提示词库 · 创作探索

提示词库

探索真实提示词和对应生成示例,按任务找到灵感,免费复制完整提示词,开启下一次创作。

示例
11,025
模型
18
来源
1277
11,025 条提示词示例 · 第 191 / 230 页
Example generated from “Web3 Ecosystem Food Chain”
GPT Image 2en

Please help me generate it.{argument name="market" default="web3"} Different levels of roles in the market, to differentiate{argument name="market" default="web3"} The ecological food chain. The image should include detailed descriptions of the roles and have an aspect ratio of [missing information].{argument name="aspect ratio" default="16:9"} .

Example generated from “Weekend Seoul travel guide poster”
GPT Image 2zh

Generate a polished one-page Chinese travel guide poster for a fast weekend trip from Nanjing to Seoul in May. Use LARGE highly legible Chinese text, short phrases only, and no paragraph blocks. Focus on shopping, skincare, and a stylish Seongsu-dong route. Layout: big title, 4 modules only (行程 / 区域推荐 / 购物清单 / 美妆护肤), each with 2 to 4 short bullet points, plus a small cute route map with icons. Clean editorial infographic style, soft pastel colors, neat spacing, high readability, modern Xiaohongshu travel card aesthetic. Avoid tiny text, avoid dense explanations, avoid garbled characters.

Example generated from “West Bengal City View”
GPT Image 2en

One{argument name="location" default="west bengal"} city

Example generated from “Western Open-World Game Screenshot”
GPT Image 2en

A highly realistic screenshot from a video game, in a style that is {argument name="game style" default="Western Open World Games"} The image is shot from a third-person, over-the-shoulder perspective. A rugged cowboy walks with his back to the camera, wearing a worn brown jacket, dark trousers, a holster at his waist, and a classic cowboy hat. To his left is a pair of horses with saddles.{argument name="horse type" default="Black horse with white spots"} The scene is set in{argument name="time of day" default="A vibrant golden hour sunset"} Below{argument name="environment" default="The dusty desert landscape is dotted with withered tall grass, dirt paths, cacti, and towering red rock terraces."} Against a hazy background, two silhouettes stand on a dirt path in the distance. The image contains a video game HUD overlay: a small circular map with a player arrow and two red dots in the lower left corner; and three lines of white text and button icons in the lower right corner, displaying "DRAW WEAPON [RT]", "HORSE CARGO [Cross D-pad]", and "TENNESSEE WALKER [LT]".

Example generated from “Wet Summer Anime Schoolgirl Portrait”
GPT Image 2en

This is a dazzling anime-style summer portrait, the protagonist being a {argument name="character type" default="Japanese high school girl"} The image shows a close-up of her profile from the chest up. She is standing outdoors, with a clear blue sky and bright white clouds in the background. She has wet hair.{argument name="hair color" default="Dark blue-black"} She has a short bob haircut, large eyes that shimmer with a blue-violet light, and a faint blush on her cheeks. Her expression is calm yet slightly aloof. Her white sailor-style school uniform shirt is soaked, appearing semi-transparent and realistically clinging to the folds. The collar is a dark navy blue sailor collar, tied with a large dark navy blue ribbon at her chest. She raises one arm, holding a transparent, textured plastic water bottle to her forehead, cool water droplets sliding down her face, neck, arm, and clothing. The image is filled with intense summer sunlight, bright highlights, rim light penetrating her wet hair, iridescent lens flares, tiny floating bubbles, and shimmering reflections. The background includes a soft-focus outdoor school pool area, showing metal railings, pool stairs, and the shimmering water. The emphasis is on the crystal-clear water bottle, the sheen of her wet skin, delicate eyelashes, subtle anime-style rendering, dramatic backlighting, the refreshing atmosphere of a summer heatwave, and the sophisticated high-end illustration style.

Example generated from “WhatsApp Group Profile Avatar”
GPT Image 2en

for{argument name="group type" default="WhatsApp Group"} Generate an avatar to decide{argument name="purpose" default="Friday dinners and restaurants"} Related matters.

Example generated from “Whimsical Animated Townsfolk Lineup”
GPT Image 2en

A clean character lineup sheet on a pure white background, showing 13 full-body stylized 3D characters in a whimsical stop-motion-inspired design with exaggerated proportions, textured fabrics, worn surfaces, muted earthy colors, and slightly creepy yet charming personality. Arrange them in 2 rows: 7 characters across the top row and 6 across the bottom row, each isolated with soft grounding shadows and plenty of white space, like a studio casting board for an animated film. Top row, left to right: 1) an extremely tall and painfully thin middle-aged man in a shabby brown suit, white shirt, narrow red tie, and checked trousers, with long fingers and slouched posture; 2) a short heavyset older woman with curly reddish hair, wearing a pink cardigan over a floral dress, sensible shoes, and carrying a brown handbag; 3) a very large burly man with tattooed forearms, wearing a red plaid flannel shirt, suspenders, dark worn pants, heavy boots, and a rough worker look; 4) a skinny teenage goth girl with pale skin, black messy ponytail, black jacket, dark shirt, skinny jeans, crossed arms, and a bored expression; 5) a lanky teenage boy in a knit beanie, layered hoodie and jacket, skinny jeans, and sneakers, hands in pockets; 6) a very tall elderly woman with an elegant beehive hairstyle, narrow face, tan blouse, scarf, long brown skirt, heels, handbag, and a cigarette held in one hand; 7) a slim teenage or young adult male with curly dark hair, wearing a striped sweater, narrow pants, and brown shoes. Bottom row, left to right: 8) a gaunt frail old person in ragged layered clothing and a stained apron, stooped forward with tired posture; 9) a thin middle-aged woman with hair in a high bun, mustard cardigan, long blue dress, flats, and a shoulder bag, holding something delicately in front of her chest; 10) a stocky police officer in a bulky navy winter uniform with fur collar, badge, utility belt, dark pants, and black shoes; 11) a chubby curly-haired boy wearing a varsity-style sleeveless jacket over a hoodie, knee-length shorts, socks, and sneakers, gripping his backpack straps; 12) a skeletal elderly man with a cap, huge nose, hunched spine, long coat, skinny trousers, and a cane; 13) a lanky mechanic or gas station worker in a dark blue jumpsuit and cap with the visible text {argument name="cap text" default="GAS"}, holding a wrench. Keep the rendering highly detailed with felt, wool, denim, knit, leather, and dust-like material realism, but maintain caricatured anatomy and expressive silhouettes. Use a subdued palette of browns, rust reds, dusty pinks, charcoal, navy, beige, and faded denim. The overall mood should feel like quirky residents of a rundown small town from a darkly comic animated feature, presented as a polished ensemble cast sheet.

Example generated from “Whimsical Childlike Doodle Reconstruction”
GPT Image 2en

Reconstruct this image as a single flat illustration infused with playful, childlike doodle elements. Use a bold, vibrant, and whimsical color palette, simplifying every detail into clean, flat shapes. Incorporate a handcrafted feel with slightly imperfect, hand-drawn outlines to create a charming, imaginative aesthetic, as if it were illustrated on a sheet of white paper. The overall style should be cute, expressive, and delightfully whimsical.

Example generated from “Whimsical Folk Art Transformation”
GPT Image 2en

Reimagine the full image as a whimsical handcrafted folk-art illustration filled with playful doodle aesthetics. Replace the original colors with a {argument name="color palette" default="fresh, vibrant palette"} featuring {argument name="tones" default="bold contrasting tones and cheerful accents"}. Convert every object and detail into simplified flat forms with soft irregular outlines, textured brush-like fills, and an intentionally imperfect handmade appearance, as if sketched and colored on clean white paper with markers and paint pens. The final artwork should feel charming, naïve, and storybook-inspired, combining decorative folk patterns, quirky shapes, and lighthearted visual energy in a cute, imaginative style.

