PROMPTS · EXPLORATION CRÉATIVE

Galerie de prompts

Explorez des prompts réels et leurs exemples générés. Trouvez l’inspiration pour votre prochaine mission créative et copiez le texte intégral gratuitement.

Exemples
11,025
Modèles
18
Sources
1277
11 025 exemples de prompts · page 192 sur 230
Example generated from “Wuxia Movie Character Posters”
GPT Image 2en

Movie character posters: Dongfang Bubai, dressed in a red robe, drinks wine against a cliff at sunset, creating a martial arts atmosphere; Dongfang Bubai, with her embroidery needles flying, stands on a cliff with long red hair, the sunset over Blackwood Cliff like blood.

Example generated from “Wuxia Movie Character Posters”
GPT Image 2zh

电影角色海报,东方不败红衣饮酒,悬崖落日,武侠意境;东方不败绣花针如飞,红衣长发立于悬崖,黑木崖夕阳如血

Example generated from “WWII D-Day Cinematic Poster”
GPT Image 2en

A high-quality World War II movie poster, using a vertical single-page layout, with a design style reminiscent of modern high-budget war film advertisements. Two lines of compressed capital letters are displayed at the top center of the poster:{argument name="top tagline" default="Courage. Brotherhood. Sacrifice."} and{argument name="second tagline" default="The price of victory."} The main composition features four American World War II soldiers in dark olive-colored uniforms and large-mesh M1 helmets: a dominant central soldier in a half-length portrait facing forward, flanked by three slightly smaller soldiers on each side (one on the left and two on the right). Their faces are shrouded in shadow, covered in dirt and grime, showing the effects of battle, with strong light and shadow effects and smoky backlighting. The background is a gloomy sky filled with smoke, sparks, embers, and two groups of aircraft: three in the upper left and five in the upper right. A line of small text is centered in the middle of the image:{argument name="based on text" default="Based on true events"} Below is a huge, bold, blocky, distressed headline:{argument name="movie title" default="D-DAY"} The two D's are separated by an asterisk. Below the title is a line of red compressed dates:{argument name="date text" default="June 6, 1944"} The lower half shows the Normandy landings from the inside of an open landing craft, framed by two dark metal ramps. Eight soldiers, carrying rifles and backpacks, wade towards the shore amidst the spray, surrounded by Czech hedgehog-like barricades, surging water, smoke, distant fires, and a bombed-out coastline. At the very bottom is a dense, tiny credits list, below which is a striking red line announcing the release date.{argument name="release text" default="Coming Soon"} Following this is a smaller line of white text: "In tribute to their sacrifice." The color scheme is charcoal gray, steel blue, and mud brown, accented with red lettering. This ultra-refined, realistic, high-contrast, emotional, epic, historical, and dramatic cinema poster design features authentic film grain, volumetric smoke, aged textures, and symmetrical composition, making it a top-tier visual art piece for blockbuster films.

Example generated from “Xianxia Cityscape from God's-Eye View”
GPT Image 2en

An ultimate God's-eye view, looking directly down at the vast and radiant expanse.{argument name="city style" default="Ancient Chinese Cities"} The city stretches to the horizon, its curvature even visible on the planet's surface. In the foreground, a solitary figure {argument name="character type" default="swordsman"} Clad in a traditional robe, I stand on the edge of the ornately decorated, dark-tiled roof of an ancient palace, gazing down. Below lies an endless complex of ancient buildings, illuminated by warm lights, bustling streets teeming with people. A soft mist swirls through the vast city. Photorealistic fantasy concept art, Chinese xianxia aesthetics, ultra-wide-angle lens, immense depth of field. A grand, solitary, and poetic atmosphere, with strong color contrasts and profound depth.{argument name="color palette" default="Deep blue and warm gold"} tone.

Example generated from “Xianxia Cityscape from God's-Eye View”
GPT Image 2en

An extreme top-down god's-eye perspective looking directly down at a vast, glowing {argument name="city style" default="ancient Chinese city"} extending to the horizon, where the curvature of the planet is visible. In the foreground, a lone {argument name="character type" default="swordsman"} in traditional robes stands on the edge of an ornate, dark-tiled ancient palace roof, looking down. Below lies an endless sea of ancient architecture illuminated by warm glowing lights, bustling streets, and dense crowds. Soft mist and fog layers drift through the sprawling metropolis. Photorealistic fantasy concept art, Chinese Xianxia aesthetics, extreme wide angle, massive depth of field. Grand, lonely, and poetic atmosphere with strong color contrast and a moody {argument name="color palette" default="dark blue and warm gold"} color palette.

Example generated from “Xianxia Sect Panorama Map”
GPT Image 2en

A grand, ultra-wide, cinematic-quality map of a fantasy world unfolds, showcasing an incredibly detailed panorama of Chinese cultivation sects, viewed from a high vantage point above a sea of ​​clouds. The scene depicts a magnificent mountain range comprised of numerous steep, rocky islands, featuring waterfalls, mist-shrouded valleys, floating boulders, ornate bridges, and layered traditional Chinese palace architecture, all bathed in a mysterious, glowing blue energy. At the center stands the grand main sect complex, labeled "Xuantian Sword Sect," perched atop the highest, glowing peak, adorned with a crystalline glow of interwoven white and blue light, towering pagodas, terraces, and ethereal aura; a smaller label nearby reads "Sect Main Hall." Nine additional marked locations are distributed around the panorama: the far left is a purple cliff area labeled "Sword-Asking Cliff," with a smaller note below stating "Place for Disciples' Training"; the middle left is a cultivation platform island shrouded in blue mist, labeled "Enlightenment Peak," with a smaller note below stating "Place for Disciples' Enlightenment"; the upper left is a solemn complex of palaces labeled "Scripture Repository," with a smaller note below stating "Collection of Supreme Techniques"; the upper right is a jagged iceberg sword formation labeled "Sword Tomb," with a smaller note below stating "The final resting place of successive generations of sword cultivators"; the middle right is a bright ceremonial building complex labeled "Martial Arts Arena," with a smaller note below stating "Sharpening the Disciples' Character"; the lower right is the Alchemy Pavilion area labeled "Alchemy Room," with a smaller note below stating "Alchemy Therapy"; the far right is a pink flower sea valley sanctuary labeled "Hundred Flowers Valley," with a smaller note below stating "A Mystical and Beautiful Secret Realm"; the lower right is the outer building complex labeled "Outer Gate," with a smaller note below stating "New Disciples' Residence." Add an inscription in the upper left corner, reading "Xuantian Sword Sect" in large gold characters, below which is a line of smaller poetry and a red seal. Employing an elegant painterly concept art style, the design features grand proportions, highlights in pale gold and ice blue, soft volumetric mist, a dramatic sky after rain, ethereal lighting, and exquisite blend of Chinese ink painting and digital landscape aesthetics. The foreground contains no figures or modern elements, presenting only a vast, legendary sect environment; the design resembles a promotional illustration for its world-building.

Example generated from “Xiaohongshu cooking tutorial card”
GPT Image 2zh

Create a Xiaohongshu-style viral cooking tutorial image in a 3:4 vertical layout for homemade scallion oil noodles. Cozy home-cooking vibe, warm inviting lifestyle aesthetic, 4 to 6 step grid layout, clean spacing, realistic food photography, soft natural lighting, slight film tone, warm color grading, visible oil sheen, steam, sauce texture, and hands interacting with the food. Add small Chinese annotations such as 切葱, 熬油, 拌面, 出锅. Avoid overcrowding or excessive text.

Example generated from “Xiaomi SU7 Exploded View Chat Screenshot”
GPT Image 2en

A screenshot of a stylish Chinese social media interface in a deep charcoal black, featuring a minimalist yet sophisticated design. A large, rounded rectangular image card is placed in the lower left center, while a small, rounded chat bubble appears in the upper right corner. The overall canvas has ample white space, creating a clean and technologically advanced feel. In the upper left of the main area, bold white Chinese text reads: "If we were to use Xiaomi cars, I would create a more sophisticated, exploded view of the entire vehicle, more like a product launch presentation." In the upper right corner, a gray, rounded bubble contains the white Chinese text: "Xiaomi Car Product Teardown Diagram." Below is a poster-style, rounded infographic card with a glossy black background. The inside of the card is turquoise.{argument name="car model" default="Xiaomi SU7"} A stunning, exploded-down product presentation of the electric sedan is created, employing a highly stimulating 45-degree front-view perspective. The body is disassembled into multiple components, floating around the chassis. The hood is open, the trunk lid is raised, the four side doors are removed and float outwards, the front bumper and fascia are separated, the headlights are displayed independently, the interior is visible, and the main mechanical and structural modules are neatly layered and arranged below and around the body. The visual style should resemble slides from a high-end launch event: cinematic studio lighting, realistic reflections, sharp details, a dark background, subtle blue highlights, and precise technical composition. In the upper left corner of the infographic card, three lines of large white Chinese titles are placed: "Xiaomi Auto SU7," "Product Teardown Diagram," and "Technological Leap, Human-Car Unity." An orange square Xiaomi-style logo is added to the upper right corner of the card. Fourteen Chinese labels surround the car, connected by thin white lines, indicating visible components and systems. The left side features six functional labels, including the integrated closed front fascia, LED headlights, front trunk design, high-voltage battery, front motor assembly, front double wishbone suspension, and Brembo four-piston brake calipers. The right side features eight functional labels, including the power tailgate, continuous taillights, frameless doors, hidden door handles, rear motor assembly, rear motor parameters, all-aluminum chassis, five-link rear suspension, and Michelin high-performance tires. Beneath the exploded, disassembled body, the battery pack, front and rear drive units, suspension subframe, wheel assemblies, braking components, front and rear seats, center console parts, chassis skid plates, and structural floor elements are clearly displayed, all meticulously organized in an explosive, floating arrangement. At the bottom of the infographic card, a horizontal information bar is added, divided into five columns of small Chinese text, titled Vehicle Dimensions, Power Performance, Driving Range, Intelligent Driving, and Safety Structure. A bold white Chinese label "Edit" is placed in the lower left corner of the card. A circular downward arrow icon is located at the bottom center, and a share/export icon is in the lower right corner. The overall composition should look like a real screenshot from a Chinese AI chat app, showcasing automatically generated car disassembly infographics with modern UI spacing, clear typography, and a luxurious black and blue color scheme.

