama. The lighting and sky color accurately reflect the {current_time}, while the street textures and atmosphere visually demonstrate the {current_weather} conditions. At street level, the environment is lifelike. Tiny detailed figures are visible going about their activities. Strong miniature effect with heavy blur at the top and bottom. 8k resolution, volumetric lighting. --ar 9:16
无缝四分屏纸张合成图像
作者: @两斤
来源: X
调了一下午，都快给我调崩溃了
---
A seamless 16:9 full-frame composite image divided into four distinct quadrants. **CRITICAL:** The four sections touch each other directly. NO GREY BACKGROUND. NO BORDERS. NO GAPS. The image is filled edge-to-edge with paper details.
**Composition:** A split-screen view looking deep into four different paper worlds.
1) **Top-Left (Luckin):** A wide view of a magical paper forest. Blue and white paper clouds and coffee leaves fill the quadrant completely. A white paper stag leaps through the density.
2) **Top-Right (Google):** A wide view of a **"Topographic Map" style paper cavern**. The "G" logo is formed by **hundreds of thin sheets** of Blue, Red, Yellow, and Green paper stacked horizontally. It looks like a sediment canyon or a contour map. Visible paper fibers and sharp cut edges. NO plastic smoothness.
3) **Bottom-Left (Coke):** A wide view of a red paper explosion. The Coke bottle is a **Negative Space Silhouette**—a deep hole void cut through dense layers of red paper. White paper ribbons span across the void.
4) **Bottom-Right (McDonald's):** A wide view of a yellow paper city. Vertical strips of yellow paper (french fries) stand densely packing the frame. The Golden Arches rise from deep within the layers.
**Global Material:** Everything is matte cardstock.
**Lighting:** Soft, even studio lighting revealing the depth of the paper cuts.
**Tech:** Octane render, 8k, --ar 16:9 --stylize 400 --no borders, frames, background wall, isolated objects, 3d plastic, seamless texture
极简雾感石版画
作者: @两斤
来源: X
感谢原作者 @tisch_eins 的灵感，我好像更喜欢这种风格！
极简构图 + 细腻胶片感 + 一丝中国水墨画的意境
Gemini说这是“Foggy Minimalist Lithograph” (雾感极简石版画风)
Banana Prompt👇
---
A four-panel grid featuring minimalist atmospheric art. The style is faded monochrome photography on high-quality matte paper.Texture & Tone:
Subtle film grain (not heavy noise).
Soft, creamy beige and desaturated grey tones (clean, not dirty).
Low contrast, dreamy, hazy atmosphere.The 4 Panels:
(Top Left): Layered misty mountains with a tiny, sharp silhouette of a modern city at the very bottom.
(Top Right): A lone traveler walking on a smooth sand dune into the fog. Clean lines, vast negative space.
(Bottom Left): The Shanghai skyline (Lujiazui) rising faintly from heavy river mist, ethereal and ghost-like.
(Bottom Right): Abstract zen geometry: concentric water ripples under a large pale sun, with a lone distant skyscraper.Negative Prompt:torn paper, dirt, stains, grunge, scratches, high contrast, harsh lines, blurry, messy, distorted.--ar 1:1 --v 6.0 --style raw
大胆鲜艳的撞色城市纸雕立体模型
作者: @两斤
来源: X
还是喜欢这种大胆鲜艳的撞色
---
1/
Eye-level straight-on view, a hyper-vibrant 3D layered paper cut-out diorama of SHANGHAI city.
The scene showcases iconic Shanghai landmarks: **The Bund waterfront with historic buildings**, **traditional Shikumen alleyways**, and **futuristic Lujiazui skyscrapers (Oriental Pearl Tower, Shanghai Tower)**.
Color Palette: **High contrast palette of Vivid Shanghai Red, Jade Green, and Matte Gold.**
Style & Texture: Playful stylized shapes with smooth curves and clean edges. Material texture is matte cardstock (no gloss).
Physics of Depth: Soft, even studio lighting creates **distinct, realistic deep drop shadows between each individual paper layer**, revealing tangible depth and physical thickness of the matte paper.
Composition: Maximalist, full bleed composition, bursting with intricate details of street life, tiny paper people, and traditional lanterns.
Details: Octane render, 8k, clean details, surreal realism. --ar 3:4 --stylize 450 --no frame, border, box, blurry, digital illustration
2/
Eye-level straight-on view, a hyper-vibrant 3D layered paper cut-out diorama of GUANGZHOU city (The Flower City).
The scene fuses traditional Lingnan culture with modern vitality:
* **Background Layer:** The iconic **Canton Tower (slim waist)** and Zhujiang New Town skyline standing tall.
* **Mid-Ground:** Bustling **Qilou arcade streets** with stained glass windows, filled with crowds enjoying **Yum Cha (Dim Sum)**, stacked steamer baskets, and tea pots.
* **Foreground:** Branches of blooming **Red Kapok flowers** and a festive **Southern Lion (Waking Lion)** head framing the view.
Color Palette: **High contrast palette of Vivid Kapok Red, Emerald Green (Banyan tree tone), and Neon Gold.**
Style & Physics:
* **Material:** Matte cardstock textures.
* **Depth:** Soft studio lighting creates **deep, distinct drop shadows** between the Qilou layers and the tower background, showing the physical spacing of the paper.
* **Vibe:** Maximalist, lively, full of fireworks and life (Yan Huo Qi).
Details: Octane render, 8k, clean edges, surreal realism. --ar 3:4 --stylize 450 --no frame, border, blurry, flat illustration
超鲜艳3D层压剪纸艺术作品
作者: @两斤
来源: X
Banana Prompt
---
1/
Eye-level straight-on view, looking directly into a hyper-vibrant 3D layered paper cut-out diorama. glowing China Red and Teal and Gold, extremely high saturation, maximalist composition, crowded festival scene with dragon dance and pagodas. Soft even studio lighting creating deep drop shadows between physical paper layers to define depth, matte paper texture, NO sun rays, NO volumetric lighting, octane render, 8k, clean details --ar 3:4 --stylize 450
2/
Eye-level straight-on view, a hyper-vibrant 3D layered paper cut-out scene illustration, full bleed composition. Glowing China Red and Teal and Gold, extremely high saturation, maximalist composition, richly detailed festival scene with dragon dance and pagodas extending to the edges. Soft even studio lighting creates distinct, realistic drop shadows cast by each individual paper layer onto the layer beneath it, revealing tangible depth and physical thickness of the matte paper. Octane render, 8k, clean details --ar 3:4 --stylize 450 --no frame, border, box, sun rays, visible light beams
https://t.co/1nF59wIImV
微缩世界与巨物美食
作者: @两斤
来源: X
微缩世界 & 巨物美食
烤鸭 + 面条 + 火锅 + 饺子
Prompt：
{
"collection_meta": {
"theme": "Miniature Worlds & Macro Chinese Cuisine",
"style_preset": "Microphotography / Tilt-Shift Realism / Cinematic Lighting",
"aspect_ratio": "3:4",
"cultural_note": "Emphasizing intricate textures of Chinese cooking methods"
},
"scenes": [
{
"id": "scene_cn_01_dumpling_assembly_line",
"subject": {
"team": "Miniature factory workers in sterile suits",
"action": "Assembling a gigantic pork and cabbage dumpling (Jiaozi)",
"tools": [
"Industrial hoses pumping meat filling",
"Heavy machinery crimping dough edges"
],
"details": "Workers slipping on greasy bits of cabbage filling, steam rising"
},
"environment": {
"macro_object": "A massive, partially folded raw dumpling",
"background": "Mountain range of stacked bamboo steamers (Dim Sum baskets)",
"lighting": "Warm, humid atmosphere, focusing on the texture of raw dough and juicy filling"
}
},
{
"id": "scene_cn_02_hotpot_expedition",
"subject": {
"team": "Miniature deep-sea explorers/divers",
"action": "Navigating a dangerous, bubbling spicy hot pot broth",
"gear": [
"Raft made of a thick lotus root slice",
"Heat-resistant diving suits"
],
"details": "One diver examining a giant Sichuan peppercorn floating by"
},
"environment": {
"macro_object": "A boiling sea of red chili oil broth with floating debris",
"surface": "Bubbling liquid surface with islands of mushroom and tofu",
"lighting": "Dramatic, intense red glow from beneath the surface, steamy and chaotic"
}
},
{
"id": "scene_cn_03_noodle_infrastructure",
"subject": {
"team": "Miniature structural engineers and builders",
"action": "Untangling and organizing massive hand-pulled noodles (La Mian)",
"tools": [
"Cranes lifting single noodle strands",
"Scaffolding built around bok choy stalks"
],
"details": "Workers abseiling down a long, oily noodle strand"
},
"environment": {
"macro_object": "A chaotic, tangled bowl of thick hand-pulled noodles",
"background": "Giant pair of chopsticks looming overhead like skyscrapers",
"lighting": "Bright, highly detailed, emphasizing the glistening, stretchy texture of noodles"
}
},
{
"id": "scene_cn_04_peking_duck_mining",
"subject": {
"team": "Miniature miners with pickaxes and headlamps",
"action": "Mining the crispy skin off a Peking Roast Duck",
"tactics": [
"Chipping away at golden-red crispy skin like valuable ore"
],
"details": "Using thick hoisin sauce as industrial lubricant for their carts"
},
"environment": {
"macro_object": "The glistening, roasted surface of a whole Peking Duck",
"surface": "Cracked, oily, lacquered skin texture",
"atmosphere": "Rich, golden hour lighting, highlighting grease and crispiness, savory atmosphere"
}
}
]
}
城市/地点名称超写实等距建筑模型
作者: @两斤
来源: X
{
"prompt": "Subject: [INSERT CITY/LOCATION NAME]. \n\nCreate a hyper-realistic isometric architectural maquette of the Subject, strictly isolated on a seamless pure white studio background. \n\n[Surface Level]: Analyze the Subject to recreate its signature architectural style, vegetation, and street atmosphere. \n- **Architecture**: Representative buildings with authentic materials (e.g., brick, glass, wood) and local design details.\n- **Street Scene**: Localized vehicles, street furniture, and scattered life details appropriate to the location.\n\n[Underground Level]: A deep vertical cross-section revealing underground elements strictly based on the Subject\'s real-world infrastructure:\n- IF it is a modern metropolis: Show complex subway tunnels, maintenance pipes, and concrete foundations.\n- IF it is a historical or rural site: Show rich soil stratification, ancient ruins, sewers, or deep root systems.\n\n[Crucial Lighting & Material Details]: \n1. **Lighting Physics**: The strong sunlight from above must cast distinct, hard shadows from the street level down onto the underground cutaway sections. The underground should be slightly darker, lit by its own dim interior lights (if infrastructure exists) or fading into shadow, creating a contrast with the bright surface.\n2. **Soil Texture**: The earth layers must look realistic and gritty—soil mixed with rocks, gravel, and roots. No smooth textures.\n3. **Photography**: Macro lens photography. Shallow depth of field (tilt-shift focus on the main building) is mandatory. The background MUST be empty white void. \n\n[Technical]: 8k resolution, Unreal Engine 5 render, highly detailed physical model aesthetic.",