Example generated from “Whimsical Programmer Almanac Infographic”
GPT Image 2en

{ "type": "Watercolor infographic Chinese calendar", "style": "Hand-drawn style, The Little Prince theme, cream background, scattered stars.", "header": { "date": "{argument name=\"date\" default=\"April 18, 2026\"}", "title": "{argument name=\"main title\" default=\"vibe coding Old Almanac\"}", "subtitle": "Saturday, the second day of the third month of the Bingwu year" }, "layout": { "sections": [ { "title": "should", "position": "top-left", "count": 5, "labels": [ "{argument name=\"first do item\" default=\"Write code based on your intuition\"}", "Let's talk to the AI ​​before we start working.", "Delete the wrong parts, keep the interesting ones.", "Give the project a romantic name", "Have a cup of coffee and let your thoughts flow." ] }, { "title": "Today's celestial phenomena & programming horoscope", "position": "top-right", "count": 4, "labels": [ "The Moon enters Gemini", "A place where inspiration flows, ideal for dialogue and creation.", "5 stars", "Inspiration 90%, Debugging 65%, Sense of Accomplishment 88%" ] }, { "title": "avoid", "position": "mid-left", "count": 5, "labels": [ "{argument name=\"first dont item\" default=\"Write a perfect architecture right away\"}", "agonizing over technology selection until dark", "Afraid of errors, I dare not run it.", "Copy and paste without thinking", "Forgot to save and submit" ] }, { "title": "Auspicious time today", "position": "bottom-left", "count": 2, "labels": [ "09:30 - 11:30 Inspiration is at its peak", "14:00 - 16:00 Knock out a surprise" ] }, { "title": "Lucky Command", "position": "bottom-left", "count": 4, "labels": [ "git add .", "git commit -m \"follow the vibe\"", "git push", "# The universe will help you compile it." ] }, { "title": "Today's Mindset", "position": "bottom-right", "count": 1, "labels": [ "You don't need to be the best programmer, just the most perceptive creator. Let your feelings guide you, and let your code blossom." ] } ], "illustration": { "position": "center-right", "subject": "The Little Prince is using a laptop on an asteroid, with a fox, a rose in a glass dome, and a mug nearby.", "quote": "{argument name=\"quote\" default=\"What is important cannot be seen with the eyes; it must be felt with code and the heart.\"} The Little Prince and the Programmer", "sticky_notes": { "count": 3, "labels": [ "Let's get running first.", "Let AI accompany you on your exploration.", "A bug can be a gift." ] } } } }

Example generated from “Whisky Bottle on Dark Bar with Bokeh”
GPT Image 2en

A tall square-shouldered amber whisky bottle with a black wax-sealed top, standing on a polished dark walnut bar counter. Single warm tungsten back-light produces a soft halo behind the bottle and a subtle reflection on the counter. Background is a deeply blurred dark wood and brass bar interior with golden bokeh. No text on the label, no brand mark, no watermark. Square 1:1 luxury spirits product photography, moody low-key lighting, photorealistic glass and liquid refraction.

Example generated from “White-Haired Anime Swordswoman Collage”
GPT Image 2en

An exquisite anime/fantasy illustration, presented in a 2x2 grid layout, showcases four scenes of the same elegant female swordsman. The character is a young woman with extremely long platinum-white hair, tied in a high ponytail with a large, deep navy blue bow. Her hair is soft, curly, and flowing, her skin is fair, her features delicate and refined, and her eyes are bright reddish-pink. She wears a magnificent white and blue gothic aristocratic dress: a white high-necked ruffled blouse with lace trim and gold embroidery, a large navy blue bow adorned with a ruby ​​brooch at the chest, puff sleeves with ribbon cuffs, a deep navy blue corset with gold details, and a voluminous navy blue skirt decorated with gold patterns, layered ruffles, and a satin sheen. She carries a katana with a dark hilt and gold embellishments. The first panel is a close-up portrait from the chest up, her body slightly turned to the side, her hair dramatically outlined against a dark, shimmering night sky. The second frame is a dynamic upper-body shot, showing the character drawing or displaying her sword horizontally, her hair flying wildly in the wind, against a cinematic, blurred cityscape background surrounded by glowing petals. The third frame is a full-body fashion portrait, where she stands elegantly in a brightly lit hall surrounded by towering, luminous buildings, showcasing the full silhouette of her dress, bow-adorned high heels, and the sword at her side. The fourth frame is a three-quarter view of her back taken at sunset or dawn on a shimmering beach, with the character looking back against a warm, soft-toned sky and sparkling water, cherry blossom petals falling in the wind. The game employs ultra-detailed anime rendering, achieving light novel cover-level image quality, with delicate fabric textures, glossy highlights, gold accents, soft halos, dramatic backlighting, floating petals, shimmering particles, an elegant and romantic atmosphere, and a harmonious color scheme of navy blue, white, gold, and pale pink.

Example generated from “Whiteboard Marker Anime Sketch”
GPT Image 2en

One{argument name="character description" default="An anime girl with shoulder-length wavy hair and a gentle smile."} A close-up portrait, wearing{argument name="outfit" default="School uniform with bow tie"} The image is a hand-drawn sketch on a glossy whiteboard, entirely using {argument name="marker color" default="blue"} Whiteboard marker artwork. The artistic style presents a sketch-like feel, with varying line thicknesses and bold marker shading, capturing a casual yet exquisitely refined anime illustration effect. Subtle light reflections in the upper right corner of the whiteboard surface enhance the realism of the "photographed artwork."

Example generated from “Whiteboard Marker Anime Sketch”
GPT Image 2en

A realistic whiteboard photograph, on which is written {argument name="marker color" default="green"} High-detail drawing with whiteboard markers{argument name="subject" default="A samurai with messy hair and a beard, praying with his hands clasped together."} The characters are adopted{argument name="art style" default="Detailed comic sketch"} The figure is depicted in a traditional style, shown from the side, with eyes closed, wearing a traditional kimono and a samurai sword at her waist. To the left of the figure, the words "..." are handwritten in capital letters.{argument name="text line 1" default="VAGABOND"} Below that, it says "{argument name="text line 2" default="MUSASHI"} The whiteboard has a glossy surface, realistic light reflections and glare on the left side, and a thin metal frame visible at the bottom edge, creating a realistic classroom or office environment.

Example generated from “Widebody Volvo 240 Spec Board”
GPT Image 2en

{ "type": "Automotive design specifications project", "subject": { "vehicle": "{argument name=\"car model\" default=\"Volvo 240\"}", "edition": "{argument name=\"edition name\" default=\"JUMDOO EDITION\"}", "body_style": "Rendering of a 2-door sedan wide-body display vehicle", "era": "1980s", "location": "Make", "stance": "Extremely low air suspension posture, aggressive wheelbase", "paint": { "primary": "{argument name=\"primary color\" default=\"warm beige\"}", "secondary": "{argument name=\"secondary color\" default=\"Satin Black\"}", "finish": "Satin/Matte Texture" }, "design_features": [ "The square silhouette of the 1980s Volvo 240", "Wide-body riveted wheel arches", "Black front lip with extremely low ground clearance", "black side skirt", "Black lower body cladding", "Black multi-spoke deep concave wheel hub", "Low aspect ratio stretch tires", "Asymmetrical headlight design, with X-shaped inserts on one side of the grille light strip.", "Black grille with Volvo diagonal logo", "A clean, smooth body with minimalist trim.", "Center-mounted dual exhaust pipe", "Visible racing bucket seats", "Small JUMDOO brand logos are located on the front license plate area, side skirts, and rear panel." ] }, "layout": { "background": "Simple light gray photography studio poster background", "overall_style": "Minimalist Swiss-style car specification sheet, clear 3D product rendering, black sans-serif font, and thin gray dividing lines.", "title_block": { "headline": "{argument name=\"headline text\" default=\"VOLVO 240 WIDEBODY\"}", "subheadline": "JUMDOO EDITION", "position": "Top left corner" }, "info_panel": { "position": "Top right corner", "items": [ { "label": "LOC:", "value": "Make" }, { "label": "MODEL:", "value": "240" }, { "label": "BODYS:", "value": "2-door sedan" }, { "label": "ERA:", "value": "1980s" }, { "label": "BEILD:", "value": "1980s" }, { "label": "CUSTOM", "value": "Wide-body stance" } ] }, "sections": [ { "title": "Vehicle view", "position": "Left side and center", "count": 6, "labels": [ "Left front 3/4 perspective", "front", "side", "back", "back", "Left rear 3/4 perspective" ], "content": "Six individual studio renders of the same vehicle, arranged in a 3x2 grid, each with a bold title and two brief captions below or to the side." }, { "title": "Material color swatch", "position": "upper right middle", "count": 3, "labels": [ "main color", "secondary color", "Satin Matte" ], "content": "Three rectangular color/material blocks, beige, beige, and black respectively." }, { "title": "Color scheme", "position": "Right sidebar", "count": 2, "labels": [ "main color", "secondary color" ], "content": "The primary color scheme is beige, and the secondary color scheme is black." }, { "title": "Side profile", "position": "Top right corner", "count": 4, "labels": [ "front lip", "Volvo diagonal processing", "Track style", "OEM + Retro Style" ], "content": "A brief list of key points describing the side and front styling features." }, { "title": "Brand logo", "position": "Right middle", "count": 4, "labels": [ "front lip", "Side skirt", "wheel arch", "trunk" ], "content": "List of key points for marking the placement area" }, { "title": "Wheels and Fit", "position": "lower right side", "count": 5, "labels": [ "Aerodynamic hubcap", "Elegant headlights", "Wheel side skirts", "Adaptability", "attitude" ], "content": "List of key points describing wheel style, fit, and stance" }, { "title": "significant details", "position": "bottom right corner", "count": 4, "labels": [ "Asymmetrical headlight treatment", "Front lip integrated red light", "Center-mounted dual exhaust", "Visible racing bucket seats" ], "content": "Key Modification Points List" } ] }, "rendering": { "camera": "Studio product photography, similar to orthographic projection, with slight perspective, consistent and soft lighting.", "quality": "High-end industrial design projects, ultra-clear realistic 3D rendering", "aspect_ratio": "16:9 Landscape" } }