Example generated from “Xiaoshu Solar Term Infographic Poster”
GPT Image 2en

Goal: Create a refined horizontal Chinese solar-term infographic poster for {argument name="solar term name" default="小暑"}, combining traditional calendar information with a fresh summer dessert and melon color palette. Canvas: 16:9 landscape poster, pale mint-white background, airy negative space, high-end editorial infographic style. Use thin black, teal, lime, peach, orange, and violet accents. The right half is dominated by a soft-focus macro fruit composition: a large cropped cantaloupe or melon slice in warm orange with a bright green rind, plus glossy grapes in lime green and purple at the lower right, with translucent circular overlays and delicate white technical arcs. Layout: Left side contains the typography and data cards; right side contains the fruit photograph-style illustration. Top-left: very large thin-line Chinese title “小暑”, below it spaced Romanization “XIAOSHU” with a dotted horizontal guide line and small circular target mark, then the date “07 / 07”. Far right edge: vertical Chinese slogan “风有热意 瓜果正甜” with a tiny sun icon beneath it. Information modules: Include exactly 7 numbered infographic sections, arranged in a clean grid with fine outlines and dotted dividers. Section 01: pill-shaped green outline card reading “二十四节气 · 第11个” with smaller English subtitle “THE 11TH OF 24 SOLAR TERMS”. Section 02: text block reading “夏季的第5个节气” with English subtitle “THE 5TH SOLAR TERM OF SUMMER”. Section 03: rounded rectangle card titled “气候关键词” with exactly 3 mini icon items: “暑气渐盛” with a sun/cloud icon, “雷雨增多” with a rainy cloud icon, and “湿度上升” with a water-drop triangle icon. Section 04: rounded rectangle card titled “应季风味” with English subtitle “SEASONAL FLAVOR” and exactly 3 mini icon items: “哈密瓜” with melon icon, “葡萄” with grape icon, and “绿豆汤” with bowl icon. Section 05: green arc-accent card titled “生活建议” with exactly 3 mini icon items: “宜清淡饮食” with leaf icon, “及时补水” with water drop icon, and “午后防晒” with sun icon. Section 06: simple text area titled “节气印象” containing the quote “盛夏序章,万物繁茂” and a small sprout icon. Section 07: simple text area titled “农事提示” containing “雨热同期,作物生长旺盛” and a small field/crop icon. Footer: Bottom-left small label “XIAOSHU” followed by tiny tick marks and a target icon. Bottom-center/right label “二十四节气” and “TWENTY-FOUR SOLAR TERMS” with a tiny globe icon and four small pagination dots. Keep all text crisp and minimal. Visual style: Minimalist Swiss-grid layout mixed with Chinese seasonal aesthetics, thin monoline icons, translucent glassmorphism circles, delicate dotted measurement lines, soft gradients, shallow-depth-of-field fruit rendering, luminous summer freshness. Main color inspiration comes from {argument name="color inspiration" default="summer desserts and watermelon colors"}. Use realistic fruit texture but keep the infographic elements flat and precise. Constraints: Use exactly the 7 numbered sections and exactly the item counts listed above. Keep the composition uncluttered, with generous blank space on the left and a dreamy macro fruit crop on the right. No extra logos, no watermark, no additional paragraphs of text.

Example generated from “Xuanwu Gate Incident Moments Post”
GPT Image 2en

A humorous and richly detailed composite image shows a modern smartphone displaying a WeChat Moments update about the Xuanwu Gate Incident on the left, and a battlefield image posted in the update on the right, extending into a grand, cinematic historical scene. The phone appears to be a black iPhone with extremely narrow bezels and a notch at the top, facing forward. On the screen, the app title bar is dark gray-black, displaying the white text "Moments," with a back arrow on the left and a camera icon on the right. Below is a source {argument name="poster name" default="Emperor Taizong of Tang"} The social media post features a square avatar of an ancient Chinese warrior in armor and a helmet. Below the name is a location icon and the text "Chang'an·Xuanwu Gate", along with a timestamp "the fourth day of the sixth month of the ninth year of Wude in the Tang Dynasty, at the hour of Si". The main text of the post reads: "I have eliminated the traitor for the country and pleaded for the state. Today at Xuanwu Gate, my merit will last for a thousand years! ⚔️💪". Below the main text is a preview of a square battlefield image that echoes the large scene on the right. Below the image is a metadata line that precisely displays three items: "38,000 views", "666 likes", and "(128)" next to the comment icon. Five comments are visible in the comment section below, each with a miniature avatar and text style that mimics the social media replies of historical figures. The commenters should include the following Chinese names: "Yuchi Gong", "Fang Xuanling", "Li Yuan", "Du Ruhui" and "Changsun Wuji", and the tone should praise the success of the action; one of the comments should imply that the emperor is shocked and the national situation is settled. On the right is{argument name="gate name" default="Xuanwu Gate"} This is an epic Tang Dynasty battlefield painting depicting the scene at sunset. The background features a massive gray stone city gate with the inscription "Xuanwu Gate." In the foreground, a mounted archer in red-tasseled helmet is drawing his bow, riding a brown horse. Seven Tang soldiers, some on horseback and others standing, are clearly visible around him, while another falls to the ground, pointing upwards. Multiple yellow battle flags bearing the large black characters for "Tang" flutter in the dust; at least four flags are visible. Arrows fly through the sky, warhorses neigh and leap, and the air is filled with dust, the warm glow of the setting sun, and the dynamic atmosphere of the battlefield. The visual style is realistic historical epic art with a painterly quality, but the overall concept possesses a comedic and anachronistic feel, presenting an ancient political upheaval as a social media feed. All visible text must be in Chinese, maintaining reasonable spacing in the mobile UI, and the mobile phone model is integrated with the extended battlefield artwork, presenting a widescreen composition.

Example generated from “Y2K Retro-Tech Youth Portrait”
GPT Image 2en

A vibrant, high-contrast Y2K fashion photograph featuring two young East Asian adults. In the foreground, a young woman smiles gently at the camera, wearing a headband.{argument name="female headwear" default="Plush pink bucket hat"} She wore a light-colored patchwork cropped top, a white pleated skirt, and a beaded necklace, holding a white smartphone. Her long, dark hair flowed gently in the wind. To her right and slightly behind, a young man was laughing excitedly, his mouth slightly open, wearing a {argument name="male headwear" default="Wearing an orange baseball cap backwards"} He was wearing a green T-shirt under a bright blue windbreaker, light khaki trousers, and a silver chain. He held a black smartphone in each hand, one hand raised high. The image was brightly lit with a flash, exhibiting slight wide-angle lens distortion, and the background was {argument name="background color" default="Bright pink"} A smooth gradient transitions to white.

Example generated from “Y2K Teen Magazine Studio Portrait”
GPT Image 2en

A studio portrait of two young East Asian adults standing back to back, with a background of{argument name="background color" default="Gradient color from bright pink to white"} The photo presents{argument name="photography style" default="Y2K's nostalgic 90s teen magazine aesthetic features soft focus, film grain, and halo effects."} On the left is a young woman with long black hair, smiling gently at the camera. She is wearing a layered style.{argument name="woman's outfit" default="A light green printed T-shirt paired with a pink long-sleeved top, a short skirt, and platform sneakers."} He was holding a smartphone. To his right was a young man with short black hair, whose expression {argument name="man's expression" default="Excited and with his mouth agape"} One hand was clenched into a fist and raised in the air. He was wearing {argument name="man's outfit" default="A loose purple T-shirt with yellow trim, loose blue jeans, and a silver chain necklace."} He was also holding a smartphone. The shooting angle was slightly higher, looking down at the subject.