"negative_prompt": "sky, city skyline, distant buildings, mountains, clouds, wallpaper, illustration, painting, flat texture, clean concrete, bright underground, cartoon, low resolution, watermarks, text",
"aspect_ratio": "2:3"
}
完美微缩纸币
作者: @两斤
来源: X
原生审美 + 垂直剪影 + 巨微反差 + 电影景深 = 完美微缩纸币
---
prompt
{
"prompt": "A dramatic, ultra-macro photograph of a genuine [INSERT CURRENCY NAME ONLY], perfectly replicating a high-end miniature art aesthetic. \n\n[Composition]: \n- The banknote is placed **diagonally**, dominating the frame with tight cropping. \n\n[Intelligent Paper Engineering]: \n- **Auto-Subject**: The **signature landscape, architectural motif, or iconic animal** traditionally featured on the reverse side of this specific banknote is identified and rendered as **intricate, multi-layered paper silhouette cutouts** standing vertically from the flat surface. \n- **Style**: The cutout follows the exact contours of the original bill\'s design. The texture is razor-thin, matte currency paper, strictly matching the **authentic color palette** of the specified bill.\n\n[Contextual Miniature]: \n- A hyper-realistic, microscopic **tiny figure or vehicle** that is **culturally and thematically appropriate** to the scenery (e.g., a boat for water, a hiker for mountains, a car for buildings) sits on the flat printed area, interacting with the paper sculpture.\n- A huge, realistic **human thumb** (showing the **fingernail/nail plate**, NOT the fingerprint) dominates the foreground corner to establish scale.\n\n[Photography - The \'Soul\']: \n- **Ultra-shallow depth of field is mandatory**. \n- **Focus Strategy**: Razor-sharp focus ONLY on the tiny miniature model and the ridges of the standing paper cutout. \n- **Bokeh**: The background, foreground text, and finger melt into a **soft, creamy bokeh** that matches the bill\'s dominant color. \n- **Lighting**: Cinematic side-lighting highlighting the intaglio ink and paper fibers.",
"negative_prompt": "flat image, illustration, cartoon, low resolution, deep focus (everything sharp), centered composition, empty background, glossy paper, plastic texture, wrong landmark, generic scenery",
"aspect_ratio": "16:9"
}
桂林风光版20元人民币
作者: @两斤
来源: X
{
"prompt": "Subject: Genuine Chinese 20 Yuan Banknote (Guilin Landscape edition). \n\n[Macro Material Analysis]: The object must be rendered with the exact physical properties of real currency paper—matte cotton-fiber rag paper, NOT glossy cardboard. \n\n[Paper Engineering]: The karst mountain scenery is delicately cut and lifted. \n- **Critical Thickness**: The cut-out paper layers must appear razor-thin (0.1mm), fragile, and slightly translucent against the light. Edges should show microscopic fibrous tearing, not clean thick cuts.\n- **Printing Texture**: The mountains are NOT solid colors. They must be composed of microscopic engraved lines (intaglio printing) and guilloche patterns. The ink should look slightly raised on the paper surface.\n\n[Scene Context]: A realistic tiny bamboo raft floats on the flat printed water. \n\n[Scale Reference]: A giant, realistic human finger with distinct skin texture presses the edge of the bill. The finger is huge compared to the tiny raft, emphasizing the miniature scale.\n\n[Photography]: Macro lens, high contrast lighting to show the texture of the paper fibers. Shallow depth of field.",
"negative_prompt": "glossy paper, thick cardboard, plastic texture, toy money, monopoly money, solid color blocks, blurred printing, low resolution, cartoon, thick edges",
"aspect_ratio": "16:9"
}
宝玉老师灵感下的蛋雕艺术
作者: @两斤
来源: X
灵感来自宝玉老师 @dotey
Inspired by the delicate art of egg carving, intricately depict a grandmother sitting in a rocking chair knitting by a window through elaborate hollow cutouts on a single, complete duck egg shell.
All carved elements are organically connected by thin, delicate shell bridges and lattice structures to maintain structural integrity, ensuring no isolated or floating segments.
The shell surface has a smooth, matte white texture. Gentle, warm backlighting illuminates the shell from within, creating a translucent glow around the cutouts and cozy silhouettes.
Background features a softly blurred rustic living room with a fireplace, providing a nostalgic and warm atmosphere, captured in ultra-detailed 8K resolution for macro-level realism, showing every tiny chisel mark.
0xFunky
Nano Banana Pro：10秒完成Binance注册教程
作者: @0xFunky
来源: X
真的玩上癮了，Nano banana pro 真的太扯
就一句prompt，只花10秒就完成了這下面的幣安註冊功略圖圖，重點是繁體中文字完全沒有出錯，還幫我把邀請碼完全沒出錯的放上去了...
Prompt: "幫我畫出一份「申請 Binance」的 step-by-step 教學DM，每一個步驟用繁體中文做說明，圖上要有幣安LOGO，同時加入我的 binance 推薦邀請碼: GRO_28502_YHQUS "
這完全是要顛覆整個行銷產業了...
@binance @cz_binance @heyibinance @binancezh
AI_only
Character Feature Illustration
作者: @AI_only
来源: X
「このキャラクターの各所を線で指し示しながら、各部分の特徴を日本語で説明した画像を作成してください」
← Nano Banana Pro
→ GPT Image 1.5
🍌つよい
Nano Banana Pro 三面图
作者: @AI_only
来源: X
NovelAI → Nano Banana Pro (三面図) → GPT Image 1.5
GPT Image 1.5 お試し
prompt1:
「この画像のキャラクターがカフェで給仕をしている様子を画像にしてください」
prompt2:
「手書き文字 (日付 (2025.12.18) + ハートマーク) と装飾を加えてチェキ風にしてください」
Latte
高端时尚杂志封面
作者: @Latte
来源: X
新做了一个高端时尚杂志封面📷，这次结合了「选区编辑」互动的视觉概念，体现具有冲击力的现代设计感，人物可以自定义。
分享提示词：
一张高端时尚杂志封面。{年轻女性时尚模特}，气质自信前卫，身体张力强，动态姿态，直视镜头。镜头为略低机位仰拍。模特双手比起取景框手势，仿佛与一个矩形选中框有互动，选中框覆盖脸部和肩部。视觉规则：只有选中框内部清晰且为自然彩色；选中框外完全灰度强像素化，无任何颜色或清晰区域。柔和漫射棚拍光。现代编辑感穿搭。版式文字：顶部居中粗体大写扁宽型无衬线标题“FOCUS”，上方小字“DECEMBER 2025”；标题左下“VOL + 随机两位数”；左下角文字块（简短时尚自信标题、短段落、条形码）；右侧“FASHION INTERVIEW”；右下角衬线体小号“THE EDIT”及大号“01-09的随机数字”。模特对标题有遮挡叠加，前后景关系，层次分明，干净现代。字体均为白色，蒙太奇风格字体，文字与图像形成强烈对比。
时尚质感短片工作流
作者: @Latte
来源: X
这个工作流可以稳定输出时尚质感短片
受到 @ReflctWillie 和 @underwoodxie96 的启发，突然有了灵感，也想试试看。整个流程比想象中更可控，而且每一步都有明确的作用。
这个视频的生成教程：
1. 生成一张基础人物图
2. 将图像导入 Nano Banana Pro，使用提示词：在一个 3×3 的网格中展示这个角色，从不同的角度呈现，单色的渐变背景，姿势随机。
如果需要特定姿势，直接在提示词中指定即可。
3. 生成后截图喜欢的关键帧，然后使用提示词：放大到 4K。得到高清图片
4. 使用 Kling 2.5 Turbo 将首尾帧变成视频，提示词：摄影机在摇臂上缓慢平稳地下降
5. 使用 @ReflctWillie 开发的速度曲线工具使用预设调整每个片段的播放节奏，最后拼接成完整视频。
对比在传统剪辑软件慢慢调试实在方便太多。
自从有了 AI 之后，每个人都可以高效地开发出解决痛点的小工具，让我们可以把更多精力投入到创意本身，而不是被技术细节困住，这个转变真的很棒。
Y2K拼贴美学城市印象海报
作者: @Latte
来源: X
分享一个我最近打磨的视觉风格：
用 Y2K 拼贴美学重新诠释城市印象，把街头文化、时尚人像和杂志排版糅合成一张纸质拼贴海报。
可以无限定制任何城市千禧年代的潮流杂志封面。背景、符号、美食、地标都会根据城市特色而变化，每次生成都是惊喜✨
提示词太长，所以放在评论区了，只需要替换城市名称就行：
提示词模版：
{
"variables": {
"CITY_NAME": "Chengdu"
},
"image_specs": {
"aspect_ratio": "4:5", "resolution": "2048x2560", "quality": "ultra", "style_strength": 0.8, "detail_level": "high", "sharpen": "medium"
},
"prompt": {
"master_visual_brief": "A high-energy Y2K-inspired editorial collage poster with a strong paper-cut and magazine print aesthetic. The entire image has a tactile paper texture with visible cut edges and layered depth. The theme centers around the city {{CITY_NAME}}. All visual elements, including background imagery, stickers, symbols, typography, and graphic decorations, are culturally and visually inspired by {{CITY_NAME}}. The composition follows a fashion magazine cover logic with dense but controlled information, playful energy, and strong visual hierarchy. The character is designed as a dominant half-body portrait occupying most of the poster.", "photography_and_character": "Character: a randomly generated young woman aged approximately 18–25. She is fashionable, attractive, and trendy, with no fixed hairstyle, hair color, facial features, or makeup. Her appearance varies naturally but always remains stylish and visually appealing. Fashion style is Y2K-inspired street fashion with playful silhouettes, layered styling, and trendy colors. Framing is a strict half-body portrait: the image is cropped at the waist or slightly above, and the lower body is not visible at all. Only the upper torso, shoulders, neck, and head are shown. Pose and gesture: the character performs a randomly selected Y2K-style dynamic pose with strong tension and attitude, including expressive arm extensions, angular elbow bends, asymmetrical shoulder twists, or forward-reaching gestures. Body language remains bold, confident, and energetic. Facial expression: the expression is slightly playful and friendly, with a hint of cuteness layered on top of confidence. Subtle smiles, softly open lips, bright eyes, or a relaxed playful look are allowed, while the overall attitude remains fashion-forward and dynamic rather than cute-only.", "camera_and_lighting": "Editorial portrait framing with a wide-angle look at close distance, optimized for half-body composition. The camera captures only the upper torso and head, with the lower body fully cropped out of frame. Perspective supports dynamic Y2K poses without distorting facial proportions. Lighting is soft, even, and magazine-style, avoiding harsh shadows. The subject is intended to be cut out and integrated into a paper collage rather than rendered as pure realism.", "graphic_design_layout": "Center-focused editorial collage layout. At the top, the city name \'{{CITY_NAME}}\' appears as the main headline in bold uppercase geometric sans-serif letters. Each letter is placed on an individual colored paper rectangle and arranged in a slight arc. The large half-body character overlaps and partially covers the headline and nearby graphic elements, creating a break-the-frame effect. Surrounding the character are floating paper stickers, speech bubbles, and cut-out graphics inspired by {{CITY_NAME}} culture, including local food, landmarks, symbols, street signs, and iconic objects. The bottom section features a full-width magazine-style collage footer composed of layered paper strips, bold button-style typography, small editorial text blocks, and thumbnail-style graphics.", "background_system": "The background is a black-and-white or desaturated urban street scene from {{CITY_NAME}}, such as crowds, architecture, or city textures. Background contrast is reduced and softened with grain so it supports the composition without competing with the main subject. The background is partially obscured by the large half-body character and collage elements and maintains a printed-paper appearance rather than photographic realism.", "materials_and_textures": "A consistent paper-based aesthetic across the entire image. Visible paper grain, halftone dots, print noise, and slight ink bleed. All elements appear as physical paper cut-outs layered together. Edges are imperfect and tactile. Stickers, typography, and characters cast subtle shadows to suggest layered depth. No glossy, metallic, or digital materials are present.", "composition_and_balance": "Clear and stable visual hierarchy: top city headline \'{{CITY_NAME}}\', central half-body character (upper torso only), dynamic Y2K-style pose with slightly playful facial expression, surrounding city-themed stickers, and bottom magazine-style collage footer. Strong overlaps between character, typography, and stickers create depth while preserving the established layout."