Example generated from “wikiHow Article Mockup”
GPT Image 2en

{ "type": "wikiHow article simulation", "header": { "background": "green", "elements": [ "Hamburger menu icon", "wikiHow logo", "Search bar with text \"Search method...\"", "5 navigation icons: PRO/PPRO, Quiz, Edit, Explore, Log In" ] }, "breadcrumbs": "Home » Categories » Computers & Electronics » Internet » CHATGPT", "article_info": { "title": "{argument name=\"article title\" default=\"How to use GPT-Image 2\"}", "metadata": "Written by wikiHow staff | Updated: June 10, 2024", "sidebar": [ "PDF Download Article Button", "Method box: 1. Use GPT-Image on Arena.ai 2." ] }, "intro_box": "GPT-Image 2 is OpenAI's most advanced AI image generator. It can create high-quality, detailed, and accurate images based on your text prompts. This wikiHow will show you how to try GPT-Image 2 for free on arena.ai.", "steps": [ { "number": 1, "title": "{argument name=\"step 1 title\" default=\"Go to arena.ai.\"}", "description": "Open your web browser and visit https://arena.ai. Arena is a platform where you can compare and try out thousands of AI models, including OpenAI's latest image generator.", "embedded_image": { "type": "Dark Mode Website UI Simulation", "logo": "arena", "navigation": [ "battle", "Side-by-side comparison", "direct dialogue", "Rankings", "Community", "Log in", "Register button" ], "hero_text": "The world's largest AI model arena", "input_ui": { "model_dropdown": "Model:{argument name=\"model name\" default=\"duct-tape-3\"}", "settings": "default", "text_input": "Enter your prompt here...", "button": "run" } } }, { "number": 2, "title": "{argument name=\"step 2 title\" default=\"Select model 'duct-tape-3'.\"}", "description": "Click the model drop-down menu and select \"duct-tape-3\". Duct-tape-3 is the name Arena uses for OpenAI's most powerful image generation model, GPT-Image 2." }, { "number": 3, "title": "{argument name=\"step 3 title\" default=\"Enter the prompt word.\"}", "description": "Enter your prompt in the box and click \"Run\". GPT-Image 2 will generate an image based on your description.", "tip_box": "Tip: GPT-Image 2 is currently the best AI image generator because it can generate extremely detailed, accurate and high-quality images, excels at text processing, can follow complex instructions, and has a deep understanding of the world." } ] }

Example generated from “WILDCAMP 巨型帐篷广告海报”
GPT Image 2en

An outdoor adventure advertisement poster featuring a rugged bearded man in full hiking gear standing confidently beside a massive orange camping tent three times taller than him, fully pitched in a dramatic forest clearing surrounded by towering pine trees beneath a deep starry night sky. The tent features a bold white “WILDCAMP” logo stitched onto the rainfly. Warm cinematic campfire lighting illuminates the scene with realistic shadows and rich outdoor textures, creating a premium adventure-commercial aesthetic. Large rugged serif typography reading “WILDCAMP” dominates the dark sky area in bold orange lettering, while the tagline “Sleep under the stars.” appears elegantly at the bottom. Small grey text in the top-right corner reads “Designed with GPT Image 2.” Photorealistic, ultra-detailed, cinematic outdoor advertising style with dramatic atmosphere and high-end commercial composition.

Example generated from “Windows 11 Focus UI Showcase”
GPT Image 2en

{ "type": "Windows 11 promotional UI mockup", "style": "The product marketing presentation is clean and modern, featuring soft frosted glass cards, delicate shadows, rounded corners, a light blue and white color scheme, a clear vector-style interface, and exquisite presentation slides aesthetics.", "scene": { "background": "A light ice-blue gradient background sets off a large, abstract Windows 11 blue ribbon swirling wallpaper at the center of the interface.", "headline": { "text": "{argument name=\"headline text\" default=\"Windows 11\"}", "subtext": "{argument name=\"tagline\" default=\"Designed for focused attention.\"}", "position": "Centered at the top" } }, "layout": { "type": "Desktop UI Display", "sections": [ { "title": "Left sidebar", "position": "Leftmost vertical parking bar", "count": 8, "labels": [ "Windows logo", "Homepage", "application", "document", "cloud", "set up", "Ava's Profile", "power supply" ] }, { "title": "Efficiency Widgets", "position": "Left column cards", "count": 3, "labels": [ "Good morning, Ava", "weather", "calendar" ] }, { "title": "Center Start Menu", "position": "Centered at the bottom", "count": 3, "labels": [ "Search bar", "Fixed", "recommend" ] }, { "title": "Right column cards", "position": "upper right, middle, and lower right", "count": 3, "labels": [ "Ava's phone", "notify", "Quick Setup" ] }, { "title": "task bar", "position": "Bottom full width", "count": 10, "labels": [ "Weather Summary", "start", "search", "Task View", "Microsoft Edge", "File Explorer", "Microsoft Store", "Outlook", "OneDrive", "Teams" ] } ], "centerpiece": "A huge Windows 11 Start Menu floats in front of a blue abstract wallpaper." }, "widgets": { "greeting_card": { "title": "Good morning, Ava", "subtitle": "Let's get today's work done efficiently.", "focus_block": { "label": "Focus", "time": "By 12:00 noon", "timer": "45 minutes", "button": "Focus timer" } }, "weather_card": { "location": "{argument name=\"weather location\" default=\"Seattle, Washington\"}", "temperature": "18°", "condition": "Cloudy turning sunny", "range": "Highest: 21°C Lowest: 13°C", "forecast_count": 5, "forecast_labels": [ "18°C now", "11:19 AM", "12:20 PM", "1:21 PM", "2:21 PM" ] }, "calendar_card": { "date": "Monday, May 20th", "events_count": 3, "event_labels": [ "Design synchronization at 11:00 AM, Teams meeting.", "Lunch with Mia at 1:30 PM, downtown.", "Project update at 3:00 PM, Meeting Room 32/1B" ], "button": "Open the calendar" }, "phone_card": { "device_name": "{argument name=\"device name\" default=\"Ava's phone\"}", "status": "Connected", "actions_count": 3, "action_labels": [ "information", "call", "photo" ], "battery": "100%" }, "notifications_card": { "title": "notify", "action": "Clear all", "items_count": 3, "item_labels": [ "Teams — 11:00 AM Design Synchronization — 2 minutes ago", "Outlook — New email from Mia — 10 minutes ago", "Calendar — 1:30 PM Lunch with Mia — 30 minutes ago" ] }, "quick_settings_card": { "title": "Quick Setup", "toggles_count": 6, "toggle_labels": [ "Wi‑Fi", "Bluetooth", "Flight mode", "Dark mode", "Focus Assistant", "Accessibility" ], "active_toggles": [ "Wi‑Fi", "Bluetooth" ], "sliders_count": 2, "slider_labels": [ "brightness", "volume" ], "wifi_network": "Contoso_5G" }, "start_menu": { "search_placeholder": "Search apps, files, settings...", "pinned_count": 18, "pinned_labels": [ "Edge", "Word", "Excel", "PowerPoint", "Outlook", "Teams", "photo", "shop", "Notepad", "Spotify", "Figma", "clock", "calculator", "set up", "document", "Movies and television", "Screenshot tool", "More" ], "recommended_count": 4, "recommended_labels": [ "Project Overview", "Second Quarter Financial Report", "Brand Guide", "My Notes" ] } }, "composition": "The product display is symmetrical, with the start menu centered, three cards stacked on the left and three cards on the right, with ample spacing. All elements appear to float slightly above the background, and the bottom taskbar spans the edge.", "rendering": "High-resolution UI concept art for desktop operating system launch pages, featuring perfect typography, Microsoft-inspired icon design, and a clean, minimalist branding." }