Example generated from “Y2K 金色时刻人像”
GPT Image 2en

candid portrait of a beautiful young blonde woman, 21 years old, glowing sun-kissed skin, wearing a baby pink velour tracksuit and butterfly clips in her hair, smiling brightly at the camera, golden hour warm light, palm trees, y2k aesthetic, glossy lips

Example generated from “Y2K 高楼浴室镜面自拍”
GPT Image 2en

A moody late-night mirror selfie captured in a luxury high-rise bathroom overlooking a glowing city skyline. A young woman with long, slightly messy black hair leans forward over a marble sink, partially obscuring one eye with loose strands. The photo is taken through a smudged mirror with visible dust particles, water spots, and imperfections, creating an authentic raw aesthetic. A powerful direct camera flash explodes in the frame, producing harsh highlights, lens flare, and a nostalgic early-2000s digital camera look. The bathroom counter is cluttered with luxury cosmetics, skincare products, jewelry, a smartphone, and everyday personal items, adding a lived-in editorial feel. Reflections of neon-lit skyscrapers and hotel windows shimmer in the background, emphasizing an urban nightlife atmosphere. Soft pale skin tones contrast against dark clothing and black hair. The composition feels intimate, candid, and effortlessly cool. Style: Y2K flash photography, candid mirror selfie, Korean editorial aesthetic, indie sleaze, hotel bathroom scene, luxury lifestyle, cinematic nightlife, disposable camera look, grainy texture, flash burn, realistic skin texture, shallow depth of field, authentic imperfections, viral Pinterest mood, high-fashion casual energy, raw unfiltered photography. Camera: direct flash, compact digital camera, 35mm equivalent, harsh lighting, slight motion blur, visible noise, natural color grading, vertical composition, ultra-realistic, high detail.

Example generated from “Yami Kawaii Idol Selfie”
GPT Image 2zh

One{argument name="idol theme" default="Dark, yandere, cute idol"} A vertical close-up self-portrait of a girl, set in a narrow, neon-pink room. The camera angle shows her arm outstretched towards the camera, with a shimmering sleeve and a blurred hand visible in the foreground. She wears a richly detailed black and pink plaid stage costume, adorned with glossy vinyl straps, lace trim, ribbons, buckles, and layered accessories: a sleeveless corset, black mesh split sleeves, multiple black armbands on one upper arm, a pink bow, and a heart-shaped badge on her chest inscribed with "病みかわいい" (I feel sick). Her long, dark brown hair is styled in a high bun, with long, straight strands framing her face and decorated with large bows. Her visible hand is close to her lips, striking a shy idol pose, wearing pink lace fingerless gloves with ruffles, long manicures, and chunky rings. A pink armband with handwritten Japanese characters is added to her upper arm, along with delicate earrings. Her face is obscured by a large rectangular mosaic in the center, covering most of her forehead and chin. The background is cluttered and decadent, filled with black and pink graffiti, plastic curtains, plush toys, and distressed decorations. The scene must contain four clearly visible Japanese texts: graffiti on the left reading "闇ー?", below it "嫌いにならないで", and further below that "私だけ見ててね". A glowing heart-shaped neon sign in the upper right corner reads "病みしか勝たん". Two plush toys are placed in the background: a light pink plush animal on the left and a black plush toy with X-shaped eyes on the right. The lighting is a melancholic nightclub-style magenta and violet with glossy highlights, creating a chaotic underground idol dressing room atmosphere. High detail, luxurious texture, glitter, lace, vinyl materials, a messy yet cute gothic atmosphere, high contrast, smartphone selfie composition, and a vertical screen ratio.

Example generated from “Yoga Book Two-Page Spread”
GPT Image 2en

{ "type": "Two-page yoga instruction manual layout", "subject": "An Asian woman with her hair tied in a high bun, wearing {argument name=\"subject outfit\" default=\"Blue sports bra and matching leggings\"}", "layout": { "left_page": { "header": { "title_english": "{argument name=\"pose name english\" default=\"Eka Pada Koundinyasana I\"}", "subtitle_english": "One-Legged Sage Koundinya's Pose I", "title_korean": "{argument name=\"pose name korean\" default=\"에카 파다 콘디니아사나 I\"}", "subtitle_korean": "한 다리로 균형을 잡는 현자의 자세 I" }, "main_image": "A large photograph of a subject performing a full arm balance pose on a gray yoga mat.", "description": "A Korean text explaining the pose", "bottom_columns": [ { "title": "effect", "count": 3, "description": "List of health benefits" }, { "title": "Ready position", "count": 3, "description": "List of preparatory actions" } ], "footer": "No.{argument name=\"page number left\" default=\"102\"} Page and chapter titles" }, "right_page": { "header": "HOW TO DO / Action Guide", "steps_grid": { "count": 4, "description": "A step-by-step guide from 1 to 4, including Korean instructions and advanced photos.", "images": [ "Step 1: Sit down and prepare your arms and legs.", "Step 2: Lean forward and place your hands on the mat.", "Step 3: Lift off the ground", "Step 4: Complete the balance posture" ] }, "tip_box": { "title": "hint", "count": 3, "description": "List of key points in Korean within the shaded box" }, "footer": "Chapter titles and numbers{argument name=\"page number right\" default=\"103\"} Page" } } }

Example generated from “YouTube 2030 Interface”
GPT Image 2en

{argument name="year" default="2030"} Year{argument name="platform" default="YouTube"} Homepage screenshot

Example generated from “YouTube Dark Mode Desktop UI Mockup”
GPT Image 2en

{ "type": "Desktop web browser UI screenshots", "environment": "{argument name=\"os environment\" default=\"Windows 11 desktop\"}", "theme": "{argument name=\"ui theme\" default=\"Dark mode\"}", "browser_window": { "tabs": [ "YouTube", "New tab" ], "url_bar": "youtube.com", "profile_icon": "{argument name=\"profile initial\" default=\"M\"}" }, "app_interface": { "header": { "logo": "{argument name=\"website name\" default=\"YouTube\"} GB", "search_bar": "search", "icons": [ "microphone", "create", "notify", "Personal information" ] }, "filter_bar": { "active_filter": "{argument name=\"active filter\" default=\"all\"}", "count": 11, "labels": [ "all", "game", "music", "live streaming", "podcast", "compilation", "science", "Lo-fi", "Thriller", "Latest Upload", "Viewed" ] }, "sidebar": { "sections": [ { "title": "main", "count": 3, "labels": [ "front page", "Shorts", "Subscription content" ] }, { "title": "I", "count": 6, "labels": [ "Your channel", "Historical records", "Playlist", "Your video", "Watch later", "Liked Videos" ] }, { "title": "Subscription content", "count": 8, "labels": [ "Linus Tech Tips", "MrBeast", "Kurzgesagt - In a Nutshell", "ColdFusion", "3Blue1Brown", "Khan Academy", "Veritasium", "Show more" ] }, { "title": "explore", "count": 7, "labels": [ "Popular", "music", "Movie", "live streaming", "game", "news", "physical education" ] }, { "title": "More YouTube services", "count": 3, "labels": [ "YouTube Premium", "YouTube Music", "YouTube Kids" ] } ] }, "main_content": { "layout": "3 columns and 4 rows grid", "count": 12, "video_cards": [ { "title": "Why building your own GPU is almost impossible", "channel": "Linus Tech Tips", "stats": "164,000 views • 1 day ago", "thumbnail": "CPU/GPU chips" }, { "title": "Survive in the circle for 100 days and win $500,000.", "channel": "MrBeast", "stats": "40 million views • 5 days ago", "thumbnail": "Mr. Beast, dressed in a blue tracksuit, and the crowd." }, { "title": "What would happen if we destroyed an asteroid with a nuclear bomb?", "channel": "Kurzgesagt - In a Nutshell", "stats": "4.2 million views • 6 hours ago", "thumbnail": "Missile hits asteroid in space" }, { "title": "How Dubai Became a $10 Billion City", "channel": "ColdFusion", "stats": "1.6 million views • 2 weeks ago", "thumbnail": "Man with Dubai skyline" }, { "title": "The beautiful mathematics of Euler's identity", "channel": "3Blue1Brown", "stats": "2.8 million views • 3 months ago", "thumbnail": "Mathematical formulas and spiral diagrams" }, { "title": "The World in 2040 - Population, Climate, and More", "channel": "Veritasium", "stats": "7.7 million views • 2 weeks ago", "thumbnail": "Aerial view of dense forest" }, { "title": "I've used Linux for 12 years, and finally...", "channel": "DistroTube", "stats": "160,000 views • 3 days ago", "thumbnail": "Laptops and smartphones with the Linux logo" }, { "title": "Black Holes Explained - From Birth to Death", "channel": "Khan Academy", "stats": "3.9 million views • 3 months ago", "thumbnail": "Glowing black hole accretion disk" }, { "title": "The problem with dietary recommendations", "channel": "Andrew Huberman", "stats": "4.4 million views • 6 days ago", "thumbnail": "The man who spoke while wearing glasses" }, { "title": "How to learn anything (based on evidence)", "channel": "Ali Abdaal", "stats": "2.7 million views • 3 months ago", "thumbnail": "Vector illustration of a man reading a book" }, { "title": "The Last of Us Part II is a masterpiece.", "channel": "Naughty Dog", "stats": "2.3 million views • 3 years ago", "thumbnail": "Ellie from The Last of Us" }, { "title": "Lofi hip hop radio - beats perfect for relaxation/study.", "channel": "Lofi Girl", "stats": "During the live stream", "thumbnail": "Lofi Girl studying at her desk" } ] } }, "taskbar": { "os": "Windows", "icons": [ "start", "search", "Task View", "File Explorer", "Edge", "Spotify" ], "system_tray": "08:34 April 3, 2024" } }