},
"constraints": {
"must_include": \[ "City name headline using {{CITY_NAME}}", "Strict half-body portrait (waist-up only)", "Lower body completely cropped out", "Y2K-style dynamic pose with strong tension", "Playful but controlled facial expression", "Paper collage and cut-out magazine aesthetic", "City-specific background and stickers related to {{CITY_NAME}}", "Full-width magazine-style collage footer", "Visible paper texture and layered depth" \], "must_avoid": \[ "Full-body view", "Visible legs, knees, thighs, or feet", "Neutral or stiff poses", "Overly cute or childish expressions", "Sexualized expressions or gestures", "Grotesque facial distortion", "Minimalist or empty layouts", "Glossy digital or 3D materials" \]
},
"negative_prompt": "full body, legs visible, knees, thighs, feet, stiff pose, neutral posture, childish expression, exaggerated cute face, sexualized expression, blurry face, deformed hands, extra fingers, bad anatomy, grotesque distortion, minimalist layout, flat image, hyper-realistic photography, glossy surfaces, plastic skin, dull colors, watermark, logo, unreadable text",
"typography_rules": {
"headline": { "text": "{{CITY_NAME}}", "font_style": "bold geometric sans-serif, uppercase", "treatment": "each letter on a separate colored paper rectangle, slightly arched, subtle shadow", "material": "printed paper cut-out" }, "supporting_text": { "style": "editorial magazine blocks, playful sticker captions, speech bubbles", "material": "paper-based printed texture" }
},
"rendering_notes": {
"depth_layers": "background city paper layer -> mid-ground half-body character -> foreground dynamic pose, expressive face, stickers, and typography", "print_feel": "strong magazine print texture with halftone dots, paper grain, and slight ink bleed", "edge_treatment": "clearly visible imperfect cut paper edges"
}
}
太空汉堡飞船降落在外星荒漠
作者: @Latte
来源: X
太空汉堡飞船降落在外星荒漠✨
分享一下我的提示词：
{
"品牌主题": "麦当劳",
"概念": "停泊在外星地面的检修中汉堡宇宙飞船",
"主体": {
"主要元素": "一个巨大的麦当劳汉堡被设计成一艘宇宙飞船，汉堡整体即为飞船主体，面包、牛肉饼与芝士构成圆润流线型船身，飞船稳稳停泊在外星星球地表，处于简洁、有序的检修状态",
"飞船装置": "汉堡下方与侧面仅保留少量简洁的金属推进口与舷窗，造型光滑克制，无外露复杂管线或零件，起落支架造型简约，与地面自然接触，舱门微微开启，作为检修状态的视觉提示",
"细节": "超微距摄影风格，汉堡看起来极其美味，面包松软微焦，牛肉饼多汁厚实，芝士被热度融化，停留在即将流下来的临界状态，拉丝明显，表面油润发亮，芝麻清晰可见，整体质感干净高级"
},
"人物": {
"类型": "微缩比例的宇航员",
"动作": "多名微缩宇航员直接在汉堡表面进行检修与检查，有的沿着汉堡侧面架设简易梯子攀爬，有的站在面包顶部观察结构，有的靠近芝士边缘进行检查或记录，部分宇航员使用安全绳固定在汉堡表面，人与汉堡形成明确互动，强化尺度对比与趣味性"
},
"环境": {
"场景": "陌生外星星球地表，作为临时停泊与检修地点",
"地貌": "外星地表以简洁的岩石、沙砾或柔和起伏的地形为主，形态抽象克制，避免杂乱细节",
"氛围": "空气清澈，少量漂浮尘埃增强尺度感，整体氛围安静、从容，具有高级电影感",
"背景": "远处简洁的外星建筑轮廓与地平线，建筑后方竖立着标志性的金色拱门标志，像霓虹灯一样柔和闪耀；天空中可见一到数颗遥远的行星，体量克制、边缘柔和，作为背景存在，增强宇宙纵深与科幻感"
},
"美学风格": {
"光影": "温暖定向光集中表现汉堡的食物质感，金色拱门霓虹光作为背景氛围光，与冷色外星环境光形成柔和对比，光影层次清晰，画面不显拥挤",
"风格": "电影级广告视觉，画面简洁、现代、克制，超现实但高度真实，强烈食欲感与轻度科幻探索感并存，具有记忆点"
},
"元信息": {
"画幅比例": "9:16",
"参考指令": "画面构图与整体风格需与上传的参考图保持一致"
}
}
乔布斯职业生涯的《指环王》主题地图
作者: @Latte
来源: X
我用 Nano Banana Pro 重新诠释了乔布斯的职业生涯，绘制成了《指环王》主题地图。
我用的提示词：
请为[乔布斯]创建一个有趣的职业生涯地图，使用[指环王]的主题。需包含相关的隐喻或比喻，将这位名人的职业经历与所选主题连接起来，使职业地图感觉连贯且有意义。中英文案，确保所有文字清晰易读。
乔布斯的一生就是一场史诗：车库创业、被驱逐、流亡、归来、改变世界。这种起伏和张力，用中土世界的框架来呈现，竟然意外地契合。
每个人心中都有一座中土世界，而真正的冒险从来不是征服已知，而是怀着敬畏之心，踏入未知的迷雾。
Stay hungry, stay foolish.
Daniel
创意 toilet 图像
作者: @Daniel
来源: X
@nurpraditya For those who want to create something like this, simply write what you want and paste this JSON profile.
I generated a toilet as an example.
{
"art_style_profile": {
"style_name": "Minimalist 3D Illustration",
"visual_elements": {
"shape_language": "Rounded edges, smooth and soft forms with simplified geometry",
"colors": {
"primary_palette": [
"Soft beige, light gray, warm orange"
],
"accent_colors": [
"Warm orange for focal elements"
],
"shading": "Soft gradients with smooth transitions, avoiding harsh shadows or highlights"
},
"lighting": {
"type": "Soft, diffused lighting",
"source_direction": "Above and slightly to the right",
"shadow_style": "Subtle and diffused, no sharp or high-contrast shadows"
},
"materials": {
"surface_texture": "Matte, smooth surfaces with subtle shading",
"reflectivity": "Low to none, avoiding glossiness"
},
"composition": {
"object_presentation": "Single, central object displayed in isolation with ample negative space",
"perspective": "Slightly angled, giving a three-dimensional feel without extreme depth",
"background": "Solid, muted color that complements the object without distraction"
},
"typography": {
"font_style": "Minimalistic, sans-serif",
"text_placement": "Bottom-left corner with small, subtle text",
"color": "Gray, low-contrast against the background"
},
"rendering_style": {
"technique": "3D render with simplified, low-poly aesthetics",
"detail_level": "Medium detail, focusing on form and color over texture or intricacy"
}
},
"purpose": "To create clean, aesthetically pleasing visuals that emphasize simplicity, approachability, and modernity."
}
}
0x卡卡撸特 | Golden.S
《诡秘之主》海报设计
来源: X
太顶了，最近在二刷《诡秘之主》，用nano banana pro做了一张海报，这美学太震撼了！
提示词：
请为影视剧/小说《需要添加的名称》设计一张高品质的3D海报，需要先检索影视剧/小说信息和著名的片段场景。 首先，请利用你的知识库检索这个影视剧/小说的内容，找出一个最具代表性的名场面或核心地点。在画面中央，将这个场景构建为一个精致的轴侧视角3D微缩模型。
风格要采用梦工厂动画那种细腻、柔和的渲染风格。你需要还原当时的建筑细节、人物动态以及环境氛围，无论是暴风雨还是宁静的午后，都要自然地融合在模型的光影里。 关于背景，不要使用简单的纯白底。请在模型周围营造一种带有淡淡水墨晕染和流动光雾的虚空环境，色调雅致，让画面看起来有呼吸感和纵深感，衬托出中央模型的珍贵。
最后是底部的排版，请生成中文文字。居中写上小说名称，字体要有与原著风格匹配的设计感。在书名下方，自动检索并排版一句原著中关于该场景的经典描写或台词，字体使用优雅的衬线体。整体布局要像一个高级的博物馆藏品铭牌那样精致平衡。
巨型人物/动物照片
来源: X
这里有个好玩的点子，你可以试试。🤣
👉用 Nano Banana Pro 制作巨型人物/动物的图片。
👉只需调整括号里的元素，尽情发挥创意吧！
👉图片生成好之后，用Veo 3.1转视频，白嫖的Gemini不用白不用。
一张高度细节化、照片级写实的[拍摄角度，例如广角镜头、俯拍镜头、超低角度仰拍镜头]照片，展现了一个巨大的[描述人物及衣着]，位于[具体地点/城市/景观]。这个巨人正在[动作 - 与环境互动，例如坐在建筑物上、跨过桥梁]。为了凸显其巨大的规模，在其[脚/手]附近可以看到微小的[微小元素：人、汽车或船只]。光照条件为[一天中的时间，例如黄金时段、正午阳光]。
Marcel
纳米香蕉Pro - 制作乐高
作者: @Marcel
来源: X
Nano Banana Pro - "Make this Lego"
Gm ☕️ https://t.co/KBfRgACIRR
Camille_1982_fr
基于上传照片的干净现代肖像网格
作者: @Camille_1982_fr
来源: X
Create your image with different expressions in chatGPT via GPT-image 1.5 using the prompt
Starting with the uploaded photo as a base reference, create a clean and modern portrait grid featuring the same person with several different facial expressions and moods. Arrange the images in a neat 3×3 layout with rounded corners and soft pastel backgrounds. Each portrait should retain the same facial features, hairstyle, and realism, changing only the expression and subtle style. Include expressions such as: natural smile, joyful laugh, calm with eyes closed, neutral and serious gaze, pensive pose with hand on chin, confident and composed expression, sad or disappointed mood, mischievous pout, and surprised reaction with wide eyes. Use soft studio lighting, highly detailed skin texture, shallow depth of field, minimal variations in wardrobe, consistent color palette, ultra-realistic editorial portrait style, high resolution, and clean aesthetics.