Example generated from “Wine Tasting Event Poster”
GPT Image 2en

Vertical 9:16 elegant wine tasting event poster. Background is a deep wine-burgundy color. Centered hand-drawn botanical illustration in gold linework of grape vines, leaves and a small cluster of grapes. Above the illustration an elegant italic serif title reads "Reserve Tasting". Below the illustration a smaller line in light cream type reads "Six wines, four courses, one evening". At the very bottom a small line of date and location "Friday · The Garden Hall". Vertical poster composition with subtle paper grain. No real winery name, no watermark. Premium boutique event poster aesthetic, magazine quality typography.

Example generated from “Winged Donkey Knight Sprite Sheet”
GPT Image 2en

Goal: Create a clean pixel-art sprite sheet for a fantasy game asset: {argument name="character concept" default="a medieval knight riding a winged donkey"}. Canvas: Square 768×768 image on a plain white background, no borders, no UI, no text. Use a crisp retro RPG pixel-art style with visible chunky pixels, dark outlines, limited palette, and subtle dithering/shading. Layout: Show exactly 8 separate sprites of the same mount-and-rider character, evenly spaced with large white gaps, arranged as 3 sprites on the top row, 2 sprites on the middle row, and 3 sprites on the bottom row. Each sprite should be roughly the same scale and read like an animation/rotation sheet. Sprite count and views: 1) top-left: rear three-quarter view facing upper-left, 2) top-center: straight rear view, 3) top-right: rear three-quarter view facing upper-right, 4) middle-left: side view traveling left, 5) middle-right: side view traveling right, 6) bottom-left: front three-quarter view facing lower-left, 7) bottom-center: straight front view, 8) bottom-right: front three-quarter view facing lower-right. Subject details: The mount is a small gray-brown donkey with long upright ears, pale muzzle, black hooves, simple bridle and saddle, and a pair of white feathered wings extended from its sides like a small pegasus. The rider is a compact armored knight wearing a rounded steel helmet with vertical visor slits, a bright red plume, a red cape or cloth drape, brown leather tack, and a blue shield with a gold cross emblem strapped at the side. Use {argument name="armor color" default="steel gray"}, {argument name="cape color" default="red"}, {argument name="shield color" default="blue and gold"}, and {argument name="mount color" default="gray-brown"}. Maintain the same proportions and costume details across all 8 sprites. Visual style: Classic 16-bit fantasy RPG sprite sheet, isometric/top-down compatible proportions, high readability at small size, black and dark-brown pixel outlines, soft shadowless presentation, minimal anti-aliasing, no blur, no gradients beyond pixel shading. Constraints: Keep exactly 8 sprites, do not add extra frames, labels, weapons, scenery, background texture, watermark, or character variations. All sprites must depict the same mounted knight and winged donkey from different directions.

Example generated from “Winter Concrete Salt Damage Close-Up”
GPT Image 2en

A hyper-realistic, documentary-style close-up photograph showcases severe salt damage on a concrete retaining wall weathered in winter. Shot from a low, oblique angle, the wall extends deep into the right side of the frame. A thin layer of snow and frost covers the top, while the ground beside the wall is dark and damp, partially covered in muddy snow. Extensive spalling and flaking are visible on the concrete surface, primarily concentrated in the upper half, with irregular areas exposing coarse aggregate of various sizes embedded in the mix. The damaged areas exhibit a rough, jagged, geographic outline with cracked edges, a pitted texture, and lost cement paste, revealing black and brown gravel. Vertical streaks of moisture and murky runoff extend downwards to the base of the wall. The overall color scheme is a mix of cool gray, charcoal gray, and soft winter white, illuminated by soft, overcast daylight, with a slight haze in the distance, and the surface texture is extremely fine. The composition should present the texture of a civil engineering inspection photo, like a real smartphone or on-site camera shot, with no people, text, labels, or exaggerated artistic processing in the picture.

Example generated from “Wireframe to Corporate Web Banner”
GPT Image 2en

Using REFERENCE_0 as the wireframe layout, generate a sophisticated and professional corporate webpage banner ad for an AI sales tool. Use a clean blue, white, and orange color palette. Transform the figures in the sketch into realistic images of two smiling Japanese business professionals (a man and a woman in suits) looking at a silver laptop in a bright office. Add [something] in the top left corner.{argument name="brand name" default="KAWAISALES"} The logo, along with the tagline "A new business support tool that utilizes AI." Map the text bubbles in the wireframe as follows: Replace the top bubble with one containing blue text.{argument name="main copy" default="Wouldn't you like to take your sales team to the next level with AI?"} The white oval shape; replace the middle bubble with {argument name="sub copy" default="From lead generation to closing, AI provides powerful support for your sales activities!"} Replace the bottom CTA box with a bright orange pill-shaped button with the text "..."{argument name="cta text" default="Experience a free demo now >"} In the bottom right center, add a blue 3D-style bar chart with an upward arrow, a target icon displaying three user silhouettes, and a white UI card with four checkmarks. Finally, add a white panel in the bottom left corner containing three blue line art icons (a clock, a rising chart, and a laptop), with a brief description below each icon.

Example generated from “Wireless Earbuds on Folded Towel”
GPT Image 2en

A pair of small white wireless earbuds resting on a neatly folded soft beige terry towel, with their charging case open behind them showing a single LED dot. Background is a clean off-white textured wall. Soft diffuse studio light from above and slightly front. Apple-style minimal product photography. Square 1:1 composition, no text, no logo on the earbuds, no watermark, no hands. Ultra sharp focus, photorealistic fabric and plastic textures.

Example generated from “Wizard's Room Illustration”
GPT Image 2en

{argument name="style" default="Handwritten magic book style"} Texture, depicting a{argument name="location" default="Wizard Room"}

Example generated from “Woman at Elegant Bar Cafe Table”
GPT Image 2en

A candid, realistic vertical photograph captures a woman seated at a rustic wooden table in an elegant upscale bar/café. Centered in the foreground, she leans forward, one elbow resting on the table, her hand supporting her chin, creating a relaxed, intimate, and editorially-inspired atmosphere. She has long, dark brown, wavy hair parted in the middle and cascading over her shoulders, and wears a fitted white ribbed long-sleeved round-neck top. Three items are clearly visible on the table in front of her: a pink cocktail in a wide-mouthed champagne glass on the left, a blue and white floral ceramic mug on the right, and a small clear glass vase containing a few dried flowers on the far right. Behind her is a warmly lit, elegant bar with marble or stone finishes, ornate classical ceiling details, shelves overflowing with bottles and glassware, soft backlighting, and a bartender in a white shirt and dark apron, slightly out of focus. The image includes dark bar stools beside the bar, and the overall color scheme is rich amber, cream, and brown. Employing shallow depth of field, the interplay of soft indoor natural light and warm bar lighting creates realistic skin and fabric textures with a delicate cinematic quality. The high-detail luxury hotel photography style, shot at eye level with a smartphone or portrait lens and in a vertical composition, creates a comfortable atmosphere that blends Parisian cafes and hotel bars.