Example generated from “YouTube Dark Mode Desktop UI Mockup”
GPT Image 2en

{ "type": "Desktop UI Prototype", "application": "YouTube Homepage", "theme": "{argument name=\"theme\" default=\"Dark mode\"}", "browser_window": { "url_bar": "{argument name=\"browser URL\" default=\"youtube.com\"}", "tabs": 1, "elements": [ "Back/Forward button", "refresh", "extensions", "Profile picture" ] }, "layout": { "sidebar": { "position": "Left side", "sections": [ { "title": "Homepage", "count": 3, "items": [ "front page", "Shorts", "Subscription content" ] }, { "title": "I", "count": 6, "items": [ "Your channel", "Historical records", "Playlist", "Your video", "Watch later", "Liked Videos" ] }, { "title": "Subscription content", "count": 8, "items": [ "Linus Tech Tips", "MrBeast", "Kurzgesagt - In a...", "ColdFusion", "3Blue1Brown", "Khan Academy", "Veritasium", "Show more" ] }, { "title": "explore", "count": 7, "items": [ "Popular", "music", "Movie", "live streaming", "game", "news", "physical education" ] }, { "title": "More YouTube services", "count": 3, "items": [ "YouTube Premium", "YouTube Music", "YouTube Kids" ] } ] }, "top_filter_bar": { "count": 11, "labels": [ "all", "game", "music", "live streaming", "podcast", "compilation", "science", "Lo-fi", "Thriller", "Latest Upload", "Viewed" ] }, "video_grid": { "layout": "4 rows and 3 columns", "count": 12, "videos": [ { "title": "Why building your own GPU is almost impossible", "channel": "Linus Tech Tips", "thumbnail": "GPU chip with red arrow" }, { "title": "Survive in the circle for 100 days and win $500,000.", "channel": "MrBeast", "thumbnail": "MrBeast is smiling, with contestants in the background." }, { "title": "What would happen if we destroyed an asteroid with a nuclear bomb?", "channel": "Kurzgesagt - In a Nutshell", "thumbnail": "Missile approaches asteroid in space" }, { "title": "How Dubai Became a $10 Billion City", "channel": "ColdFusion", "thumbnail": "A man stands beside the Dubai skyline with the text '$0 TO $10B' displayed." }, { "title": "The Mathematical Beauty of Euler's Identity", "channel": "3Blue1Brown", "thumbnail": "Mathematical formulas and spirals on a black background" }, { "title": "The World in 2040 - Population, Climate, and More", "channel": "Veritasium", "thumbnail": "A lush forest with the text 'The World in 2040'" }, { "title": "After using Linux for 12 years, finally...", "channel": "DistroTube", "thumbnail": "Laptop and smartphone connected by a blue arrow." }, { "title": "Black Holes Explained - From Birth to Death", "channel": "Khan Academy", "thumbnail": "The glowing orange accretion disk of the black hole" }, { "title": "Dietary recommendations", "channel": "Andrew Huberman", "thumbnail": "A man wearing a black shirt; the text reads 'the problem with diet advice'." }, { "title": "How to learn anything (based on evidence)", "channel": "Ali Abdaal", "thumbnail": "Flat illustration: A man reading a book, surrounded by colorful leaves." }, { "title": "The Last of Us Part II is a masterpiece.", "channel": "Naughty Dog", "thumbnail": "Ellie's resolute portrait, featuring the game logo." }, { "title": "Lofi hip hop radio - beats perfect for relaxation/study.", "channel": "Lofi Girl", "thumbnail": "An anime girl is studying at her desk, with the city skyline at sunset visible through the window." } ] } }, "os_environment": { "os": "{argument name=\"operating system\" default=\"Windows 10\"}", "taskbar_position": "bottom", "visible_icons": [ "start", "Search bar", "Task View", "File Explorer", "Edge", "Spotify" ], "system_tray": [ "network", "volume", "Battery", "A clock displaying the time 08:34 on April 3, 2024." ] } }

Example generated from “YouTube Gaming Stream Despair Scene”
GPT Image 2en

A popular Japanese YouTuber{argument name="content" default="Game streaming"} Scene. In{argument name="game name" default="Tetris"} In the video, the long bar seemed poised to eliminate all the blocks, but its placement was slightly off. The streamer screamed in despair, his face contorted in disbelief, and the comment section descended into chaos. (Japanese)

Example generated from “YouTube Live Presentation UI Mockup”
GPT Image 2en

{ "type": "YouTube desktop dark mode UI mockup", "header": { "logo": "YouTube", "search_bar": "search", "icons_count": 5, "icons": [ "search", "microphone", "create", "notify", "Personal information" ] }, "video_player": { "top_left_badge": "live streaming", "left_side": { "subject": "{argument name=\"presenter description\" default=\"A man wearing a green sweater standing on a wooden podium\"}", "podium_logo": "OpenAI" }, "right_side_presentation": { "text_elements": [ "OpenAI", "Introduced with great fanfare", "{argument name=\"product name\" default=\"GPT-Image-2\"}", "{argument name=\"tagline\" default=\"More realistic. More practical. More creative.\"}" ], "sample_images_count": 4, "sample_images": [ "Mountain lakes with boats", "A female portrait with dappled light and shadow", "A cute robot carrying a lantern in the forest", "Starry Night Cafe paintings" ] }, "bottom_controls_count": 10, "bottom_controls": [ "pause", "Next", "volume", "live streaming", "Red progress bar", "subtitle", "set up", "Mini Player", "Cinema mode", "full screen" ] }, "video_details": { "title": "{argument name=\"video title\" default=\"OpenAI Live Stream: GPT-Image-2 Officially Launched\"}", "channel": { "name": "{argument name=\"channel name\" default=\"OpenAI\"}", "verified": true, "subscribers": "1.36 million", "button": "subscription" }, "action_buttons_count": 5, "action_buttons": [ "Likes 12,000", "I don't like 497", "share", "keep", "More" ], "description_box": { "stats": "95,237 people are currently watching; the live stream started 7 minutes ago.", "tags": "#OpenAI #GPTImage2 #AI", "text": "Welcome to our special live stream event where we'll be introducing GPT-Image-2, our latest and most advanced image generation model. Watch new feature demos, live demonstrations, and hear from the team… and more." } } }

Example generated from “YouTube Live Stream Announcement Mockup”
GPT Image 2en

{ "type": "YouTube live streaming interface demo", "theme": "Dark mode", "layout": { "top_bar": { "components": [ "Hamburger menu", "YouTube logo", "Search bar", "microphone icon", "Create icon", "Notification icon", "Personal profile picture" ] }, "video_player": { "status": "The LIVE logo is displayed in the top left corner.", "left_side": "One{argument name=\"presenter description\" default=\"The man wearing a green sweater\"} Standing in front of a wooden lectern with a spiral logo, they are communicating with gestures, and a laptop is placed on the lectern.", "right_side_presentation": { "text_elements": [ "OpenAI", "Introduced with great fanfare", "{argument name=\"product name\" default=\"GPT-Image-2\"}", "{argument name=\"tagline\" default=\"More realistic, more practical, and more creative.\"}" ], "image_gallery": { "count": 4, "panels": [ "Mountain lakes with small boats", "A close-up portrait of a woman with dappled light and shadow", "A cute robot carrying a lantern in the dark forest.", "Impressionist paintings of city streets illuminated at night" ] } }, "bottom_controls": { "components": [ "Play/Pause", "Next", "volume", "LIVE Text", "Red progress bar", "CC", "Set gears", "Mini Player", "Cinema mode", "full screen" ] } }, "video_info": { "title": "{argument name=\"video title\" default=\"OpenAI Live Stream: GPT-Image-2 Officially Launched\"}", "channel_section": { "name": "{argument name=\"channel name\" default=\"OpenAI\"}", "badge": "Authentication checkmark", "subscribers": "1.36 million subscribers", "button": "Subscribe (white capsule button)" }, "action_buttons": { "count": 5, "labels": [ "Likes (12,000)", "Dislike (497)", "share", "keep", "..." ] }, "description_box": { "metadata": "95,237 people are currently watching; the live stream started 7 minutes ago.", "hashtags": "#OpenAI #GPTImage2 #AI", "body_text": "Welcome to our special live stream event where we'll be showcasing OpenAI's latest and most advanced image generation model, GPT-Image-2. Learn about new features, watch a live demo, and hear from the team…and more." } } } }