Kaushik Shivakumar
Nano Banana Pro 全美地图，标注各州缩写（水彩风格与创意插画）
来源: X
Asked Nano Banana Pro 🍌 for a full map of the US with all states labeled with their abbreviations (in watercolor style & creative illustrations). This is actually quite a challenging task for image generation models, and this is the first model I've seen that gets this right. https://t.co/tABg05SJ5n
バナーナ
画像生成の未来とチャンス
作者: @バナーナ
来源: X
画像生成はどうせすぐ飽和する。
参入障壁が低いから、誰でも出来るし、競争もすぐ激化する。
っていう未来はたぶん正しい。
でも逆に言えば、誰でも出来る今ってめちゃくちゃチャンスでもある。
「画像生成で稼ごう」とか「マネタイズしよう」とか、いきなり考えなくて良い。
ガチの初心者なら、まずは何も考えず遊んでみるのが一番。
日本語でもそれっぽいの出るし、英語でも中国語でもプロンプト次第でどんどん面白くなる。
いちいち正解探さなくて良い。ただ触ってみる。それだけで十分。
無料で使えて、生成制限もないGoogleのImageFXとかから始めるのがオススメ。
ツールもプロンプトも、まずは触ることから全部始まる。
ちな、このポストの画像で使ったプロンプトはリプ欄に貼っとく👇
■プロンプト（必要に応じて変えてください）
{
"image_size": {
"width": 1152,
"height": 2048,
"aspect_ratio": "9:16",
"orientation": "portrait"
},
"prompt": {
"subject": "若い日本人成人女性のクローズアップポートレート",
"composition": {
"framing": "顔のアップ、額から顎までを中心にした構図",
"angle": "やや斜めからの視線、カメラ目線に近い",
"crop": "肩は一部のみ写り、背景は最小限"
},
"facial_features": {
"eyes": "大きめで丸みのあるダークブラウンの瞳、自然なまつ毛",
"skin": "滑らかで明るい肌質、薄いそばかすや小さな肌の点がわずかに見える",
"lips": "自然なボリューム感、赤みのあるツヤ感リップ",
"nose": "小ぶりで自然な形"
},
"hair": {
"color": "ダークブラウン〜ブラック",
"style": "前髪あり、自然に下ろしたストレート寄りの髪",
"texture": "やや柔らかく、細めの毛束感"
},
"makeup": {
"style": "ナチュラルメイク",
"details": "薄いアイメイク、肌になじむチーク、控えめなハイライト"
},
"lighting": {
"type": "ソフトライト",
"direction": "正面〜やや横から",
"effect": "影が少なく、肌の質感をきれいに見せる"
},
"background": {
"color": "白〜淡いグレー",
"texture": "無地、ボケ感あり",
"distraction": "なし"
},
"style": {
"realism": "高い写実性、写真風",
"mood": "落ち着きがあり清潔感のある印象",
"quality": "高解像度、肌や目のディテールが明瞭"
}
},
"negative_prompt": [
"イラスト",
"過度なメイク",
"不自然な肌加工",
"強い影",
"派手な背景",
"イラスト風",
"低解像度",
"ぼやけ"
]
}
1LittleCoder💻
印度男子瑜伽步骤与美学的信息图表
作者: @1LittleCoder💻
来源: X
GPT-Image-1.5 is one of the best for Infographics Design!
Give the topic or the text content - ask it to make infographics
Example Prompt-
"create an infographic with an indian man with different yoga steps and details with beautiful typography that matches with Yoga style"
The same thing would have taken Graphic designers more a handful few hours, but today, it's just less than a minute generation!
Wan 2.6 文本转视频 ASMR 新作
作者: @1LittleCoder💻
来源: X
New ASMR👑 dropped - Wan 2.6 Text-to-Video!
Prompt 👇🏽
清澈碧水中的参考图像
作者: @1LittleCoder💻
来源: X
GPT-Image-1.5 is good for Product Photoshoots. There's something about its consistency I like. Imho, it respects the original image much more than Mr. NanoBanana Pro!
Prompt:
[Reference Image] fully submerged in crystal-clear, turquoise water, captured in ultra-high-resolution underwater photography. Sunlight penetrates the surface above, creating intricate caustic light patterns that ripple and dance across the subject and surrounding water. The scene conveys pristine clarity with zero particulate matter, emphasizing a sense of suspended weightlessness and serene motion. Fine details are frozen using high-speed capture, with subtle bubbles and flowing fabric or hair enhancing the feeling of aquatic elegance. The overall aesthetic is clean, refreshing, and ethereal, with soft natural color grading, high dynamic range, and cinematic realism.
1RPC.ai
Nano Banana Pro的圣诞雪花球角色
作者: @1RPC.ai
来源: X
Christmas snow globes with characters in kawaii form by Nano Banana Pro
Prompt ↓ https://t.co/jUeG89Sd5A
✨ Pulikesi✨
人工智能专家形象
作者: @✨ Pulikesi✨
来源: X
1. Open ChatGPT / Gemini AI / Grok AI
2. Upload your picture
3. Enter this prompt—
{
"pose": {
"masculine": {
"p": "Leaning back against a textured wall; grounded, self-contained posture; squared shoulders, chest subtly forward; head level or slightly lowered; arms resting naturally at sides; calm, guarded gaze; stoic, controlled mood; minimal movement; strong presence; dramatic directional lighting emphasizing structure without exaggeration.",
"n": "exaggerated posing, softness, overt sensuality, dramatic gestures, emotional excess, submissive posture, theatrical expression"
},
"feminine": {
"p": "Leaning subtly forward against a textured wall; relaxed, fluid posture; softened shoulders and gentle curves; head tilted slightly back or to the side; arms close but relaxed; sleeves loosely rolled, fabric falling naturally; open, warm gaze; intimate, sensual mood; soft yet dramatic lighting highlighting natural contours.",
"n": "rigid posture, aggressive stance, exaggerated dominance, stiffness, harsh expression, artificial posing, over-stylization"
}
},
"wardrobe": "Lightweight off-white linen shirt or dress-style top; relaxed, fluid silhouette; visible fabric weave; softly wrinkled texture; open, banded collar; loosely rolled sleeves; natural drape emphasizing movement and tactile realism.",
"environment": "Richly textured vivid red wall; bold saturated color palette; high color depth; strong contrast with realistic skin tones.",
"face_lock": {
"enable": true,
"priority": "max",
"strength": 0.85,
"rules": "Preserve uploaded facial identity exactly; no alteration to bone structure, proportions, age, ethnicity, or unique asymmetries; allow only lighting, color, and micro-expression variation."
},
"camera": {
"lens": "85mm",
"aperture": "f/1.8",
"dof": "shallow"
},
"lighting": "Dramatic directional key light with soft rim light; cinematic shadow depth; sculpted highlights.",
"grading": {
"palette": "vivid",
"contrast": "high",
"tone": "cinematic"
},
"quality": "Photorealistic RAW; ultra-detailed skin and fabric; natural imperfections preserved.",
"negative_global": "face alteration, identity drift, symmetry correction, beauty standardization, plastic skin, excessive retouching, washed-out colors, flat lighting, cartoon, anime, low detail, blur, distortion",
"aspect_ratio": "2:3",
"style": "raw"
}
现代电影式肖像摄影
作者: @✨ Pulikesi✨
来源: X
1. Open ChatGPT
2. Upload your picture
3.Enter the prompt —
"A cinematic studio portrait of the person in the uploaded photo, framed from the chest up, facing the camera with a calm, confident presence.
Dramatic dual-tone lighting with cool teal and warm golden tones creating a modern, cinematic mood.
Natural skin texture, sharp focus on the eyes, soft shadows, and smooth color transitions.
Hair styled naturally, wearing a minimal dark outfit with a contemporary aesthetic.
Clean studio background with subtle gradients.
Professional photography, shallow depth of field, cinematic color grading, ultra-realistic, high detail, 4K quality.