Example generated from “Woman Dancing in Black Dress”
GPT Image 2en

A dramatic full-body photograph depicts a woman dancing powerfully in a dimly lit, upscale lounge or bar. She wears a fitted black evening gown with a high slit, the velvety fabric billowing with her movements. She has long, straight hair.{argument name="hair color" default="black"} Her hair flew to one side, her figure slender and athletic, adorned with delicate hoop earrings. Capture her confident and stylish dance moves: one arm gracefully raised overhead, the other bent downwards, her torso twisting to the side, emphasizing elegance and dynamism. The scene is dimly lit and cinematic, with warm amber-colored practical lighting, soft bokeh, and blurred spectators in the background, including several seated onlookers, and a glowing television screen on the left. The overall atmosphere is cool, sensual, and sophisticated, with strong contrast, shallow depth of field, realistic skin texture, subtle movements of the skirt, and magazine-like lighting. Realistic style, dynamic nightlife photography, vertical composition.

Example generated from “Woman Embracing a Bear in Winter”
GPT Image 2en

A cinematic, photorealistic shot of a young woman with {argument name="hair style" default="Long, wavy brown hair with small braids"} Wearing{argument name="clothing style" default="Rustic, historical clothing with fur trim and leather pouches"} She knelt in the snow, gently embracing a huge, docile animal.{argument name="animal type" default="Brown Grizzly Bear"} The woman and the bear both had their eyes closed, immersed in a peaceful and affectionate embrace. The bear rested its massive head against the woman. The background was winter.{argument name="environment" default="Pine forest in snow"} The lighting is soft and natural. Tiny snowflakes fall on the woman's hair and the bear's thick fur. High detail, 8K resolution, fantasy-realistic style.

Example generated from “Woman Reading at European Street Café”
GPT Image 2en

A documentary-style photograph with a cinematic quality, depicting a {argument name="subject" default="young women"} She sat alone at a small, weathered metal coffee table beside a narrow, cobblestone street in an old European town. Her hair was shoulder-length {argument name="hair color" default="Medium brown"} She had wavy hair, wore small gold hoop earrings, a soft beige knit sweater, and loose blue jeans. She sat with her legs crossed, one hand holding a printed ceramic coffee mug, the other gently touching the pages of an open paperback book on the table. A smartphone sat to the left of the table. The background featured a warm stone building with wooden shutters, green plants outside the shop, and a sign that read…{argument name="cafe sign" default="CAFÉ LUNA"} A hanging sign. Several pedestrians are in the distance, with a softly blurred background. Shot from eye level, shallow depth of field, natural golden morning light, soft bokeh, realistic and delicate skin and fabric textures, warm neutral tones, an intimate travel editorial style, and a vertical composition.

Example generated from “Woman Riding Motorcycle on Cliff Road”
GPT Image 2en

Using REFERENCE_1 as the character base and REFERENCE_0 as the scene/atmosphere reference, the same woman is regenerated into a pixel art-style action shot: she is riding a motorcycle on a narrow cliffside coastal road at sunset. The composition is changed from a close-up portrait to a wider, cinematic full-body perspective, with the motorcycle traveling towards the observer along a winding road carved into the steep coastal cliffs. Her brown leather jacket and overall styling are maintained, but her hair is allowed to flow in the wind, and she is posed with both hands gripping the handlebars. The dramatic drop in ocean elevation on the left, the rocky cliffs on the right, and the warm orange-purple sunset sky are emphasized. The retro 16-bit pixel art aesthetic is preserved.

Example generated from “Woodcut Print Graphic Design Poster”
GPT Image 2zh

请将我上传的每一张照片分别制作成一张独立的高级设计海报,不多图拼接,每张照片单独输出。整体采用3:4竖版构图,上下两个区域高度严格1:1,各占画面50%。 上半部分保留原始照片,保持主体结构、真实质感、自然光影和原有色彩氛围,仅进行轻微高级摄影调色,使其具有艺术出版物与展览摄影质感。为适配画幅,可自然扩展天空、地面或环境背景,但不得拉伸、扭曲或改变主体。 下半部分提取照片中最具识别性的**主体、轮廓、姿态与叙事关系**,先理解其核心寓意、文化属性或情绪关系,再重构为具有民俗图录、古物拓片与现代平面设计气质的图形画面。不要把照片中的元素逐一翻译,也不要添加泛泛的传统纹样;只保留**一个核心视觉母题与极少量真正有叙事意义的辅助形态**,通过剪影、负形、局部放大、形态合并与符号化转译,使图形既能对应原物,又能表达其背后的意义。 构图建立在严谨的**几何骨架与比例系统**之上,根据主体自身的结构提炼圆、方、三角、弧线、轴线或重复比例作为“几何母体”,再让所有主体、辅助图形、文字和留白与其发生对齐、切线、共轴、嵌套、镜像、比例递进或边界裁切关系。版面可以自由分布、偏心、越界或不对称,但必须共享同一套结构逻辑,形成清晰的视觉重心、层级、节奏与负空间;远看秩序明确,近看才发现传统印痕与细节变化,避免随机散点 and 装饰性堆砌。 配色从上方照片中提取最具辨识度、生命力与精神气质的颜色,重新归纳为清爽克制的纸面印刷色组。以明亮浅色纸底形成呼吸感,用一个深色建立视觉骨架,再以1–2个源自原图的主题色组织主体和少量强调,通过面积比例、叠印与明度关系建立层次,颜色应干净、有精神,避免灰脏做旧、固定民俗配色和杂乱多色。 材质采用**木刻拓印、干刷丝网印与旧式图录印刷**的混合质感,断墨、擦痕、露底和不均匀墨量主要存在于图形内部,几何边界、重要轮廓和整体版式仍保持清晰准确,形成“粗粝物质感 × 严谨几何秩序”的对比。 文字仅作为构图中的辅助信息,从照片的主题、寓意或文化属性中提炼少量标题与微型注释,可顺应几何轴线、负形或图形边界自由排布,与主体形成对齐、嵌套或穿插关系,但不抢夺视觉中心。 整体呈现**寓意先行、图形克制、几何严谨、拓印粗粝、留白清晰、传统文化感与现代设计秩序并存**的视觉气质。每一个出现的图形都必须有来源、有寓意、有结构作用;删除不能增强主题理解的装饰元素,避免素材拼贴、传统符号乱入、平均排版、写实插画和模板感

Example generated from “Wooden Bookshelf Prompt Test”
GPT Image 2en

A wooden bookshelf consisting of three shelves: On the top shelf, there should be one book, on the second shelf, there should be three books, and on the bottom shelf, there should be seven books.

Example generated from “World Champion Stadium Victory Poster”
GPT Image 2en

Create a cinematic vertical sports victory poster featuring {argument name="subject" default="a male soccer champion"} as a full-body figure standing triumphantly on top of a gigantic glowing soccer ball at the center of a futuristic packed stadium. The subject wears a black soccer kit with gold trim, black shorts, black knee socks, and gold cleats, posed confidently with one foot on the ball and one arm raised high holding the golden FIFA World Cup trophy. Use dramatic rain, wet reflective ground, blue-gray storm clouds, heavy stadium floodlights, mist, and a heroic low-angle perspective. The oversized soccer ball should dominate the lower foreground, with black and metallic gray panels separated by bright golden glowing seams and a warm reflection on the rain-soaked floor. In the background, show a massive cheering crowd and exactly 2 large blue digital stadium screens, one on the left and one on the right, both displaying the bold text {argument name="screen text" default="WORLD CHAMPION 2026"}. Also include repeated ribbon-board text around the stadium reading the same championship phrase. In the sky above the athlete, include exactly 4 major firework bursts: one smaller gold burst on the upper left, one blue burst near upper center-left, one large bright blue-white burst at upper center-right, and one gold burst on the upper right, with additional thin spark trails but no extra main bursts. Style the image as ultra-realistic, high-contrast, glossy, cinematic advertising photography with volumetric light beams, water droplets, lens flare, and epic World Cup celebration energy. Keep the composition centered and symmetrical, portrait aspect ratio 9:16, no watermark, no logos beyond subtle kit-style emblems, and make all text legible.

Example generated from “World Civilization Spatiotemporal Map”
GPT Image 2en

Accurately visualized in a popular science style{argument name="subject" default="A 5000-year time map of world civilization"} It also includes both time and space dimensions.