Example generated from “YouTube Live Stream Announcement Mockup”
GPT Image 2en

{"type": "YouTube Live Stream UI mockup", "theme": "dark mode", "layout": {"top_bar": {"components": ["hamburger menu", "YouTube logo", "search bar", "microphone icon", "create icon", "notifications icon", "profile avatar"]}, "video_player": {"status": "LIVE badge in top left", "left_side": "A {argument name=\"presenter description\" default=\"man in a green sweater\"} standing at a wooden podium with a spiral logo, gesturing with hands, laptop on podium.", "right_side_presentation": {"text_elements": ["OpenAI", "INTRODUCING", "{argument name=\"product name\" default=\"GPT-Image-2\"}", "{argument name=\"tagline\" default=\"More Realistic. More Useful. More Creative.\"}"], "image_gallery": {"count": 4, "panels": ["mountain lake with a boat", "close-up portrait of a woman with dappled lighting", "cute robot holding a lantern in a dark forest", "impressionist painting of a glowing city street at night"]}}, "bottom_controls": {"components": ["play/pause", "next", "volume", "LIVE text", "red progress bar", "CC", "settings gear", "miniplayer", "theater mode", "fullscreen"]}}, "video_info": {"title": "{argument name=\"video title\" default=\"OpenAI Live: Introducing GPT-Image-2\"}", "channel_section": {"name": "{argument name=\"channel name\" default=\"OpenAI\"}", "badge": "verified checkmark", "subscribers": "1.36M subscribers", "button": "Subscribe (white pill button)"}, "action_buttons": {"count": 5, "labels": ["Like (12K)", "Dislike (497)", "Share", "Save", "..."]}, "description_box": {"metadata": "95,237 watching now, Started streaming 7 minutes ago", "hashtags": "#OpenAI #GPTImage2 #AI", "body_text": "Join us for a special live event as we introduce GPT-Image-2, our latest and most advanced image generation model. See new capabilities, live demos, and hear from the team ...more"}}}}

Example generated from “YouTube Livestream Interface Mockup”
GPT Image 2en

{ "type": "YouTube Live UI", "top_nav": { "logo": "YouTube Premium", "search": "{argument name=\"search query\" default=\"bilal fraiha\"}", "icons": 3 }, "player": { "subjects": [ "{argument name=\"female guest\" default=\"Sydney Sweeney\"} Wearing a white cardigan", "A bearded man wearing a beige jacket is laughing." ], "bg": "Sofa, two silver play buttons, and the 'SARDI' ram logo.", "overlays": { "chat": { "pos": "Left side", "count": 15, "desc": "Colored usernames, white text" }, "goal": { "pos": "Top right corner", "text": "Tonight's goal: 0 to 25" }, "banner": { "pos": "Bottom Center", "text": "K {argument name=\"streamer name\" default=\"MOREBILAL\"}" } }, "controls": { "count": 10 } }, "details": { "title": "{argument name=\"video title\" default=\"FULL STREAM | سيدني سويني مع بلال\"}", "channel": "{argument name=\"channel name\" default=\"More Bilal No Filter\"}", "buttons": 5 } }

Example generated from “YouTube Livestream Interface Mockup”
GPT Image 2en

{ "type": "YouTube livestream UI", "top_nav": { "logo": "YouTube Premium", "search": "{argument name=\"search query\" default=\"bilal fraiha\"}", "icons": 3 }, "player": { "subjects": [ "{argument name=\"female guest\" default=\"Sydney Sweeney\"} in white cardigan", "bearded man in beige jacket laughing" ], "bg": "couch, 2 silver play buttons, ram logo 'SARDI'", "overlays": { "chat": {"pos": "left", "count": 15, "desc": "colored usernames, white text"}, "goal": {"pos": "top right", "text": "TONIGHT'S GOAL: 0 to 25"}, "banner": {"pos": "bottom center", "text": "K {argument name=\"streamer name\" default=\"MOREBILAL\"}"} }, "controls": {"count": 10} }, "details": { "title": "{argument name=\"video title\" default=\"FULL STREAM | سيدني سويني مع بلال\"}", "channel": "{argument name=\"channel name\" default=\"More Bilal No Filter\"}", "buttons": 5 } }

Example generated from “YouTube Unboxing Thumbnail Prompt”
GPT Image 2en

Thumbnail of a YouTube unboxing video; the video's theme is {argument name="topic" default="Open all overdue bills"} ,{argument name="quantity" default="100 consecutive copies"}

Example generated from “YouTube Video Interface Mockup”
GPT Image 2en

{ "type": "Desktop video platform UI model", "theme": "Dark mode", "header": { "logo": "YouTube IN", "search_bar": "Centered, with a magnifying glass icon", "right_icons": { "count": 4, "items": [ "microphone", "+ Create Button", "Notification bell", "User Avatar" ] } }, "video_player": { "scene": "In the studio setting, two tech bloggers sit at a table.", "subjects": [ "{argument name=\"left person\" default=\"Marques Brownlee (MKBHD)\"} Wearing a black T-shirt", "{argument name=\"right person\" default=\"Arun Maini (Mrwhosetheboss)\"} Wearing a green shirt", "{argument name=\"robot model\" default=\"Tesla Optimus humanoid robot\"} Standing among them" ], "props": [ "Black mug with logo", "Background screen with glowing blue neon M logo" ], "controls_bar": { "progress_bar": "Red, near the starting position", "timestamp": "0:01 / 18:47", "left_icons": { "count": 3, "items": [ "Play/Pause", "Next song", "volume" ] }, "right_icons": { "count": 6, "items": [ "CC Subtitles", "Setting with 4K logo", "Mini Player", "Cinema mode", "Screen mirroring", "full screen" ] } } }, "video_details": { "title": "{argument name=\"video title\" default=\"Tesla Optimus Robot Review: The Future is Here!\"}", "channel_info": { "name": "{argument name=\"channel name\" default=\"Mrwhosetheboss\"}", "badges": [ "Authentication check mark" ], "subscribers": "20.5 million subscribers", "button": "White subscription button" }, "action_buttons": { "count": 6, "items": [ "Likes (215,000) / Dislikes (4,200) Capsule Button", "share", "download", "Editing", "keep", "More(...)" ] } }, "description_box": { "metadata": "3,278,456 views. Ranked #7 in the trending topics on April 12, 2024.", "text": "MKBHD and I reviewed Tesla's Optimus humanoid robot. We covered its design, mobility, AI, real-world testing, and its implications for the future!", "links": [ "Marques's channel: @mkbhd" ] } }

Example generated from “Yuzu Sparkling Water Advertisement”
GPT Image 2zh

{ "type": "Commercial beverage advertising posters", "visuals": { "model": "A young Asian woman, eyes closed, face slightly tilted back, expression refreshed, wet hair, water droplets on her skin, wearing a semi-transparent white top.", "product": "One white-yellow aluminum can with condensation, labeled '{argument name=\"product name\" default=\"YUZU SPARK\"} ' and'{argument name=\"brand name\" default=\"KIRARA\"} ', printed with a pomelo illustration", "props": "Two whole yellow grapefruits with green leaves, floating water droplets", "background": "Clear blue sky, bright sunshine" }, "layout": { "top_left": "'with sun rays icon'{argument name=\"brand name\" default=\"KIRARA\"} 'logo'", "left_column": [ "Large vertical text:'{argument name=\"main catchphrase\" default=\"A new me, scented with yuzu.\"} '", "With bright yellow vertical lettering: 'A gentle reward for your hard work every day.'", "Yellow cursive text: 'Refresh!'" ], "right_column": [ "White handwritten text: '{argument name=\"handwritten text\" default=\"A single swirl that will untie me.\"} '", "Yellow circular badge: 'NEW'" ], "bottom_left": [ "Yellow circular badge:{argument name=\"flavor profile\" default=\"3% domestically produced yuzu juice\"} '", "Small print: 'Clean and light. A moment of gentle sweetness as a reward.'", "Minor disclaimer: *Image is for illustrative purposes only." ], "footer": "yellow width, inclusion'{argument name=\"product name\" default=\"YUZU SPARK\"} ', 标语 'A gentle burst of yuzu's light carbonation.' 以及角框文字 'Limited quantity on sale now'" } }

Example generated from “Zar⭕on #38”
GPT Image 2en

A stylish man wearing a tan suit and sunglasses is featured in a creative split image, framed by two ornate gold frames. In the top frame, he is pouring Coca-Cola from a bottle. The cola streams downward, passing through the gap between the frames. In the bottom frame, the same man, also in a tan suit and sunglasses, is shown with a joyful expression as the stream of cola splashes into a glass mug he is holding, creating a dynamic splash.The background behind the frames is a softly blurred night cityscape with tall buildings and lights, suggesting.

Example generated from “Zelda Map in Minecraft Style”
GPT Image 2en

Generate a{argument name="base style" default="Minecraft"} The game screenshots, in a style that builds the world into{argument name="map theme" default="The Legend of Zelda: Tears of the Kingdom"} The map scene.