Aspect ratio: 3:4 "
サトシ
SynClub活动首次使用Nano Banana Pro
作者: @サトシ
来源: X
SynClubのイベントで初めてNano Banana Proを使ったけど、流石。
キャラ二人を選んで「街角でサンタのコスプレをしながらクリスマスケーキの売り子のアルバイトを二人でしている」
これだけでちゃんと街はクリスマス仕様になっているし、何より今まで弱かった日本語の文字が綺麗に出ている。 https://t.co/vrkdCbiui4
369 Labs
GPT Image 1.5 与 Nano Banana Pro
作者: @369 Labs
来源: X
Left: GPT Image 1.5
Right: Nano Banana Pro
Prompt below👇
角煮星丸
Nano banana Pro 小技巧分享
作者: @角煮星丸
来源: X
⭐️Nano banana Proの小技共有
奥に小さくキャラを置きたいんだけどなー
って時に使えるかもしれないテクニック。
①背景の元を用意する
②bananaプロでキャラを消す
③ペイントツールでキャラを配置したいところを適当に塗る
④bananaプロで塗り領域とキャラを置き換える
プロンプトはリプ欄に！
无茶振り海报选手权
作者: @角煮星丸
来源: X
✨ 無茶振りポスター選手権✨
prompt
--------
このキャラクターでめっちゃかっこいいポスターを作って。
ポーズとか構図も自由に考えていいからとにかくめっちゃカッコよくして。
Silent Belly, Holy Nightっていうテキストをカッコよく入れて。
--------
上記の舐めてる丸投げ日本語プロンプトで、各社モデルにキャラの正面立ち絵のキャラ参照画像を渡して、カッコいいポスター画像を3枚ずつ生成してもらいます。見せられないレベルの破綻した画像は作り直して減点です。
一番バリエーション豊かでカッコよかったモデルが優勝です。
採点基準は私の気分です。
画像から参加した4モデルがわかったら凄い。
⚠️ガチの性能比較じゃなくてお遊び企画ですから参考にしないで下さい。
ちゃんとした生成結果が欲しいならプロンプトは真面目に書きましょう。
Kling O1生成海报
作者: @角煮星丸
来源: X
⭐️Kling O1でポスター画像生成⭐️
Kling O1の画像生成がどうも凄そうだというので
前にNano banana Proでやって大好評だった
例のプロンプトをそのまま流し込んでみた！
いいねー！
ちょっとだけテキストが不安定かも…？
キャラはだいぶ変わるな…
プロンプトの引用元はリプ欄に👇
ダディ: 生成AIスタート教室
高市政権AI基本計画
作者: @ダディ: 生成AIスタート教室
来源: X
高市政権が発表した「AI基本計画」
「概要」でも文字がぎっしりで、受け付けない人もいるかもしれません…
ChatGPTで要約
↓
Gemini（Nano Banana Pro）で画像まとめ
これだけでかなり見やすくなります👀
Geminiへのプロンプト
「この内容を詳細に日本語で図解して。グラレコ風にして。本当に超詳細に小さい字でもいいから。キャラクターは添付の画像を使って。」
ミスターVR / Mr.VR🕶
Google AIメガネ発表
作者: @ミスターVR / Mr.VR🕶
来源: X
【速報】GoogleがAIメガネ発表、Gemini搭載で2026年登場
・Google×Samsungが共同開発
・スピーカー、マイク、カメラ内蔵
・Geminiで見ているものに質問可能
・リアルタイム翻訳対応
・Google Maps、YouTube Music等が視界に表示
・撮影した写真をNano Bananaでその場で編集可能
・ディスプレイなし版とディスプレイ付き版の2種類
・Android/iOS両対応
・単眼版はわずか49g
・「まずメガネとして」設計、一日中着用可能
・デザインはWarby Parker、Gentle Monsterが担当
Gemini統合のAIメガネは最強すぎる
動画字幕つけました
AI.comb(コム)
创建这种构图时的prompt
作者: @AI.comb(コム)
来源: X
こういう構図を作る時、とりあえず入れておきたい
cute and playful girl, squatting on the ground, full body close up
おはよう木曜日 ALTに✨
作者: @AI.comb(コム)
来源: X
おはよう木曜日
promptはALTに✨
ultrarealistic, colorful Ink floating in the Multi-dimensional space, back ground is building in megacity, a realistic asian model, thick lips, conceptual photography design, dimensional multilayering, Canon EOS 5D Mark IV, Canon EF 16-35mm f/2.8L, moody atmosphere, 16K UHDP
意味はない
作者: @AI.comb(コム)
来源: X
意味はない
promptには入れたけど🤣
ALTで確認してね 笑
にじv6 Default
スタライズ 中
prompt↓
a ultra-realistic asian beauty, "くじら", Nostalgic photos taken with a Polaroid camera, cheers, party, Canon EOS 5D Mark IV, Canon EF 16-35mm f/2.8L, f/5.6, ISO 800, 1/120 sec, moody atmosphere, 16K UHDP, Lomography, taken with a LOMO, using LC-A, Film used cinestill 800T
おはよう金曜日画像
作者: @AI.comb(コム)
来源: X
おはよう金曜日
画像promptも使用してます(ALT参照)
以下のpromptに自分の过去作を组み合わせています．
colorful Ink floating in the Multi-dimensional space , hyperrealistic painting created in the style of JohnDry Brendon using RenderMan, a realistic asian model, thick lips, conceptual photography design, dimensional multilayering, Canon EOS 5D Mark IV, Canon EF 16-35mm f/2.8L, moody atmosphere, 16K UHDP
よん
Gemini 3与nano banana pro的卓越之处
作者: @よん
来源: X
Gemini 3すごい！nano banana proすごい！ってなってる人、一旦おちつこ。
何がどうすごいのかちゃんと理解している？公式の活用事例すら読んでない人、実は多いんだよね。
とりあえず把握したい人は、nano banana proでまとめ画像作ったから見て👀
ちゃんと知りたい人はリプ読んで↓
こーたろー/営業×AIの人生再設計士
マーケティングの5ステップ円環状図解
来源: X
Nano Banana Proって
使いこなせれば最強なんだけど
指示のクオリティーに左右されます！
この画像も以下のプロンプトで
作成できますのでご参考に↓
〜プロンプト〜
「マーケティングの流れ」を、5ステップの円環状（サークル図）で 時計回りに表現した図解を作成してください。
【全体スタイル】
・白背景のシンプルでビジネス向けの図解
・円形のフローチャートを時計回りに5分割
・ブルー系のアクセントカラーで統一
・日本語テキストを正確・見やすく描画
・スライド資料にそのまま貼れる整ったデザイン
【円環の5ステップ（時計回り）】
円の外周に以下5項目を配置し、太めの矢印でつないでください。
1. 市場理解 - 市場・競合・顧客を調査
2. ターゲット設定 - 狙う顧客層を決める
3. 価値提案（バリュープロポジション） - 選ばれる理由をつくる
4. 施策設計（4P＋集客導線） - 商品・価格・売り方・伝え方を組み立てる
5. 分析と改善（PDCA）
- データを見て改善し次のアクションへ
【レイアウト条件】
・円の外側に角丸ボックスで「タイトル＋短い説明」を配置
・各ステップには線画アイコンを1つ付ける
・中心に「マーケティングの5ステップ」と記載
・矢印は時計回りの流れが明確にわかるように配置
・余白多めで洗練された構図 【禁止事項】 ・人物写真は使用しない
・実在企業名やロゴを入れない
【完成イメージ】
マーケティングの基本プロセスが5つの要素で ひと目で理解できる円環状のビジネス図解。 スライドや提案資料にそのまま使える品質。
#NanoBananaPro #プロンプト
Nehuen
梦幻IMAX 70mm电影肖像
作者: @Nehuen
来源: X
Amazing results based on a prompt by @rjchicago with some changes for Nano Banana Pro!
PROMPT: A dreamy IMAX 70mm film portrait, mid-shot, of a young woman with a gentle, distant expression.
Her hair is in a loose, chic updo with soft tendrils framing her face, wearing a fitted ribbed black wool turtleneck.
The scene is dramatically lit by opposing magenta and teal colored gels, creating volumetric haze and soft bloom halation around highlights on her skin and hair.
The background is a wash of pastel city light bokeh from a Parisian window at night.
Film grain present, extremely shallow depth of field (f/1.2), natural skin textures, ethereal and elegant mood, cinematic color grade.
852話(hakoniwa)
nano banana pro同人誌封面绘制
作者: @852話(hakoniwa)
来源: X
nano banana proで2キャラの絵入れて「この二人の同人誌の表紙」って入力すると下の絵が出る
流石に女性向け同人誌あるある度が高くて笑った https://t.co/8aRdXuxWDo
就是玩AI
街头霸王游戏人物志艺术升华版
作者: @就是玩AI
来源: X
刷到@servasyy_ai黄总的这个街头霸王游戏人物志的提示词，我突然想到了一个万能用法，不单单是游戏了，很多方面都可以做艺术升华。改了一个通用版本（带有立体感）提示词:指南手册通用版提示词：看起来有那种商业渲染味道了，放进PPT里面观感绝佳！！喜欢的可以尝试各种元素，都可以通用！！！
[BRAND NAME]: [品牌名称，例如 Nintendo / Riot Games / FromSoftware / Klei Entertainment 等]
A high-end, glossy concept art magazine editorial photograph of a unique, unexpected functional object conceptualized and designed by [品牌名称]: a premium collector's edition [手册/图鉴/攻略书/生存手札/帝国典籍/远征编年史 等自定义名称] for [游戏完整名称，例如 The Legend of Zelda: Tears of the Kingdom / League of Legends / Elden Ring] on [平台，例如 Nintendo Switch 2 / PC / PlayStation 5], crafted as an ultra-luxurious, sculptural artifact that captures the [游戏核心氛围关键词，例如 epic mystical grandeur / vibrant strategic depth / dark gothic whimsy / high-octane adrenaline] with refined minimalist elegance – a thick hardcover tome opened to reveal intricately illustrated pages of [列出主要内页内容，例如 world maps, boss strategies, ability trees, character profiles, crafting recipes, lore timelines 等].
**1. The Concept & Object (AI Invention):**
The [手册名称] reimagines the [游戏] guide as a [形容词 + 物件类型，例如 sacred relic / forbidden grimoire / imperial war manual / high-stakes tactical codex]: substantial yet [氛围形容词，例如 enigmatic / commanding / adventurous / ominous] (approx. [尺寸，例如 240x300mm] when open), with a [闭合方式，例如 magnetic clasp / heavy magnetic closure flap] featuring an embossed glowing [标志性元素，例如 Erdtree / Hextech crystal / Master Sword / Horizon Festival logo] on the cover, [丝带颜色] silk ribbon bookmark, reinforced [材质] corners, and die-cut windows revealing layered [纸张风格，例如 vellum / parchment / silk-scroll] pages. Interior spreads showcase hyper-detailed [主要内容，例如 raid layouts / Summoner's Rift topography / Constant biomes / Mexico map biomes] with [具体功能，例如 wipe mech timelines / optimal ward spots / jungle clear paths / boss phase breakdowns], [其他内容] charts, and narrative timelines [连接游戏元素，例如 of the Golden Order / of Runeterra regions / from the Shattering]. It feels like a [比喻，例如 relic unearthed from the Erdtree’s roots / forbidden archive from the Institute of War] while serving as the ultimate functional companion for [玩家身份，例如 Tarnished / summoners / survivors / raiders].
**2. Materials & Details (Hyper-Premium):**
Constructed from ultra-premium, highly tactile materials inspired by [游戏美学关键词，例如 Tears of the Kingdom’s epic fantasy / League’s futuristic-fantasy / Don\'t Starve’s gothic whimsy]: hand-bound cover in [主色调，例如 deep midnight-blue / imperial-red / charcoal-black] Italian calfskin or premium Alcantara with subtle [质感，例如 pearlescent shift / cracked patina / silk brocade weave] and [箔色] foil-embossed title "[手册标题，例如 TEARS OF THE KINGDOM COMPANION ARCHIVE]", interior pages on thick [克重，例如 160gsm] textured [纸张类型，例如 cotton paper / vellum / rice paper] mimicking [风格，例如 ancient stone tablets / silk scrolls / aged explorer notes] with faint [水印元素，例如 glowing rune / ink brush / shadow] watermarks, spot UV gloss highlighting [高光元素，例如 Zonai devices / ability icons / nightmare fuel / car liveries], precision metallic foil accents ([颜色] for [元素，例如 rose gold for constructs / chrome silver for hextech]), hand-stitched binding with visible [对比色] contrast thread, and tactile embossing for [纹理，例如 sky island topography / web patterns / tentacle spikes / road textures]. Hyper-realistic details: visible leather grain and [氛围] patina, microscopic paper fiber [质感，例如 antiquity / vibrancy], delicate page curls with [阴影风格] shadows, holographic foil [标志性符号，例如 Elden Ring sigils / League emblems], chamfered [边缘材质] edges, and complex layered contrasts between matte [纸张], glossy illustrations, and shimmering [箔色] embellishments.
**3. Photography & Lighting (Cinematic Studio):**
Shot on a medium format Phase One camera with a 100mm macro lens. Extremely shallow depth of field, razor-sharp focus on hero details (open spread with [主要焦点，例如 sky island map / Summoner's Rift map / Constant biome map], [箔色] foil [元素], embossed [标志], stitching, glowing [效果]) and creamy, smooth bokeh background. Sophisticated studio softbox lighting: gentle enveloping fill light with precise rim lighting to accentuate leather contours, paper tactility, metallic reflections, and subtle [氛围] glows on illustrated elements.
**4. Environment:**
Seamless, impeccably clean studio cyclorama background in a pure ultra-light pastel tone ([氛围色调，例如 epic pale sky azure / intense hextech blue / moody lunar gray / vibrant sunset orange]), completely free of shadows.
**5. Layout & UI Elements (Strict Placement):**
- **Bottom Right Corner:** Small, understated, monochrome gray [品牌logo，例如 Nintendo / Riot Games / Klei Entertainment] logo paired with subtle [标志性图标，例如 glowing Zonai dragon / League pentagon / shadow hand].
- **Bottom Left Corner:** Small, minimalist monochrome gray text describing the product. Font: Manrope Regular with very tight tracking (kerning) and balanced line spacing. Format:
"CONCEPT STUDY: [手册名称].