Example generated from “World Cup Celebration Portrait”
GPT Image 2en

Use the uploaded image as the face reference. Ultra-realistic portrait of the same man kissing the {argument name="trophy" default="golden FIFA World Cup trophy"} with eyes closed, holding it with both hands. Wearing a {argument name="jersey" default="Portugal national team jersey inspired by Cristiano Ronaldo #7"}. {argument name="background" default="Clean white studio background"}, soft professional lighting, golden reflections from the trophy, detailed skin texture, realistic beard and hair, Canon EOS R5, 85mm lens, shallow depth of field, sports editorial photography, emotional championship moment, ultra-detailed, photorealistic, 8K, masterpiece quality.

Example generated from “World Cup Champion Tunnel Poster”
GPT Image 2en

Create a cinematic vertical sports poster showing {argument name="character name" default="a confident male football champion"} walking slowly toward the camera through a dark, modern FIFA World Cup player tunnel. He is centered full-body, wearing a black streetwear-inspired football outfit: black long-sleeve jersey with gold trim, a gold crest, small tournament emblems, the word {argument name="shirt text" default="CHAMPION"} across the chest, black jeans, and black sneakers. His face should be intentionally soft/blurred as if identity-preserved from a reference, while the pose feels calm, victorious, and powerful. He holds one golden World Cup-style trophy in his left hand at chest height. The tunnel walls and ceiling are covered with deep navy LED screens displaying large white text reading {argument name="event text" default="WORLD CUP 2026"}, repeated on the ceiling and both side walls, plus small infinity-like tournament logos and subtle red, cyan, and magenta ribbon graphics. At the tunnel exit behind him, create a glowing stadium entrance with intense white floodlights, green pitch visible at the bottom, and a cheering crowd in soft focus. Add exactly 9 floating foreground/background sports elements: 4 golden World Cup-style trophies positioned upper left, mid-right, lower left, and lower right; and 5 black-and-white soccer balls positioned upper right large cropped, left-center small, lower-left mid, bottom-right large cropped, and one additional small ball near the left wall. Use dramatic volumetric lighting, glossy reflections on the floor, strong backlight halo, shallow depth of field, high-contrast blue-and-gold color grading, realistic photographic detail, and a triumphant commercial advertising style. Vertical 9:16 composition, no watermark, no extra logos beyond the tournament-style marks, keep text legible and bold.

Example generated from “World Cup Manga Football Poster”
GPT Image 2en

Goal: Create a black-and-white manga comic poster for a FIFA World Cup 2026 football star, featuring {argument name="character name" default="Hakan Çalhanoğlu"} as a heroic chibi-style midfielder sprinting forward with a ball. Canvas: Square 1:1 poster, high-contrast monochrome ink, dense manga halftone shading, speed lines, explosion bursts, thick black outlines, exaggerated comic typography, energetic sports-card composition. Main subject: Center the footballer in a dynamic running pose, oversized head and compact athletic body, wearing a dark Türkiye national team kit with crescent-and-star chest badge, jersey number {argument name="jersey number" default="10"} on the shirt and shorts, Nike-style swoosh, dark socks, cleats, and a football at his feet. Hair is spiky and dramatic. Place a plain vertical gray rectangular censor/placeholder block over the face, covering most facial features while leaving hair, ears, body, and kit visible. Add a small Türkiye shield crest above the head with crescent, star, football, and “1923”. Layout: Huge central headline at the top reading “{argument name="headline surname" default="ÇALHANOĞLU"}” in massive block letters with drop shadow. Above it, a jagged caption bubble reads “THE MAESTRO!” Put the FIFA World Cup 2026 logo at top left. Use a giant pale “26” shape behind the player, plus explosive rays radiating from the center. Comic callouts: Include exactly 8 comic text bursts around the player: 1) left upper burst: “VISION! TECHNIQUE! PRECISION! PERFECTION!!!”; 2) left diagonal sound effect: “BOOM!”; 3) right upper burst: “THE PITCH IS HIS CANVAS!”; 4) right middle burst: “THE TURKISH MAESTRO!”; 5) right lower-middle burst: “WHAT A LEGEND!!”; 6) lower right huge slanted burst: “UNSTOPPABLE!!”; 7) bottom right burst: “SET PIECES ARE ART!”; 8) top center burst: “THE MAESTRO!”. Info card: Add one rectangular player card at lower left with the text “HAKAN ÇALHANOĞLU”, “10 | MIDFIELDER”, “TÜRKİYE”, “THE GENERAL OF MIDFIELD!”, and three stars along the bottom. Bottom timeline: Along the bottom edge, include exactly 6 small World Cup icon badges with labels: 1) 2002 FIFA WORLD CUP (KOREA/JAPAN), 2) 2006 FIFA WORLD CUP (GERMANY), 3) 2010 FIFA WORLD CUP (SOUTH AFRICA), 4) 2014 FIFA WORLD CUP (BRAZIL), 5) 2018 FIFA WORLD CUP (RUSSIA), 6) 2022 FIFA WORLD CUP (QATAR 2022). Style constraints: Pure black, white, and grayscale only; no color. Make the composition extremely busy and celebratory, like a Japanese sports manga cover mixed with a vintage football trading card. Use crisp linework, halftone dots, ink splatters, debris, action rays, and bold readable lettering. Add a small artist signature near the right side of the large background number.

Example generated from “World Cup Stadium Fan Transformation”
GPT Image 2en

Transform the uploaded photo into a realistic World Cup stadium fan scene. Place the person naturally among thousands of supporters in a packed football stadium during a major international tournament. The person should be wearing the official national team jersey of {argument name="country" default="your country"}, sitting or standing among fans who are also supporting the same nation. Maintain the person’s facial identity, hairstyle, and recognizable features while creating a realistic candid sports photography look. Capture the excitement, emotion, and atmosphere of a live World Cup match, with authentic stadium seating, crowds, flags, scarves, team colors, and match-day energy. The subject should blend naturally into the crowd as if photographed by a professional sports photographer during a real game. Use realistic lighting, shallow depth of field, natural facial expressions, and documentary-style sports photography. The image should feel authentic, cinematic, highly detailed, and indistinguishable from a real photograph taken during an international football tournament. Ultra-realistic, professional photography, realistic crowd interactions, authentic fan culture, natural clothing folds, realistic stadium environment, high detail, 8k quality.Ratio 4:5

Example generated from “World Cup Throne Portrait”
GPT Image 2en

Create a cinematic vertical luxury football champion portrait in a dark studio arena. A {argument name="subject identity" default="young adult man with short dark hair"} sits centered and facing forward on an enormous black iron throne inspired by a fantasy sword throne, made from dozens of overlapping steel blades and swords fanning upward behind him like a crown. He wears a {argument name="outfit" default="plain black long-sleeved shirt, black jeans, and black-and-red high-top sneakers"}. His posture is powerful and relaxed: seated with legs apart, right hand gripping a gleaming {argument name="main trophy" default="FIFA World Cup trophy"} held upright near his knee, left forearm resting on the throne armrest. Place the throne on a circular glossy black metallic platform. Surround him with exactly 6 visible football awards and props on the platform: 2 large gold cup trophies standing at the far left and far right edges, 2 taller narrow gold trophies beside his feet, and 2 metallic gold footballs in the front left and front right corners. Add additional smaller gold football ornaments integrated into both throne armrests. Background: a vast dark stadium or arena with blurred crowd seating, cool blue stadium floodlights on both sides, deep black upper background, and subtle haze. Lighting should be dramatic and high contrast, with teal-blue rim light from the stadium and warm red-orange highlights on the throne, trophies, shoes, and platform edges. Style: ultra-realistic, premium sports editorial, heroic symmetrical composition, sharp metallic textures, glossy reflections, shallow depth of field, 9:16 vertical poster framing. Keep the face recognizable if a reference is provided, but do not add text, logos, watermarks, or extra people.