Example generated from “Zombie Survival RPG Anime Ad Poster”
GPT Image 2zh

Create a dramatic Japanese-style zombie survival RPG mobile game advertisement poster, employing a style that blends anime and cinematic elements. The poster features a vertical smartphone poster layout with high-density advertising design. The background is a brutal post-apocalyptic city at dusk, including abandoned streets, smoke, sparks, debris, damaged buildings, neon lights, and a horde of zombies approaching from behind. In the center of the image, a young anime woman sits on the ground in a slightly languid posture, her legs bent, one hand gripping a black pistol vertically. She has shoulder-length, straight, dark brown to black hair, pale skin, and wears an oversized, dirty white T-shirt stained with blood. The T-shirt features the striking Japanese calligraphy "働きたくね〜" vertically printed on it. Her facial expression area is handled simply and unobtrusively to ensure the composition focuses on the T-shirt, weapon, and atmosphere. Around her, several zombies in tattered clothing reach out from the background on the left and right, grabbing forward. The game's main visual design is impactful, comprising eight clearly defined advertising areas. Area 1: The top left corner game logo area contains three text elements: a large, distressed English title "Dead or work," a smaller Japanese subtitle "Zonbi Sabair RPG," and a heavy black splatter background with white and red text. Area 2: The right-side vertical slogan area contains two text elements: on a rough black stroke bar, the upper text is white Japanese "動かない、ただ," and the lower text is large red vertical text "生き残れ." (Life is shattered). Area 3: Function list panel in the middle left, including 4 Line function, each line has a white icon inside a black square and yellow/white Japanese copy; the labels are "ゾンビを下して生き Yan" "びろ", "拠PointをStrengthening・カスタマイズ", "Nakama's collaborative effort" and "デイリーミッションでReward GET". Area 4: Bonus activity panel in the middle right, containing 1 title and 3 Bonus item; the title is in a bright jagged yellow explosion box that reads "Now the beginning of the day, Deluxe Special Bonus!" below. There are three item frames labeled "Surface × 5000", "Rea Weapon (7 Days)", and "Surface Weapon × 10". Area 5: The middle and lower game screenshot bar, including 4 thumbnails with captions; the captions are "リアルタイムバトル!", "Base is free"にカスタム!", "Nakama とレイドボスにpickめ!" and "広大なオープンワールド!". Area 6: Bottom call-to-action banner with two text elements, featuring a large yellow and red Japanese phrase “今すぐダウンロード!” and a smaller white/red subtitle “あなたの选択が、生死を分ける.”. Area 7: Bottom right store badge with two badges displaying App Store and Google Play-style download buttons. Area 8: Bottom right QR code area with a QR code within a white square. Overall Aesthetics: Ultra-refined, high-contrast, dirty, intense, and commercially sophisticated gacha/mobile game ad combining anime character art with a realistic zombie horror background. Utilizes striking distressed fonts, black brushstroke textures, red warning colors, yellow promotional highlights, metallic UI borders, and cinematic orange-gray lighting.

Example generated from “Zoomed Fake Pixel Art Grid Analysis”
GPT Image 2en

A closely magnified comparison image showcases the rendering effect of pseudo-pixel art. The image presents a magnified view of a low-resolution scene, overlaid with a fine, bright blue square grid across the entire frame. The composition is an abstract close-up of a pixel art-style environment: sandy brown and ochre blocks on the left and center, stepped diagonal blocks in deep navy blue and black on the right, and soft olive green and slate blue shapes above the background, accented with a few brighter gold highlights. At first glance, each edge appears stepped and blocky, but the colors soften and blend within the obvious squares rather than perfectly aligning with the grid, creating a pixel art illusion that doesn't stand up to close scrutiny. A small white crosshair cursor is included near the upper left quadrant, hinting at the image analysis process in the editor. The overall look resembles a screenshot from a graphics program or social media demo, intended to illustrate how AI can mimic the aesthetics of retro game art when it cannot generate truly grid-perfect pixel art; no visible text labels should be included in the image.

Example generated from “Zoomed Fake Pixel Art Grid Analysis”
GPT Image 2en

A tightly zoomed-in comparison image demonstrating fake pixel-art rendering, showing a low-resolution-looking scene enlarged on screen with a thin bright blue square grid overlay across the entire frame. The composition is an abstract close crop of a pixel-art-style environment: sandy tan and ochre blocky forms on the left and center, a dark navy and black stepped diagonal mass on the right, muted olive-green and slate-blue shapes along the upper background, and a few lighter golden highlights. Every edge looks staircase-like and blocky at first glance, but the colors softly blur and blend within the apparent squares instead of aligning perfectly to the grid, creating the impression of pixel art that breaks down under inspection. Include a small white crosshair cursor near the upper left quadrant to suggest image analysis in an editor. The overall look should feel like a screenshot from a graphics program or social media demonstration about how AI can imitate retro game art aesthetics without producing true grid-perfect pixel art, with no text labels visible.

Example generated from “Zoomer UI Design Concept”
GPT Image 2en

GPT Image 2 is an operating system designed for Zoomer, with an ever-increasing {argument name="emotion" default="anxiety"} Mood Level

Example generated from “Zoomer UI Design Concept”
GPT Image 2en

gpt image 2 operating system for zoomers with increasing levels of {argument name="emotion" default="anxiety"}

Example generated from “ΛRMIN | AI #187”
GPT Image 2en

A luxury [BRAND] brand advertisement featuring three stylish [ATHLETES / MODELS / PEOPLE] arranged in a single horizontal 1×3 grid, each posed in their own separate architectural window frame. Composition & framing: 3 vertical window niches aligned side-by-side horizontally, Each character shown in a waist-up mid-shot, Fully inside the frame, with hands and props naturally breaking the bottom/side frame edges, No legs or lower body visible, hands anatomically correct, no missing fingers, no deformed limbs. All 3 frames must be fully visible, no cropping at the top or bottom, Clean, symmetrical, editorial alignment, consistent camera perspective, photorealistic fashion photography, DSLR, 85mm lens, shallow depth of field, natural skin texture, real pores, realistic fabric texture Character slots (left to right): [CHARACTER 1] wearing [OUTFIT 1], holding [OBJECT 1], in [POSE/ACTION 1], [EXPRESSION 1], [ACCESSORIES 1]. [CHARACTER 2] wearing [OUTFIT 2], holding [OBJECT 2] or doing [GESTURE 2], hands centered, breaking the frame edge outward (not sideways), [EXPRESSION 2], [ACCESSORIES 2]. [CHARACTER 3] wearing [OUTFIT 3], holding/carrying [OBJECT 3], projecting [VIBE 3], [EXPRESSION 3], [ACCESSORIES 3]. Style: Authentic [BRAND] premium styling, couture-inspired detailing, expressive editorial fashion poses, glamorous, iconic, fashion-forward campaign vibe. Backdrop & materials: Elegant [BACKGROUND COLOR + MATERIAL] wall backdrop with subtle textures, refined architectural detailing, premium finish. Lighting & camera: High-fashion bright studio lighting, soft shadows, high contrast contours, realistic reflections, commercial studio photography, no CGI, no illustration, crisp product highlights. Aspect ratio & layout control: Ultra-wide horizontal composition, 1×3 grid, all three niches visible, balanced spacing.

Example generated from “ΛRMIN | AI #20”
GPT Image 2en

A [person type] standing full-body near the edge of the frame, slightly to the left, in [real-world incomplete / empty / damaged environment]. The camera is positioned behind and slightly to the right of the person, capturing them from a three-quarter rear angle. The person stands calmly and thoughtfully, looking toward the space ahead. The environment is [describe the real space as it exists now: empty, unfinished, decayed, rigid, lifeless]. The atmosphere feels [emotional tone: quiet, dry, historical, minimal, raw]. Natural light illuminates the scene, creating realistic shadows and depth. Floating hand-drawn sketch lines appear over and within the environment, aligned precisely in perspective with the real space. These sketches represent [what is missing / what once existed / what could exist in the future]. The sketch lines are black, clean, and hand-drawn, resembling [architectural / interior / restoration / urban planning] concept sketches made with pencil or ink. The strokes are expressive, slightly imperfect, and organic — not rigid or mechanical. The sketched elements occupy the foreground and midground, appearing closer to the person than expected. Some sketch lines subtly overlap the person’s silhouette and the real surfaces, visually merging imagination with reality. The sketches float naturally within the scene, seamlessly blending with the environment and conveying the idea of [vision / restoration / transformation / potential] coming to life.

Example generated from “いにしえ@AI Director & Creator|Will Oldgram #21”
GPT Image 2mixed

画像生成: 商品広告写真, 夏にぴったりの季節商品, 炭酸飲料, 名称="夏柑SODA", 形状=ペットボトル500ml, 2025年に飲料広告として高CTAだったデザインをリサーチしてから画像仕様をデザインして生成, アスペクト比=3:4

Example generated from “みどり🐲Midori Tatsuta #18”
GPT Image 2en

Cute Y2K style affordable cosmetics advertising image designed for Gen Z. Using vibrant color schemes, including neon colors. Aspect ratio is 3:4.