MATERIAL: [主要材质组合，例如 ITALIAN CALFSKIN, GOLD FOIL, TEXTURED VELLUM PARCHMENT].
SS25."
通用类别新年提示词
作者: @就是玩AI
来源: X
应@Banderhu1980 推友建议，出一套通用类别新年提示词。两个超级大类别Nano Banana Pro通用提示词如下：
超级大类别1: 写实高级肖像类（Photorealistic & Artistic Portraits）
这些提示词支持单人/多人，多种照片质感（超写实、电影、复古、梦幻），强调真实皮肤纹理、光影反射、节日细节（如烟花、香槟、倒计时）。可上传真人照片作为参考，实现高度个性化还原。｛Prompt：
based on the uploaded reference photo (if any), advanced [photorealistic/cinematic/retro/dreamlike] portrait of [a young woman / a handsome man / a group of diverse friends / a romantic couple / a family] welcoming New Year 2026, [solo / multiple characters in elegant outfits] [toasting champagne / hugging joyfully / watching fireworks], festive midnight atmosphere with exploding multicolored fireworks reflecting on skin, falling confetti, city skyline or countdown clock in background, ultra detailed realistic skin pores, hair strands, fabric textures and joyful expressions, [warm golden glow / high-contrast dramatic lighting / soft bokeh], 8k masterpiece｝
超级大类别2: 卡通插图类（Cartoon & Illustration Styles）
这些提示词覆盖所有主流画风（动漫、迪士尼、皮克斯、Chibi/Kawaii、复古卡通、混合），支持单人/多人，强调可爱夸张表情、活力节日元素。可上传真人照片作为参考，转为卡通版本（保留可辨识面部特征）。
｛Prompt：
based on the uploaded reference photo (if any), transform into [anime / Disney / Pixar 3D / chibi kawaii / vintage cartoon / mixed art style] illustration of [a cute character / multiple adorable friends / a group of whimsical figures] celebrating New Year 2026, [solo big expressive eyes / group cheering with party hats] [holding sparklers / toasting / blowing horns], vibrant festive background with confetti explosion, balloons, 2026 banner and magical fireworks, [detailed digital art / soft pastel colors / exaggerated joyful expressions / polished animation style], high energy cute and fun atmosphere, masterpiece｝
压箱底的超级人像办法分享
作者: @就是玩AI
来源: X
！！📦📦📦压箱底的超级人像办法分享给大家，还是再讲一下捏脸这个事情，其实 你得按照香蕉逻辑控图，保持人物面容姣好的秘诀就是按照图层写提示词，秘诀我放在了附图4（自行补充一些负面提示词即可）
中文提示词：
写实摄影，超清晰特写，同一位女性的真实生活照片，高度一致的角色，每次面部和身体都完全相同，杰作，最佳质量，8k分辨率，照片级真实感，
主体：惊艳的中国美女，优雅的古典与现代魅力结合，精致细腻的面部特征，柔和经典的鹅蛋脸或瓜子脸，脸颊带有自然的柔和丰盈感以呈现优雅的年轻外观，自然的低发际线，浓密蓬松的黑色长发（直发或带有古典韵味的微卷），头发被水浸湿，丝滑的发丝紧贴面部和颈部，
眼睛：高度细节化、大而富有表现力的杏仁眼或丹凤眼，柔和的双眼皮，柔和迷人的目光，结合了古典温柔与微妙的诱惑魅力，能唤起安静的渴望，浓密修长的自然睫毛，优雅的卷曲，深邃清晰的深棕色或黑色虹膜，带有柔和诗意的高光和水润的倒影，
面部：高挺精致、笔直而柔和的鼻梁，完美的古典比例，微妙的高光，小巧精致的鼻尖，饱满自然的樱桃唇或玫瑰花蕾唇，带有柔和的光泽色调和优雅的丘比特弓形唇线，明亮温柔的微笑，露出完美洁白整齐的牙齿，充满温暖魅力，精致柔和的下颌线，具有古典的柔和轮廓，
皮肤：无瑕、明亮如玉的瓷肌，带有自然的细微瑕疵，散发出光泽湿润的光泽（因水而增强），湿皮肤，大量逼真的水滴，妆容没有晕染，
妆容：精致细节的中式优雅自然妆容，细长的拱形眉毛，柔和的眼线突出眼型，自然的红润脸颊，微妙的高光，带有光泽色调的嘴唇，强调古典的新鲜性感，
场景：浸泡在游泳池的水中，只有裸露的肩膀和锁骨露出水面，水面以上不可见任何衣物或肩带，真实的人体，自然的纤细优雅比例，奢华的室内或室外游泳池，水位在上胸部，
灯光：柔和温暖的黄金时刻自然阳光，反射在水面上，在湿润的皮肤和妆容上形成闪烁的高光，
技术：面部和眼睛对焦清晰，高度细节化的真正湿皮肤纹理，水珠与妆容的互动，自然的光影，浅景深，使用佳能 EOS R5 全画幅相机拍摄，9:16 纵横比.
负面提示词：动漫，动漫风格，动漫脸，2D，插画，绘图，绘画，卡通，漫画，韩漫，日漫，3D渲染，CG，CGI，辛烷值渲染，虚幻引擎，Blender，Maya，3D模型，游戏角色，VTuber，虚拟头像，玩偶般，蜡像，塑料皮肤，玻璃皮肤，过度平滑的皮肤，磨皮，假皮肤，修饰过的脸，液化变形，过度的美颜滤镜，抖音滤镜，TikTok滤镜，美颜App滤镜，偶像海报，KOL脸，素材模板，过度对称的脸，完美对称，千篇一律的网红脸，AI生成，人工的，明显的AI伪影，低质量，模糊，畸形，扭曲，丑陋，糟糕的解剖结构，多余的肢体，缺失的肢体，变异的手，画得不好的脸，不一致的脸，不同的脸，脸部变化，多张脸，低分辨率，JPEG伪影，噪点，过曝，欠曝，水面以上可见衣物，可见肩带，内衣肩带，比基尼肩带，文字，水印，标志，签名，画框.
冬季雪景与韩国偶像
作者: @就是玩AI
来源: X
冬季雪景提示词：Photorealistic, raw photo, social media amateur phone camera shot, anti-AI artifact, 8k UHD, hyper-detailed, film grain texture, natural skin pores, realistic skin micro-details, best quality, masterpiece
1girl, solo, 18-year-old Korean top-tier high-visual K-pop female idol face, extremely beautiful, flawless symmetrical face, large sparkling doe eyes, delicate features, snow-white porcelain skin, smooth flawless skin with natural subtle imperfections, glowing fair complexion, slight natural blush, no makeup filter effect
Korean sweet beauty style, slightly tipsy makeup, glossy pink lips, subtle rosy cheeks, long eyelashes, visible aegyo-sal, natural eye creases
slim tall model physique, very long slender limbs, long legs, large bust, voluptuous yet elegant figure, natural body proportions, no over-retouching
long dark wavy hair, loose flowing hair blowing gently in wind, realistic hair strands, natural hair volume
wearing large fluffy black fur hat, Russian ushanka style, soft luxurious fur texture with individual fur fibers, no plastic-looking fabric
tight black bodysuit, form-fitting black thermal jumpsuit, long sleeves, high-cut leotard style bottom, minimalist sleek fashion, emphasizing natural curves, realistic fabric wrinkles and folds
back view, 3/4 rear angle, walking away in snowy field but turning head back over shoulder, looking directly at viewer with gentle warm smile, soft eye smile, eye contact, candid natural pose, playful and sweet expression, dynamic body movement
snowy winter landscape, vast open snow-covered field, clean white snow ground with natural unevenness, leafless winter trees and forest edge in distant blurred background, shallow depth of field, ar9:16
overcast sky, soft diffused natural daylight, cool blue-toned color temperature, serene chilly winter atmosphere, natural environmental light, no harsh lighting
负面提示词
ugly, deformed, noisy, blurry, low contrast, distorted, cartoon, anime, 3d render, painting, illustration, CGI, digital art, AI-generated look, plastic skin, doll-like features, wax figure texture, extra limbs, bad anatomy, bad hands, missing fingers, mutated, dull skin, rough skin, tanned skin, dark skin, small eyes, asymmetrical face, text, watermark, logo, interface elements, bright sunlight, warm tones, oversaturated colors, harsh shadows, low resolution, compression artifacts, over-retouched skin, smooth plastic fabric, unnatural body proportions, flat chest, small bust, underage, childlike features, perfect airbrushed skin, fake fur texture, stiff hair, static pose, sharp artificial background
看得出区别吗？
未知内容
作者: @就是玩AI
来源: X
https://t.co/BrOXAT4bcz
公布提示词答案
作者: @就是玩AI
来源: X
Subject：Cinematic hyper-realistic movie stills in comic page format, full comic page with 5 panels and thick black panel gutters, ultra-detailed photorealistic rendering, shot on 70mm IMAX film with professional cinema camera, glossy neon candy palette (electric cyan, magenta, violet, acid lime), extreme glitch-sparkle particles and star confetti explosively drifting across every panel, heavy halftone + screenprint texture overlay, ultra high contrast cinematic lighting with luminous volumetric gradients and god rays. Subject: photorealistic live-action young adult woman with the beautiful face of a Korean K-pop female idol, highly refined and symmetrical facial features, flawless porcelain skin, large sparkling doe eyes with perfect eyeliner, small V-shaped face, delicate nose, plump lips, professional idol-grade makeup and lighting, long flowing vibrant blue hair, delicate butterfly/star earrings, wearing a simple black long-sleeve mini dress and knee-high boots, walking through a surreal massive crystal forest with enormous icy shards and intense prism caustics and rainbow refractions. She looks curious and deeply awed. Panel plan (strict layout): Top-left: extreme close-up portrait, hair dramatically sweeping in wind, galaxy glitter and lens flares behind her face. Top-right: wide full-body cinematic shot in the crystal forest, dramatic soft rim lighting and volumetric glow around her hands. Middle-left: dramatic foreshortened hand reaching directly toward camera, massive prismatic energy streaming like intense rainbow aurora ribbons with heavy particle effects. Bottom-left: two hands in elegant mid-gesture, overwhelming sparkling cosmic light pouring and overflowing between fingers with intense sparkle and bokeh. Bottom-right: hand unleashing a massive glitchy neon energy burst, explosive light rays, pixel fragments, chromatic aberration, and shockwave effects. Typography (optional): huge onomatopoeia in bold gradient comic lettering with Japanese text, thick white outline and heavy drop shadow: 「星の光！」「幻視！」「粒子嵐！」 Maximum cinematic special effects, intense lens flares, anamorphic bokeh, motion blur, dynamic camera angles, epic scale, ultra-detailed VFX, vibrant, energetic, breathtaking, photorealistic Korean K-pop idol face seamlessly integrated into hyper-stylized surreal environment.