Example generated from “World Cup Trophy Reveal Perspective”
GPT Image 2en

Ultra-realistic sports photography, documentary realism, premium commercial campaign style, immersive 9:16 vertical composition.\n\nThe camera is positioned deep inside an opened luxury gift box, creating a dramatic first-person perspective from the hidden treasure itself. A magnificent golden FIFA World Cup trophy rests on elegant dark velvet fabric in the lower foreground, with realistic metallic reflections and authentic textures.\n\nAbove the opened box stands [PLAYER NAME], leaning forward and looking directly into the box with an expression of determination, ambition, and hunger for glory. He reaches one hand naturally toward the trophy, symbolizing his dream of winning the FIFA World Cup 2026.\n\nSTRICT IDENTITY PRESERVATION OF [PLAYER NAME]. The player must be instantly recognizable as the real [PLAYER NAME], with authentic facial structure, eye shape, nose, lips, jawline, hairstyle, beard style, skin texture, and expression. Preserve his real-world appearance with extreme accuracy. No generic AI face.\n\nAuthentic human anatomy, realistic proportions, detailed skin pores, realistic eyes, natural hair strands, genuine facial expression, lifelike appearance, true-to-life colors.\n\nOfficial national team jersey with realistic fabric folds, stitching, texture, sweat details, and accurate colors.\n\nThe official FIFA World Cup 2026 logo is elegantly integrated into the composition with realistic physical appearance. The overall scene incorporates the authentic FIFA World Cup 2026 visual identity and official tournament color palette, creating a premium championship atmosphere.\n\nWarm stadium lights in the distance, subtle crowd bokeh, realistic depth of field, natural shadows, cinematic composition, premium sports advertising photography.\n\nCaptured with a Canon EOS R3 and RF 70-200mm f/2.8L lens, Getty Images realism, Reuters sports photography quality, authentic football photography, shallow depth of field, realistic lighting, true-to-life materials.\n\nLuxury black gift box interior, premium velvet textures, elegant packaging, realistic cardboard material, soft shadows, natural golden reflections.\n\nPhotorealistic, highly detailed, realistic skin texture, realistic hands, realistic fingers, authentic athlete appearance, physically accurate materials.\n\nNo generic AI faces, no face distortion, no altered facial features, no stylized appearance, no cartoon style, no CGI look, no plastic skin, no wax skin, no fantasy effects, no exaggerated proportions, no deformed hands, no extra fingers, no toy figures, no LEGO style, no miniatures.\n\nNo watermarks, no unrelated text, no third-party branding.\n\n16K ultra-high-resolution, photorealistic, premium commercial sports photography, FIFA World Cup 2026 atmosphere.\n\nAspect Ratio: 9:16

Example generated from “World Cup Tunnel Champion”
GPT Image 2en

Create a vertical 9:16 cinematic sports poster of {argument name="character name" default="a confident male football champion"} walking slowly toward the viewer through a dark futuristic FIFA World Cup player tunnel into a packed stadium. He is shown full body, centered, wearing an all-black modern athletic tracksuit with a zip jacket, black pants, and black shoes, holding the golden FIFA World Cup trophy in one hand at chest height. The scene is rainy and dramatic, with wet reflective tunnel flooring, strong backlighting from the stadium entrance, bright white floodlights, mist, sparks, and explosive golden pyrotechnics behind him. The tunnel walls are dark blue-black metal panels with glowing digital World Cup graphics and readable-style branding, including FIFA, Qatar 2022, and World Cup trophy emblem panels, without making the text the main focus. Add exactly 4 floating soccer balls around the subject: one large ball on the lower left foreground, one smaller ball on the upper left side, one large ball on the mid-right wall area, and one ball near the lower right foreground, all with motion streaks. Add exactly 6 visible fan or photographer figures leaning in from the tunnel edges: two on the left side and four on the right side, some cheering with arms raised and one holding a bright camera flash. The stadium beyond the tunnel is crowded and glowing, with a curved roof structure visible at the top. Use high-contrast teal-and-gold color grading, ultra-realistic detail, dramatic depth of field, rain streaks, lens flares, glossy reflections, and a heroic World Cup entrance mood. Keep the subject’s face as {argument name="face treatment" default="front-facing but anonymized/blurred"}; if a real identity is provided, preserve facial identity exactly. Avoid cartoon style, avoid extra limbs, avoid distorted trophy shape, and avoid cluttered or unreadable typography.

Example generated from “World Cup Victory Tunnel Scene”
GPT Image 2en

victorious {argument name="event" default="FIFA World Cup"} tunnel scene featuring the uploaded face as reference for identity preservation, the subject walking slowly and confidently through a dark modern player tunnel toward a glowing stadium entrance, wearing a {argument name="outfit" default="black streetwear-inspired football outfit"} with a long-sleeved top, dark jeans, and light sneakers, holding a golden trophy tightly in the left hand as if just won the final, while several metallic trophies and footballs appear suspended around him in a surreal cinematic arrangement, some floating slightly above the ground and others reflecting in the glossy tunnel floor, the tunnel walls lined with dim LED panels, abstract World Cup graphics, and blurred flashes of cheering crowd light from the stadium ahead, intense cool blue overhead lighting mixed with warm red uplight creating a powerful contrast across the face and outfit, smoke haze drifting through the corridor, dramatic shadows, glossy reflections, shallow depth of field, high-end sports poster composition, ultra realistic details, monumental atmosphere, premium cinematic grading, designed to feel legendary, viral, and unforgettable.

Example generated from “World of Warcraft Style Raid”
GPT Image 2en

A{argument name="player count" default="20"} The human-run raid is currently in combat, and the opponent is {argument name="character" default="Sam Altman"} ,use{argument name="year" default="2004"} The style of World of Warcraft from 2000. Brief description.

Example generated from “World of Warcraft Style Raid”
GPT Image 2en

a {argument name="player count" default="20"} person horde raid is fighting {argument name="character" default="Sam Altman"} in {argument name="year" default="2004"} world of Warcraft style. One shortted.

Example generated from “WSJ-Style Business Newspaper Front Page”
GPT Image 2en

A photorealistic newspaper, taken from a three-quarter view from above, sits folded on a dark wood desk, its style mimicking a famous American business daily from the 1980s to the 2000s. The newspaper features a clear, lifelike texture of newsprint, subtle paper creases, and slightly worn folded edges. Warm morning light casts diagonal window shadows across the pages. The masthead occupies the top, written in bold, large, classic serif capital letters.{argument name="newspaper name" default="THE WALL STREET DAILY."} Below it is a thin header containing details of the small financial newspaper: "DOW JONES", "News Corp", the date line "MONDAY, APRIL 20, 2026 – VOL. CCLXXXV NO. 90", "WSJ.com", a star rating line, and the price "$5.00". The main news is located in the upper left center, with a large, bold serif headline.{argument name="main headline" default="Cook to Step Down as Apple CEO; Ternus Named Successor"} Below is a smaller serif subheading: “Transition caps 15-year run; company pivots to AI-driven future.” Add a line of small capital letters byline including the two journalists: “By Tim Higgins” and “And Joanna Stern.” The main body is filled with dense, fully readable multi-column news articles about Tim Cook’s departure as Apple CEO, John Tenus’s succession, Apple’s growth, AI strategy, and corporate transformation; the text should read like solid financial news, not placeholder gibberish. To the right of the main article is a black-and-white, woodcut-style portrait of a male executive, wearing a shirt and jacket, centered in the rectangular image area; his face is partially obscured by a light gray square above his eyes and nose, creating a modern ironic effect. In the upper right corner, there is a sidebar titled "What's New," containing two subheadings: "Business & Finance" and "World-Wide," each followed by a brief summary in small serif font. "Business & Finance" contains three points, and "World-Wide" contains four. All content must be fully formatted and clearly readable. Below the main news, a second-level banner article spanning the entire page is added, titled "Silicon Valley Braces for New Chapter in AI Arms Race," attributed to "By Liz Hoffman," and accompanied by several short articles about large tech companies, startups, OpenAI, Google, Meta, chips, platforms, and the AI ​​competition. To the right of this lower-middle area, a compact market table is inserted, titled "Markets," with four rows of labels: "DOW JONES," "S&P 500," "NASDAQ," and "10-YR. TREASURY," and columns for index/value/change/percentage, using a realistic newspaper table format and actual figures. At the bottom of the front page, there are three preview boxes: the left box is labeled "Heard on the Street," with the headline "Why Investors Are Rethinking Big Tech," and is marked "B12"; the middle box is labeled "Markets," with the headline "Earnings Season Lifts Stocks Despite Rate Uncertainty," and is marked "B1"; the right box is labeled "Inside," with the headline "Private Credit Sees Inflow Surge as Yields Stay High," and is marked "B14." All typography is kept clean, sharp, and fully readable, with consistent alignment in an authentic newspaper grid layout, thin lines separating columns, justified text, a classic serif font, and realistic print contrast. The newspaper is slightly rotated clockwise on a tabletop. In the environment, a white ceramic mug containing black coffee is placed in the upper right corner, and a pair of dark-rimmed glasses sits near the lower right edge of the newspaper. The upper left corner contains the cropped edge of a dark cylindrical object, resembling a camera lens or a flat-bottomed mug. Overall feel: High-end editorial mockup front page, serious business journalism aesthetics, believable print photography feel, ultra-high detail, natural shadows, high resolution.