Example generated from “ヨシダ@AIエンタメ #46”
GPT Image 2mixed

日本語の架空広告をテーマに、SNSでバズることを目的としたインパクトの強いビジュアルを作成してください。 1:1の正方形キャンバスに、4種類の広告を「2×2グリッド」で配置してください。 ジャンル: ①テック(AI・アプリ) ②美容・コスメ ③飲食 ④フィットネス すべて日本語コピー入り・架空ブランドで作成。 【バズるための演出要件】 ・一瞬で目を引く強烈なキャッチコピー ・感情を動かす要素(驚き・共感・焦り・欲望) ・コントラストの強い配色(ネオン、ビビッドカラー、黒×蛍光など) ・大胆なレイアウト(はみ出し・斜め配置・大きすぎる文字) ・サムネ映えする構図(中央強調・顔アップ・視線誘導) ・SNS広告っぽい“あざとさ”(矢印・強調枠・吹き出し) 【デザインスタイル】 ・2025年トレンドのSNS広告風 ・TikTok / Instagram広告風 ・視認性最優先(スマホで見ても読める) ・写真風 or 高品質イラスト ・CTA入り(例:今だけ無料・限定・残り3日など) 【ランダムバズ要素】 以下の①〜③から、それぞれ1つずつランダムに選び、必ずキャッチコピーに反映すること。 ①禁断ワード系(いずれか1つ) ・知らないと損 ・99%が間違ってる ・今すぐやめて ②ギャップ演出(いずれか1つ) ・見た目かわいいのに中身がヤバい ・普通に見えて実は… ・これ、○○じゃありません ③数字・限定(いずれか1つ) ・3日限定 ・残り1枠 ・無料→0円 【重要】 ・4つそれぞれ明確に違う雰囲気にする ・グリッドは崩さない ・文字は読みやすく配置する ・“スクロールを止める違和感”を意図的に入れる 出力:4広告が1枚にまとまった高インパクトビジュアル

Example generated from “一张中文健身信息图”
GPT Image 2zh

请生成一张中文健身信息图,主题为:【xxx】。 要求这张图既专业又实用,适合普通成年人作为训练参考。默认对象为无严重伤病的健康成年人;如果没有额外说明,默认训练目标为“增肌 + 基础力量提升”,默认训练水平为“新手到中级之间”,默认训练场景为“普通健身房”,默认单次训练时长控制在 40–60 分钟内。 请根据【训练主题】自动判断输出类型: 1)如果【训练主题】是某个肌群或身体部位(例如:胸肌、背阔肌、肱二头肌、腹肌、肩部、腿部等),请输出一张“该部位训练计划信息图”。 2)如果【训练主题】是某个动作或技能目标(例如:引体向上、俯卧撑、双杠臂屈伸、深蹲等),请输出一张“动作解锁 / 进阶训练计划信息图”。 整张图请采用清晰、现代、专业、易读的中文信息图风格,竖版排版,视觉简洁,重点突出,适合社交媒体分享或训练参考卡片。不要写成长篇大论,每个模块用简洁短句呈现,数字信息要醒目。 这张信息图必须包含以下内容: 【A. 标题区】 - 主标题:直接写【训练主题】训练计划 / 解锁计划 - 副标题:自动补充适用人群、目标、训练场景、建议时长 例如:适合新手 / 增肌导向 / 健身房版 / 45分钟 【B. 训练目标区】 用简洁语言说明: - 这次训练主要针对什么 - 主要目标是什么(增肌 / 力量 / 技能解锁 / 核心控制等) - 本次训练的重点刺激或能力提升方向 【C. 热身区】 给出 2–4 个热身建议,简洁列出即可,例如: - 动态活动 - 目标肌群激活 - 轻重量预热组 每项可附一句说明 【D. 主训练区】 这是核心部分,请列出 4–6 个主要训练动作。 每个动作都要包含以下信息: - 动作名称 - 训练作用 / 针对部位 - 组数 × 次数(或时间) - RIR 建议 - 每组间休息时间 - 动作关键要点(1–2 条) - 常见错误(1 条即可) 请确保动作安排合理: - 先复合动作,后孤立动作 - 整体训练量适中 - 新手不要安排过度极限训练 - 主动作通常建议 RIR 1–3 - 孤立动作可建议 RIR 0–2 - 如果是腹肌或核心类动作,可用“秒数 / 次数”形式 - 如果是技能类动作,请优先安排“前置能力动作 + 过渡动作 + 目标动作尝试” 【E. 进阶 / 解锁逻辑区】 根据主题自动生成: - 如果是肌群训练:写“如何渐进超负荷”,例如达到次数上限后再加重量、优先保证动作标准等 - 如果是动作解锁:写“分阶段进阶路径”,例如从悬垂、肩胛引体、离心训练、弹力带辅助,到标准动作完成 【F. 替代动作区】 请给出 2–3 个替代动作,适用于以下情况: - 没有器械 - 家庭训练 - 当前能力不足 - 某些动作做不了 【G. 执行提醒区】 请给出 4–6 条简洁提醒,例如: - 动作标准优先于重量 - 不要每组都练到力竭 - 同肌群建议间隔 48–72 小时 - 疼痛不等于正常发力 - 睡眠不足时可适当减少训练量 【H. 恢复建议区】 简洁说明: - 训练后恢复重点 - 蛋白质 / 睡眠 / 恢复间隔建议 - 1 句风险提醒(如有明显疼痛应停止并评估) 【I. 视觉设计要求】 - 整体为单页中文信息图 - 竖版排版 - 风格现代、清爽、专业、健身感强 - 使用模块化卡片布局 - 重点数字(组数、次数、RIR、休息)要醒目 - 可加入简洁的人体肌群图标、哑铃、杠铃、引体向上等小图标 - 颜色保持高级、干净、有运动感 - 中文文字必须清晰、准确、易读 - 避免过多装饰,强调实用性与执行性 请最终输出为“一张完整的信息图内容”,而不是只给普通段落文字。

Example generated from “一张手绘风格的城市美食地图,以台州为主题”
GPT Image 2zh

一张手绘风格的城市美食地图,以台州为主题。画面以鸟瞰视角的手绘简化城市地图为底,标注椒江、路桥、黄岩等区域和灵江、台州湾等水系地标,不追求精确比例而是追求可爱的水彩手绘感。地图上分布着12个美食地点的精致手绘小插画:1. 椒江老粮坊的蛋清羊尾(金黄蓬松的蛋泡甜点撒着糖粉,筷子夹起拉丝)2. 临海紫阳古街的食饼筒(一个饱满的麦饼卷切开露出肉丝、蛋皮、米面等丰富馅料)3. 三门的青蟹(一只肥硕的青壳大蟹张着大钳子,旁边一小碟姜醋)4. 温岭石塘渔港的海鲜面(粗瓷大碗浓白鱼汤面铺满虾、蛏子、小黄鱼)5. 路桥的糟羹(一锅稠厚的五彩羹,芥菜、冬笋、香干、牡蛎粒粒可见)6. 玉环坎门的炊圆(三四个白胖糯米团子卧在笼屉里,旁边酱油碟滴着麻油)7. 黄岩的麦虾(陶锅里面疙瘩配蛤蜊、青菜翻滚冒泡)8. 仙居的八大碗(八只粗陶小碗围成一圈——土鸡、溪鱼、豆腐皮俱全)9. 天台的饺饼筒(几卷金黄酥脆的薄饼整齐码放,露出红烧肉和豆面馅)10. 临海的麦油脂(竹盘上摊着薄如蝉翼的饼皮卷着肉末、豆芽、鸡蛋丝)11. 温岭的嵌糕(厚实的年糕饼中间嵌着红烧肉和油条,正在铁板上滋滋作响)12. 椒江的姜汁调蛋(一只青花碗里琥珀色姜汤卧着嫩滑蛋花,撒几粒核桃碎)。每个插画约占地图5%面积,旁边用手写体标注店名和一句推荐语如“阿婆凌晨四点就起来和面”“本地人认准这口锅”。地图边缘用手绘藤蔓、杨梅枝和小海鲜(虾、蟹、贝壳)装饰形成边框。右下角有一个手绘指南针(标注“东海”方向)和图例说明。左上角标题“台州·山海食光地图”使用胖圆的手绘美术字,用杨梅和小黄鱼点缀装饰。整体画风为水彩+彩铅混合的手绘质感,颜色以杨梅红、姜黄、海蓝、翠绿为主,图片比例1:1。