D+奥特曼限时发售通用版
作者: @就是玩AI
来源: X
# Prompt:
masterpiece, best quality, ultra detailed, 8k, high resolution, photorealistic cinematic photography, movie-grade shot, professional cinema camera, Arri Alexa look, dramatic film lighting, sharp focus, shallow depth of field, realistic lens flare and bokeh, volumetric god rays, subtle film grain, hyper-realistic textures and reflections,
(character focus:1.65),
1girl/1boy, solo, [角色名称] [作品名称] [genderbend/full form/human host] version, [头部描述 e.g. sealed metallic helmet head with glowing visor OR realistic human face from reference], [身材描述 e.g. voluptuous feminine body OR muscular male body], [胸部/身材强调 e.g. extremely large breasts deep cleavage exposed OR athletic build],
(strictly side profile view, body turned exactly 90 degrees sideways:1.8),
(direct intense eye contact with viewer OR glowing visor eyes locked on viewer:1.9), eyes sharply focused forward,
wearing [服饰/铠甲/套装完整描述 e.g. iconic silver purple red metallic armor suit with deep V-cut chest OR blue armored dress], [关键退片 e.g. glowing color timer on chest, Excalibur in hands], ultra dynamic violent battle pose, floating mid-air, ferocious forward lunging stance unleashing powerful energy blast, overwhelming explosive [光效颜色 e.g. multi-colored light / pink flame / golden] aura bursting from body, hair/armor/cape violently torn backward by massive force, 100% faithful to original [作品] design with [optional: sexy fan redesign OR hyper-realistic tokusatsu style],
(full body [realistic/reference] from uploaded photo if applicable:1.65),
[optional: use uploaded full-body real person photo as reference for exact face/body/suit/pose, IP-Adapter full weight 0.8~0.9, perfect fusion into cinematic hybrid style, realistic skin pores natural imperfections blended with dramatic lighting]
(environment & core visual),
ancient ruined [场景 e.g. city battlefield / fantasy temple / apocalyptic landscape] in real world, photorealistic crumbling architecture, scattered debris, dark apocalyptic night atmosphere, extreme dramatic cinematic contrast, thick realistic smoke dust volumetric fog, god rays piercing through clouds, professional film lens distortion, movie-grade epic composition,
giant ominous [反派/威胁 e.g. Golza silhouette / demon king shadow] looming in distant background, massive shadowy presence, glowing ominous red eyes, immense oppressive aura,
(special effect - ultra violent explosion:1.55),
apocalyptic violent radial detonation of shattered glass originating exactly from character's center, realistic physical glass physics, razor-crisp refraction, blinding specular reflections, violent prism rainbow chromatic aberration, deep micro-cracks spiderweb fractures, savage jagged edges, cinematic slow-motion feel,
(glass fragments dramatic size gradient from center to periphery:1.75),
(largest massive sharp glass blocks and huge jagged chunks clustered tightly right around the character body, extremely large fragments with brutal thickness deep cracks near center:1.65),
(gradually decreasing in size outward: medium shards mid-distance, small sharp pieces further, tiniest glittering micro-fragments at edges creating immense cinematic depth:1.65),
tens of thousands of razor-sharp glass pieces blasting in every direction at hypersonic speed, intense motion blur trails speed lines debris streaks, violent outward shockwave burst as if reality torn apart by energy,
each glass shard contains hyper-detailed pure anime/tokusatsu style cinematic screenshot from [作品名称] (vivid colors, dramatic composition, iconic emotional moments:1.6), strictly anime/tokusatsu art style inside fragments only, NO realistic elements inside shards, famous scenes such as [列出 5-10 个经典名场面, 用逗号分隔], and other unforgettable highlights,
massive violent burst of [光效描述 e.g. multi-colored light particles / pink explosive flames], blinding energy sparks, raging shockwave rings, volumetric light beams shattered wildly through chaos, cinematic lens effects,
Negative prompt: human face exposed on helmet head, visible mouth nose (unless intended), flat chest small breasts (if large bust wanted), covered cleavage (if exposed wanted), full cartoon, pure anime on realistic parts, chibi, smooth doll skin, plastic texture, no pores, no natural imperfections, deviation from canon design, anime screenshots inside shards turned realistic, low quality, blurry, deformed, ugly, bad anatomy, extra limbs
overall: overwhelming violent cinematic composition, absolute destructive power, extreme chaotic motion with powerful radial depth perspective, god-level movie contrast and dramatic lighting, ultra sharp hyper-detailed photorealistic rendering, masterpiece quality, consistent perfect side profile, strongest center-to-periphery size gradient in exploding glass fragments (biggest at center, smallest at edges), glass shards filled with pure iconic [作品] anime/tokusatsu scenes in surreal contrast to realistic movie world
漫画书破窗主题提示词
作者: @就是玩AI
来源: X
In the ultra-realistic 8K scene, the person in the photo walked out and broke the panel frame of a giant open Japanese version of the comic book called "｛就是玩AI}" lying on the floor of the aesthetic bedroom. The comic page depicts a surreal 4-panel Japanese version of the black-and-white comic, with screen tone, bold ink lines and authentic Japanese layout (read from right to left). According to the content and character design in "{Your Name}", it features a comic version with the same person trapped in it, wearing, styling
裸眼3D时装风格生图
作者: @就是玩AI
来源: X
裸眼3D时装风格，生图应该是12层了，不是九宫格了，稍微做了修改
{
"title": "时尚杂志 9 宫格叠层写真",
"prompt": "cold-beauty early-20s Chinese/Korean fashion model, sharp jawline, high cheekbones, aloof and minimalist expression, very editorial face",
"outfit": "oversized black cashmere V-neck sweater (slightly loose fit), high-waisted wide-leg pure white tailored trousers, black leather loafers with subtle gold horsebit",
"layout_and_composition": "Surreal creative collage. Background is a CLEAR and SHARP 3x3 grid of 8 different lifestyle poses, separated by DISTINCT THICK WHITE LINES. FOREGROUND is dominated by a single, massive, hyper-realistic full-body shot of the same model in the absolute center. The central figure overlaps and breaks the white grid boundaries, appearing to be on a separate front layer.",
"central_figure_pose": "Dynamic full-body shot from head to toe. The model is walking forward. Her head touches the very top edge of the canvas, and her shoes touch the very bottom edge. She occupies the maximum vertical space, creating a \'pop-out\' 3D optical illusion.",
"background_grid_poses": "8 distinct variations in the background slots (sitting, standing, looking side, etc.). IMPORTANT: All faces in the background grid must be sharp, clear, and fully visible. No blur.",
"hair_and_accessories": "neat low bun with slightly messy front strands, small gold hoop earrings, thin gold chain necklace",
"environment": "Bright minimalist indoor studio with concrete walls and green plants. Both the background grid and the foreground figure are fully illuminated and sharp.",
"lighting": "Even, bright studio lighting to ensure clarity on all faces in the grid. Subtle rim light on the central figure to aid separation.",
"camera": "Deep depth of field (f/16) to keep BOTH the background grid figures and the foreground figure in sharp focus. No bokeh, no background blur.",
"mood": "Breaking boundaries, high-fashion editorial, surreal depth, quiet luxury, confident, clear and graphic",
"negative_prompt": "blurred background, depth of field, bokeh, out of focus faces, messy composition, missing grid lines, low resolution, deformed limbs, merging bodies, dark shadows",
"technical": {
"aspect_ratio": "9:16",
"notes": "Crucial: The background grid lines must be white and visible. The faces in the background grid must be high resolution and clearly visible, not blurry.",
"recommended_model": "Flux 1.1 / Midjourney V6 / Stable Diffusion XL"
},
"midjourney_command": "/imagine prompt: a sharp 3x3 fashion grid collage with distinct white grid lines, featuring a chinese model in black sweater and white trousers in various clear poses. OVERLAID by a massive hyper-realistic full-body 3D cut-out of the same model walking forward in the center. All faces sharp and in focus, deep depth of field, no blur, pop-out 3D effect, editorial photography, 8k --ar 9:16 --v 6.1 --stylize 250"
}
发型设计提示词生成器
作者: @就是玩AI
来源: X
这是一个能帮你设计想要的发型设计说明书风格的提示词，可以根据你想要的发型类别生成对应的修剪程序和最终展示效果，男女都可以适配，关键词已匹配。
{
"prompt": "Generate an extremely detailed [hairstyle name, such as: American front thorn] hairstyle design guide, the image ratio is 21:9. Each view should be a high-resolution, realistic 3D rendering image. The model is [model description, for example: Asian young man, medium figure, long face type], and the background is a pure white studio environment to ensure that the details of the hairstyle are visible in high definition, such as single hair texture, color gradient (if there is high or low light), solid The complete steps are as follows: 1. B) Use straight scissors and fine-toothed combs, hold the comb horizontally as a guide line, and cut the scissors in parallel (illustration: measure from behind the ear to the neck, centimeter ruler). Layer creation: a) Vertical partitioning, raise the hair at a 90-degree angle. Final shape: a) Blow-dry: use a round comb to lift from the root to create volume (illustration: wind arrow). Nursing), potential risk warnings and professional certification badges; ensure that the guidelines are comprehensive and printable for professional training, salon use or DIY enthusiasts.",
"parameters": {
"aspect_ratio": "21:9",
"quality": "The description text uses Song Simplified Chinese, ultra-high resolution, 8K, extremely detailed texture and rendering"
},
"negative_prompt": "Blury or low-resolution details, cartoon or abstract style, non-Simplified Chinese font, asymmetrical or unprofessional hairstyle, messy or colored background"
}
电视节目同步角色设计
作者: @就是玩AI
来源: X
This prompt can be synchronized with the characters you want to cushion according to the broadcast content of the TV program. If you are interested, The original prompt is random, and the role can be changed according to the next instructions.