Example generated from “WSJ-Style Business Newspaper Front Page”
GPT Image 2en

A photorealistic overhead three-quarter view of a folded financial newspaper on a dark wooden desk, styled like a prestigious American business daily from the 1980s–2000s, with crisp realistic newsprint texture, subtle paper wrinkles, slightly worn folded edges, and warm morning sunlight casting diagonal window-shadow bands across the page. The masthead dominates the top in large classic serif capitals reading {argument name="newspaper name" default="THE WALL STREET DAILY."}. Directly beneath it is a thin header strip with small financial-paper details: "DOW JONES", "News Corp", the date line "MONDAY, APRIL 20, 2026 – VOL. CCLXXXV NO. 90", "WSJ.com", a star rating row, and the price "$5.00". The main lead story spans the upper center-left with a huge bold serif headline reading {argument name="main headline" default="Cook to Step Down as Apple CEO; Ternus Named Successor"}. Below it is a smaller serif subheadline: "Transition caps 15-year run; company pivots to AI-driven future". Add a byline line in small caps with 2 reporters: "By Tim Higgins" and "And Joanna Stern". Fill the lead article with dense, fully legible multi-column newspaper body text about Tim Cook stepping down as Apple chief executive, John Ternus succeeding him, Apple's growth, AI strategy, and corporate transition; the text should read like authentic financial journalism, not placeholder gibberish. On the right half of the lead story, place a black-and-white engraved stipple portrait of a male executive in collared shirt and jacket, centered in a rectangular image area; the face is partially obscured by a pale gray square censor block over the eyes and upper nose, creating a modern satirical effect. At the top-right, include a sidebar box titled "What’s New" with exactly 2 subsection headers, "Business & Finance" and "World-Wide", each followed by bullet-point briefs in small serif text; include exactly 3 bullets under "Business & Finance" and exactly 4 bullets under "World-Wide", all fully typeset and legible. Beneath the main story, add a secondary banner article across the page titled "Silicon Valley Braces for New Chapter in AI Arms Race" with a byline "By Liz Hoffman" and several short columns of text about big tech, startups, OpenAI, Google, Meta, chips, platforms, and AI competition. To the right of this lower-middle section, insert a compact markets table titled "Markets" with exactly 4 rows labeled "DOW JONES", "S&P 500", "NASDAQ", and "10-YR. TREASURY", with columns for index/value/change/percent and realistic numeric values in tight newspaper tabular formatting. Along the bottom strip of the front page, include exactly 3 teaser boxes: left box with section label "Heard on the Street" and headline "Why Investors Are Rethinking Big Tech" plus page marker "B12"; center box with section label "Markets" and headline "Earnings Season Lifts Stocks Despite Rate Uncertainty" plus page marker "B1"; right box with section label "Inside" and headline "Private Credit Sees Inflow Surge as Yields Stay High" plus page marker "B14". Make all typography clean, sharp, perfectly legible, and consistently aligned in authentic newspaper grid layout with thin rules between columns, justified text, classic serif fonts, and realistic print contrast. The newspaper sits slightly rotated clockwise on the desk. In the environment, show a white ceramic coffee cup filled with black coffee in the top-right corner and a pair of dark-framed eyeglasses resting near the lower-right edge of the paper. In the top-left corner, include the cropped edge of a dark cylindrical object, likely a camera lens or tumbler. Overall mood: premium editorial mock front page, serious business-news aesthetic, believable print photography, ultra-detailed, natural shadows, high resolution.

Example generated from “Wuxia MMORPG Ancient Town Screenshot”
GPT Image 2en

A meticulously detailed, cinematic-quality screenshot of a third-person martial arts MMORPG is shown. The scene is set at dusk in a bustling ancient Chinese water town market street, with the perspective positioned behind the player character in the foreground center. The player character is a wandering swordsman, dressed in a dark gray and blue Hanfu-style robe with embroidered patterns, his long black hair tied in a low ponytail, carrying a sword sheath on his back, facing the crowded street ahead. The environment is a beautifully rendered Ming or Song Dynasty-style town, featuring traditional wooden shops, tiled roofs, hanging red lanterns, pink cherry blossom trees, damp, reflective cobblestone streets, a soft golden sunset, and townspeople dressed in ancient costumes. In the right foreground, three elegant female geishas in semi-transparent light-colored Hanfu stand outside an ornately decorated teahouse or brothel, draped with red silk curtains; one is playing the pipa, one sits in a light green and beige robe, and the other stands in a flowing rose-pink robe. The largest shop sign at the top features gold Chinese characters on a black plaque, and vertical shop banners with Chinese characters hang on both sides of the street. The overall composition should present the quality of a top-tier AAA Chinese fantasy RPG, with realistic character proportions, dense scene narrative, dramatic yet natural lighting, atmospheric depth, and complex clothing textures. It includes a complete Chinese game HUD overlay: a circular character portrait in the upper left corner displaying level 36 and two horizontal health and mana progress bars; a quest tracking panel on the left with four quest entries under each tag title; a semi-transparent chat box in the lower left corner displaying multiple lines of Chinese world chat content; a circular minimap in the upper right corner with location text; a vertical menu on the right with six icon buttons; a skill bar at the bottom center with ten slots, including eight square skill icons and two circular end slots with keyboard key labels; and a bottom menu row with six navigation icons. The game features ultra-detailed, photorealistic simulations of a fantasy game interface, exquisite UI, an immersive and crowded street scene, and a 16:9 aspect ratio.

Example generated from “Wuxia MMORPG Gameplay Screenshot”
GPT Image 2en

Screenshots from the "Snowy Path of the Fierce Blade" themed ancient ARPG MMO open-world game: Li Chungang comprehends swordsmanship before the gates of Longhu Mountain. With a cry of "Swords, come!", ten thousand swords fly forth, converging from all directions like a tidal wave. Xu Fengnian stands by, facing off against Zhao Danping. The scene is grand and awe-inspiring, showcasing the full power of a terrestrial immortal. On-screen text: "If Heaven had not given birth to Li Chungang, the way of the sword would have remained in eternal darkness," "Swords, come!" High-definition, epic feel, cinematic lighting and shadows, complete MMORPG game UI. Screenshots from the "Snowy Path of the Fierce Blade" themed ancient ARPG MMO open-world game: The final battle atop Wudi City. Old Huang transforms into Sword Nine Yellow, his clothes stained with blood as he unleashes "Sword Nine: Six Thousand Miles," golden sword energy crisscrossing the heavens and earth. In the distance, Wang Xianzhi faces his opponent with both hands, a tragic and fierce display of power, a peak sword move burning with life. On-screen text: "Sword Nine: Six Thousand Miles," "Anyone who dares to challenge Beiliang, ask the sword in my hand." High-definition, epic BOSS battles, complete MMORPG game UI. Screenshots from the "Snowy Path of the Sword" themed ancient ARPG MMO open-world game: Li Chungang stands atop a shattered mountain peak, cleaving the heavens with a single sword strike. The sword light soars to the heavens, the gates crack, chains shatter, the earth trembles, and Xu Fengnian watches from afar, the scene awe-inspiring and sacred, brimming with a defying-the-heaven spirit. On-screen text: "Severing destiny, severing shackles," "One sword opens the heavens." High-definition, epic feel, ultimate Xianxia (fantasy martial arts) grand scenes, complete MMORPG game UI. Screenshots from the "Snowy Path of the Sword" themed ancient ARPG MMO open-world game: A battle at the outer altar. Li Chungang unleashes a domineering sword strike, a golden sword pillar descends from the sky, the immortal opposite him kneels instantly, the earth collapses, the ruins of the heavenly palace float, the sword intent suppresses all living beings, an invincible aura. On-screen text: "One sword, immortals kneel," "With Li Chungang here, there are no immortals under heaven!" High-definition, epic Xianxia BOSS battles, complete MMORPG game UI.