Example generated from “一张采用分层蒙太奇构图的电影海报”
GPT Image 2mixed

[中文] “一张采用分层蒙太奇构图的电影海报。背景为日落时分的海滨小镇,平静的海面倒映着耀眼的日光眩光,薄雾笼罩的天空中有远处飞鸟,沿海公路旁立着电线杆剪影。左侧中景处,一位身着深灰色外套、留着深色卷发的中年男子站在混凝土海堤边,神情忧郁地低头凝视,被傍晚的阳光逆光勾勒轮廓。右侧前景主体为一张大幅特写年轻女子侧脸肖像,她望向右侧,身穿带白色条纹的深色水手校服,湿润的黑发贴在脸颊,柔和漫射光线下,一滴泪珠从她脸颊滑落。画面下方中央前景处,一只柴犬抬头朝右侧望去,红棕色毛发被温暖的轮廓光点亮。画面最底端为一条横向电影胶片,内含五幅独立矩形场景缩略图:女孩与柴犬在海滩、女孩骑车望向海面、女孩与男子坐在室内桌前、男子与女孩在海滩面对面站立、女孩拥抱柴犬的特写。画面叠加指定文字:左上角为深青绿色大号衬线字体标题《风间静语》,下方副标题为「—— 致那日的你 ——」;标题下方为小号深色衬线正文:“逝去之物,不复归来。然而,只要心灵稍稍相连,我们便能再度直面明日。” 画面右侧中部为深色衬线字体文字:“曾有一段时光,是你教会我如何生活。我永不会忘。” 左下角为大号白色文字:“10 月 31 日 周五 影院上映”。右下角为小号白色无衬线字体演职人员表:“主演:福波真子 / 桐嶋秀作 原作与剧本:柴野麻吕 导演:今仓七海 主题曲:SyVa《看得见海的地方》(Dogstar★唱片) 制作:《夕凪之尾》影视伙伴 制作公司:DABUSHIBANU-NU 发行:GOODSHIBALERS ©2026《夕凪之尾》影视伙伴”。 分段提示词: 图层索引:0 片段:“背景为日落时分的海滨小镇,平静海面倒映耀眼日光眩光,薄雾天空中有远处飞鸟,沿海公路旁有电线杆剪影。” 图层索引:1 片段:“左侧中景处,身着深灰色外套、留深色卷发的中年男子站在混凝土海堤边,神情忧郁低头,被傍晚阳光逆光照射。” 图层索引:2 片段:“右侧前景主体为大幅特写年轻女子侧脸肖像,她望向右侧,身穿带白条纹的深色水手校服,湿润黑发贴脸,柔和漫射光下一滴泪珠滑落脸颊。” 图层索引:3 片段:“画面下方中央前景处,一只柴犬抬头望向右侧,红棕色毛发被温暖轮廓光点亮。” 图层索引:4 片段:“画面最底端为横向电影胶片,内含五幅独立矩形场景缩略图:女孩与柴犬在海滩、女孩骑车望向水面、女孩与男子坐在室内桌前、男子与女孩在海滩面对面、女孩拥抱柴犬特写。” 图层索引:[5,6,7,8] 片段:“画面叠加指定文字:左上角为深青绿色大号衬线字体《风间静语》,下方副标题「—— 致那日的你 ——」;其下小号深色衬线正文:“逝去之物,不复归来。然而,只要心灵稍稍相连,我们便能再度直面明日。” 右侧中部深色衬线文字:“曾有一段时光,是你教会我如何生活。我永不会忘。” 左下角大号白色文字:“10 月 31 日 周五 影院上映”。右下角小号白色无衬线字体演职信息:“主演:福波真子 / 桐嶋秀作 原作与剧本:柴野麻吕 导演:今仓七海 主题曲:SyVa《看得见海的地方》(Dogstar★唱片) 制作:《夕凪之尾》影视伙伴 制作公司:DABUSHIBANU-NU 发行:GOODSHIBALERS ©2026《夕凪之尾》影视伙伴”。 负面提示词: “平光照明,无质感表面,对称构图,底部留白空荡,文字缺失,翻译文字,改写文字,3D 渲染,卡通风格,高对比生硬阴影,干涩头发,明亮欢快表情” [English] A cinematic movie poster utilizing a layered montage composition. In the background, a coastal town at sunset with calm ocean water reflecting a glowing sun glare, distant birds in a hazy sky, and the silhouette of utility poles along a coastal road. In the left midground, a middle-aged man with dark wavy hair in a dark grey jacket stands near a concrete sea wall, looking downward with a melancholic expression, backlit by the late afternoon sun. Dominating the right foreground is a large, closely cropped profile portrait of a young woman looking right; she wears a dark school sailor uniform with white stripes, has wet dark hair clinging to her face, and a single tear rolls down her cheek under soft, diffuse lighting. In the lower center foreground, a Shiba Inu dog looks upwards toward the right, its reddish-brown fur catching warm rim lighting. Along the very bottom edge is a horizontal film strip of five distinct rectangular scene thumbnails: a dog and girl on a beach, a girl on a bicycle looking at the water, a girl and man sitting at an indoor table, a man and girl standing facing each other on a beach, and a close-up of a girl hugging a Shiba Inu. Overlaid on the image is specific text. In the top left, large dark teal serif text reads 'The Quiet Between Winds' with a subtitle below reading '— To You, That Day —'. Below that, smaller dark serif body text reads 'What is lost will not return. And yet, when hearts connect, even just a little, we can face tomorrow again.'. On the mid-right side, dark serif text reads 'There was a time when you taught me how to live. I won't forget it.'. In the bottom left, large white text reads 'OCTOBER 31 FRI. IN THEATERS'. In the lower right corner, small white sans-serif credit text reads 'Starring: Mako Fukunami / Shusaku Kirimine Original Story & Screenplay: Shibano Maruo Director: Nanami Imakura Theme Song: SyVa \"Umi no Mieru de\" (Dogstar★RECORDS) Production: \"Yūnagi no Shippo\" Film Partners Production Company: DABUSHIBANU-NU Distribution: GOODSHIBALERS ©2026 \"Yūnagi no Shippo\" Film Partners'." segmented: - layer_index: 0 segment: "In the background, a coastal town at sunset with calm ocean water reflecting a glowing sun glare, distant birds in a hazy sky, and the silhouette of utility poles along a coastal road." - layer_index: 1 segment: "In the left midground, a middle-aged man with dark wavy hair in a dark grey jacket stands near a concrete sea wall, looking downward with a melancholic expression, backlit by the late afternoon sun." - layer_index: 2 segment: "Dominating the right foreground is a large, closely cropped profile portrait of a young woman looking right; she wears a dark school sailor uniform with white stripes, has wet dark hair clinging to her face, and a single tear rolls down her cheek under soft, diffuse lighting." - layer_index: 3 segment: "In the lower center foreground, a Shiba Inu dog looks upwards toward the right, its reddish-brown fur catching warm rim lighting." - layer_index: 4 segment: "Along the very bottom edge is a horizontal film strip of five distinct rectangular scene thumbnails: a dog and girl on a beach, a girl on a bicycle looking at the water, a girl and man sitting at an indoor table, a man and girl standing facing each other on a beach, and a close-up of a girl hugging a Shiba Inu." - layer_indices: [5, 6, 7, 8] segment: "Overlaid on the image is specific text. In the top left, large dark teal serif text reads 'The Quiet Between Winds' with a subtitle below reading '— To You, That Day —'. Below that, smaller dark serif body text reads 'What is lost will not return. And yet, when hearts connect, even just a little, we can face tomorrow again.'. On the mid-right side, dark serif text reads 'There was a time when you taught me how to live. I won't forget it.'. In the bottom left, large white text reads 'OCTOBER 31 FRI. IN THEATERS'. In the lower right corner, small white sans-serif credit text reads 'Starring: Mako Fukunami / Shusaku Kirimine Original Story & Screenplay: Shibano Maruo Director: Nanami Imakura Theme Song: SyVa \"Umi no Mieru de\" (Dogstar★RECORDS) Production: \"Yūnagi no Shippo\" Film Partners Production Company: DABUSHIBANU-NU Distribution: GOODSHIBALERS ©2026 \"Yūnagi no Shippo\" Film Partners'." negative: "flat lighting, untextured surfaces, symmetrical composition, empty bottom margin, missing text, translated text, paraphrased text, 3D render, cartoon, high-contrast harsh shadows, dry hair, bright cheerful expressions

Example generated from “三甲医院真实门诊处方笺”
GPT Image 2mixed

[中文] 一张三甲医院的门诊处方笺,医生潦草的手写字,包含真实合理的 诊断、药品名、剂量,右下角有医生签名和科室章。 [English] An outpatient prescription sheet from a Grade 3A hospital, doctor's illegible handwriting, containing realistic and reasonable diagnosis, drug names, dosages, with a doctor's signature and department stamp in the bottom right corner.

Example generated from “上海地铁站台晨光”
GPT Image 2en

A candid photograph of a young woman standing on a Shanghai metro platform during the summer morning commute, authentic daily life photography, natural candid moment. Half-body framing, shot at eye level from moderate distance — the arriving train a motion-blurred streak behind her, the yellow safety line at her feet. East Asian young woman in her early 20s. Almond-shaped eyes with natural double eyelids, slightly elongated eye corners — eyes directed downward, absorbed. Straight refined nose with a delicate bridge. Skin tone fair to light beige (NC10–NC20) — skin subsurface scattering visible under cool overhead station light, specular micro-highlights on cheekbones and nose ridge, fine foundation powder grain perceptible. Summer fresh look, no heavy coverage, lightly tinted lip. She wears a relaxed oversized white shirt dress, collar lightly open, sleeves loosely rolled above the elbow. A structured natural tan canvas tote hangs from her shoulder, a physical novel tucked under one arm — pages open at a saved page. She stands with her weight on one hip, gaze downward at the open book, entirely absorbed, occupying her own stillness inside the morning rush. Background: white and grey tiled metro wall, overhead route signage board, arriving train motion blur to one side, other commuters blurred and distant. Two or three stray hairs displaced by the train's air displacement on arrival, natural unplanned imperfection, not geometrically symmetrical. Cool overhead fluorescent lighting with a slight greenish cast, flat frontal illumination offset by a warm spill from the platform entrance behind. Absorbed in focus, eyes downward, unaware — a private moment inside a public space, the Shanghai morning. Subtle ISO 400 film grain in shadow areas, photographic noise texture not CG render smoothness. Aspect ratio 2:3. No watermark, no text overlay, not cartoon, not digitally painted, not illustration, not anime.

Example generated from “不知火舞的小红书主页”
GPT Image 2mixed

[中文] 生成不知火舞的小红书主页截图 [English] Generate a screenshot of Mai Shiranui's Xiaohongshu homepage

Example generated from “专业设计师打造角色写真集”
GPT Image 2mixed

[中文] 请用这个角色制作一本专业设计师打造的照片集。语言为日语。 根据喜好加入提示词会让它更丰富多彩… ・丰富的场景 ・信息量较多 [English] Please use this character to create a photo book crafted by a professional designer. The language should be Japanese. Adding prompts according to your preferences will make it more colorful and rich ・Rich scenes ・Large amount of information