Prompt:
1. 提示词：温馨真实的客厅，沙发上坐着一个人（参照图中的人，人脸100%完全一致，五官细节绝对同一人），他原本正聚精会神盯着前方电视机，身体微微前倾，突然听到声音或感觉到有人在拍他，于是猛地回头直视镜头，眼神惊讶或带着玩味的笑，关键核心：此时他已经实时、完美、100%同步cosplay电视屏幕里正在出现的那个随机角色，服装、发型、道具、妆容、姿势细节全部与电视里当前角色一模一样（电视随机播放全世界任何影视/动画/游戏经典角色，比如哈利波特、灭霸、杰克船长、初音未来、死侍、艾莎、乔克、汪达、甘道夫、鸣人、沃尔特·怀特、任何角色都可能），连衣服褶皱、道具磨损、光影都完全一致，回头瞬间头发微动，表情生动，电视蓝光打在侧脸，超现实却又极其真实的高级场景，电影感极强，8K，摄影大师构图，锐利细节，真实皮肤纹理
负面提示词：
模糊，低质，两个人，电视里出现人脸，畸形手，多余肢体，文字，水印，服装不一致，道具错误，姿势不同步，没回头，正脸看电视，眼神涣散。
2. Cinematic cozy living room, one person (100% identical to reference image, perfect face swap) sitting on couch, originally leaning forward completely absorbed in watching TV straight ahead, suddenly hears something and whips their head around to look directly at the viewer with a surprised/playful/shocked expression, caught in the act. Crucial rule: at this exact moment they are perfectly cosplaying in real-time whoever is currently on the TV screen — any random iconic character from movies, shows, anime, games (Harry Potter, Thanos, Captain Jack Sparrow, Hatsune Miku, Deadpool, Elsa, Joker, Scarlet Witch, Darth Vader, Naruto, Walter White, literally anyone), 100% identical costume, hairstyle, makeup, props, fabric details, wear and tear, pose frozen mid-turn, hair in motion, TV glow illuminating half the face, surreal yet hyper-realistic, masterpiece, 8K, sharp focus, photorealistic skin, dramatic side lighting
现代希尔顿酒店雨夜客房摄影风格
作者: @就是玩AI
来源: X
{
"prompt": "Night hotel style",
"tips": "Highly detailed real photos of the bedroom of the modern luxury Hilton Hotel on a rainy night, camera-level style photos, warm ambient lighting. The foreground is a neat but slightly wrinkled white bed with daily necessities scattered on it: a black jacket with gold buttons, a white shirt, a dark blue skirt, a milky white turban, a high school student\'s schoolbag, an exquisite gold necklace and randomly arranged earrings, an elegant small handbag and a pair of hanging Transparent black stockings on the edge. In the background, a large frosted glass shower door is covered with condensed and gentle rain patterns. Behind it is a soft female silhouette, with elegant long hair (no corners or exposed features) behind the blurred glass. In the rain, she elegantly raises her arms to wash her hair, elegantly and relaxed.",
"analysis": {
"posture": "The silhouette naturally turned his back to the audience, gently raised his arms to his head, in the classic \'just washed his hair\' posture, relaxed shoulders, elegant and calm, not sexy.",
"background": "Clean Hilton hotel room, large bed, simple decoration, floor-to-ceiling windows with the sound of rain, frosted shower area integrated into the bedroom, soft neutral color palette.",
"lighting": "The warm bedside lamp on the left (2700K) + the cold diffuse light on the right (5600K), through the frosted glass, creates beautiful light and dark and soft edge light on the outline.",
"gesture": "The daily attitude of someone after taking a shower after coming back from the rain is natural and peaceful.",
"style": "Clean real-life realistic, photo style, similar to the sense of movie-level lens",
"mood": "Comfortable, slightly melancholy, romantic rainy night atmosphere, completely safe work"
},
"parameters": {
"aspect_ratio": "16:9",
"resolution": "8k",
"quality": "Masterpiece, the best quality, super meticulous",
"negative_prompt": "nsfw, nude, underwear, underwear, sexual pose, horn, tail, low resolution, blur, text, logo, watermark, exposed clothes, porn"
},
"producer": "94vanAI"
}
开心鸵鸟
专业生活方式照片，主题靠在白色墙上
作者: @开心鸵鸟
来源: X
感觉还是差点意思，该怎么改呢？
gemini Nano Banana Pro prompt: Professional lifestyle photo using the uploaded image, subject leaning against a white wall. Beside them, a high-contrast polygon art illustration of their face painted on the wall.
Aakash Gour
Nano Banana Pro 爆红！
作者: @Aakash Gour
来源: X
🚨Nano Banana Pro is going viral!
People are using a new Gemini prompt to make bright, fun 3D caricatures.
Users on X are posting shiny, colorful AI portraits — and the trend is growing fast.
Here’s the viral prompt:
“A highly stylised 3D caricature of the person in the uploaded image, with expressive facial features and playful exaggeration. Smooth, polished look with soft lighting and a bold colour background.”
Give it a try and see what you get.
Aaliya
快速创建演示或网站
作者: @Aaliya
来源: X
Imagine creating a full presentation or website faster than making tea.
Gamma @Gammaapp + Nano Banana Pro just made it real.
This update is dangerous for procrastinators (and game-changing for creators).
使用Nano Banana Pro创作的广告
作者: @Aaliya
来源: X
Created these advertisements with Nano Banana Pro 😍
Prompt: Cinematic 3D action-packed advertisement for [Product], captured in an intense mid-motion moment with dramatic studio lighting, dynamic particle effects, and high-impact slow-motion energy. Ultra-hyperrealistic rendering, razor-sharp details, glossy commercial finish, atmospheric depth, and powerful contrast. Viral-ready composition with the brand logo seamlessly integrated into the scene and a sleek, modern slogan positioned cleanly beneath. High-end blockbuster commercial aesthetic.
荒井健一 AK
《三国演义》名场面3D海报设计
作者: @荒井健一 AK
来源: X
これいいね。
prompt：请为影视剧/小说《三国演义》设计一张高品质的3D海报，需要先检索影视剧/小说信息和著名的片段场景。
首先，请利用你的知识库检索这个影视剧/小说的内容，找出一个最具代表性的名场面或核心地点。在画面中央，将这个场景构建为一个精致的轴侧视角3D微缩模型。风格要采用梦工厂动画那种细腻、柔和的渲染风格。你需要还原当时的建筑细节、人物动态以及环境氛围，无论是暴风雨还是宁静的午后，都要自然地融合在模型的光影里。
关于背景，不要使用简单的纯白底。请在模型周围营造一种带有淡淡水墨晕染和流动光雾的虚空环境，色调雅致，让画面看起来有呼吸感和纵深感，衬托出中央模型的珍贵。
最后是底部的排版，请生成中文文字。居中写上小说名称，字体要有与原著风格匹配的设计感。在书名下方，自动检索并排版一句原著中关于该场景的经典描写或台词，字体使用优雅的衬线体。整体布局要像一个高级的博物馆藏品铭牌那样精致平衡。
Abdullah AI
超现实主义夸张肖像
作者: @Abdullah AI
来源: X
Hyperrealistic Caricature Portrait
Google Gemini ai Nano Banana 3 Pro ✨
Prompt ⤵️
{
"style": {
"type": "hyperrealistic caricature",
"lighting": "soft studio lighting",
"mood": "serious, dramatic",
"details": "exaggerated facial proportions, smooth skin texture, painterly realism"
},
"subject": {
"type": "older man",
"features": {
"hair": "light blond, voluminous, swept to the side with dramatic flow",
"face": "large exaggerated cheeks and chin, deep frown lines, pronounced brows, slightly drooping eyelids",
"expression": "stern, pouting, serious"
},
"clothing": {
"outfit": "dark formal suit",
"shirt": "light-colored dress shirt",
"tie": "light blue textured tie"
},
"pose": "straight posture, facing forward"
},
"background": {
"type": "plain dark gradient",
"atmosphere": "studio portrait, minimalistic"
},
"camera": {
"angle": "straight-on",
"depth_of_field": "shallow, soft background",
"focus": "sharp on
Abhi
青年男子 mugshot
作者: @Abhi
来源: X
1. Open Gemini / Chat GPT
2. Upload your Picture
3. Prompt :
Photorealistic mugshot portrait of a young man, early 20s, use face from the provided image ,masculine features, sharp jawline, intense tired eyes, long straight black hair slightly messy and falling over face, pale skin with dirt smudges, bruises and grime on face and neck, serious sullen expression, wearing dirty orange prison jumpsuit with stains and smudges, gray undershirt visible, standing in front of police height chart wall with "POLICE" text and measurement lines, one hand raised touching his hair behind head, other hand holding a black and white striped film clapperboard at chest level with multiple silver rings on fingers, clapperboard text clearly readable: circled "ARRESTED", "For illegally Money", "12-28" and "25-26", high detail, dramatic lighting, realistic textures, gritty atmosphere, professional photography --ar 3:4 --v 6 --q 2 --stylize 250
风暴海边的英雄独白
作者: @Abhi
来源: X
1. Open Gemini / Chat GPT
2. Upload your Picture
3. Prompt :
A hyper detailed cinematic scene featuring use uploaded face 100% as reference as a solitary figure gazing out at a tempestuous sea. The camera captures the hero in the center of the frame with absolute clarity and dramatic presence. Clad in a heavy, dark waxed canvas jacket with brass hardware and a deep hood, over a thick fisherman's knit sweater, and rugged denim jeans, the textures of the weather-beaten materials are palpable. Standing on windswept, rocky cliff edge overlooking a churning, grey ocean under an angry, cloud-filled sky. Sea spray catches the dim light, and gulls circle overhead. heavy atmosphere, dramatic and brooding, with the raw power of nature emphasized. Use language inspired by premium photography and expert level photo manipulation techniques similar to high end Photoshop to achieve top quality realism without CGI or 3D graphics. Create a one of a kind, unforgettable scene with unmatched detail, cinematic power, and full focus on the hero. no CGI or 3D graphics or art. use Photoshop manipulation if the scene illogical and imaginary.
海边日落时分专注弹奏吉他的男子
作者: @Abhi
来源: X
1. Open Gemini / Chat GPT
2. Upload your Picture
3. Prompt :
Create an ultra realistic photo with 8k sharp details, cinematic quality, natural light, soft focus of the man in the attached photo. Copy his face and hair 100%, do not change any of the facial feature.
A man dressed in earthy, functional clothing: a dark teal or charcoal linen button-down shirt with the sleeves rolled up, paired with olive-green cargo pants. On his feet, he wears tan suede chukka boots. He is deeply focused on playing a standard six-string acoustic guitar with a natural wood finish. He is sitting cross-legged on the edge of a vehicle's roof rack. His head is bowed toward the guitar strings. The overall mood is tranquil and solitary. The scene is set in a coastal or cliffside location during \"golden hour.\" The primary prop is a classic silver Land Rover Defender. A black metal roof rack serves as the subject's seat. The vehicle is parked facing the ocean. The lighting is soft and directional, coming from the setting sun on the right side of the frame. This creates a gentle glow on the subject's face and the body of the guitar, while leaving parts of the car in a soft shadow. The background consists of a vast, calm ocean and a soft, hazy sky transitioning from pale blue to a warm orange-pink. To the left, a dark, silhouetted cliffside adds depth to the landscape. The photographer used a medium-wide shot with a slightly low angle, looking up at the subject.
There is a shallow depth of field, which slightly blurs the distant horizon. The color grading is "moody" with desaturated blues and warm highlights. Negative Prompt:
No oil painting. No drawings. No fantasy. No anime. No cartoon. No 3D render. No soft diffused illuminating skintone.
简约紫色背景人像摄影
作者: @Abhi
来源: X
1. Open Gemini / Chat GPT
2. Upload your photo
3. Prompt :
Use the uploaded image as a STRICT identity reference. Preserve 100% of the person’s face and identity. Do not change facial features, proportions, skin texture, or appearance.
Ultra-realistic editorial studio portrait of the same person sitting on a bright purple bench in a minimalist studio. Oversized black t-shirt, black pants, crisp white sneakers. Casual pose: leaning back slightly, one arm resting on the back of the bench, the other relaxed on the leg. One sneaker extended toward the camera, sole clearly visible, dominating the foreground with strong perspective.
Plain lilac/purple background, no text, no logos, no graphics, no clutter. Soft balanced studio lighting with natural highlights and shadows, emphasizing textures of clothing, sneakers, and bench. Low-angle shot, wide-angle lens (20–28mm). Subject’s face sharp and proportional, natural wide-angle perspective only.
Bold monochromatic purple aesthetic, modern editorial fashion photography, clean studio look, ultra-detailed, photorealistic, 4K quality.
现代圣诞老人特写
作者: @Abhi
来源: X
1. Open Gemini / Chat GPT
2. Upload your picture
3. Prompt :
Transform the original attach