GPT Image 2 prompts

9:16 vertical, {argument name="grid layout" default="nine-panel 3x3 collage"} of a {argument name="subject" default="Korean female idol"} portrait series. Same girl in every frame: consistent face, proportions, long slightly tousled dark hair. Natural skin texture with subtle imperfections, no retouching or smoothing. Clean idol-style minimal makeup, soft glow.

Outfit: {argument name="outfit" default="simple white shirt with short bottoms"}, youthful and styled. Same outfit across all frames.

Background: minimal indoor studio, plain wall or soft window light. Focus on the subject.

Lighting: soft diffused natural light, low contrast, airy tones, gentle film-like softness.

Camera: intimate handheld-feel portrait photography, minor grain, slight motion blur in some frames, slightly imperfect framing.

Grid layout (3 rows, 3 columns):
Top-left: standing relaxed, looking slightly away
Top-center: facing camera, mid-motion with hair or body movement
Top-right: slight side angle, soft candid gaze
Mid-left: looking slightly upward, thoughtful expression
Mid-center: close-up portrait, direct eye contact, gentle idol smile
Mid-right: body slightly turned, candid mid-motion
Bottom-left: seated or leaning casually, relaxed posture
Bottom-center: back partially turned, looking over shoulder toward camera
Bottom-right: close to frame, playful or soft expression

Aesthetic: Korean idol photobook and photocard vibe. Intimate, soft, natural everyday charm. Ultra-realistic 8K detail, subtle analog film grain, soft dreamy tone.

Reimagine {argument name="person" default="[NAME]"} as a dreamlike cloud portrait, keeping their face, expression, and defining features clearly recognizable while transforming the form into soft billowing clouds against a {argument name="background" default="bright blue sky"}.
The portrait should look like the face is materializing from or melting into the clouds, with gentle diffused natural light casting soft highlights and airy shadows for depth and realism.
Avoid sharp edges, visible skin texture, or hard details, keeping the transition symbolic and organic.
Preserve facial proportions, eyes, smile, and distinctive features through the cloud structure.
Style: dreamy, ethereal, cinematic, surreal
Lighting: volumetric sunlight, soft glow, natural
Color palette: sky blue, white, soft gradient tones
Mood: serene, uplifting, peaceful
Layer clouds naturally around and within the face for a smooth, seamless transition. Keep the background a clean blue sky with soft gradient clouds.
Ultra-realistic cloud texture, high resolution, seamless blending, no watermarks, no text.

Create a 9:16 image in the "Neon Doodle Gallery Snapshot" style. Subject: [SUBJECT]. Subject action: [SUBJECT_ACTION]. Prop or product: [PRODUCT_OR_PROP]. Location: [kashmir]. Background elements: [wooden interior ]. Main handwritten text: "[Focus mode on]". Secondary handwritten text: "[keep going]". Accent symbol: [ACCENT_SYMBOL]. Wardrobe style: [WARDROBE_STYLE]. Use a realistic candid phone-photo as the base layer. The setting should feel specific and ordinary: visible walls, art, shelves, labels, tables, lamps, posters, people, bags, shadows, grain, and imperfect handheld framing. Draw a loud digital marker layer directly on top of the photo. Wrap the main subject with a thick hot-pink contour and a cyan offset glow. Add yellow-orange monster spikes, horns, rays, fins, or sunburst shapes around the silhouette. Scatter rough hand-drawn symbols around the frame: stars, paw prints, spiderweb corners, halos, abstract eyes, plants, flowers, scribble underlines, tally marks, arrows, hearts, and sticker-like blobs. Place rough uppercase handwritten marker text in open areas, using white, yellow, or lime green. The text should feel funny, personal, distracted, and student-made. Preserve the contrast between a real candid photo and chaotic handmade doodles. Avoid watermarks, usernames, platform logos, creator IDs, app marks, QR codes, clean vector-only illustration, fully illustrated backgrounds, polished ad layout, luxury editorial styling, perfect typography, empty sterile locations, identifiable celebrities, and tiny unreadable text. --- VARIABLES --- [ACCENT_SYMBOL] — star, paw print, spiderweb, halo, abstract eye, plant, flower, underline, arrow, tally mark, or scribble [BACKGROUND_ELEMENTS] — real photo details such as wall art, labels, shelves, posters, tables, lamps, signage, crowds, fabric, shadows, and phone-camera grain [LOCATION] — art gallery, campus hallway, library, studio critique room, classroom, night market, cafe, bookstore, museum, or city wall [MAIN_TEXT] — large hand-drawn caption or emotional headline [PRODUCT_OR_PROP] — notebook, tote bag, coffee, phone, headphones, sketchbook, jacket, snack, poster, camera, book, or exhibition card [SECONDARY_TEXT] — small handwritten notes, repeated words, short joke, date-like label, or study annotation [SUBJECT] — main person, group, student, artist, friend, commuter, shopper, or quiet candid figure [SUBJECT_ACTION] — looking at art, studying, walking, waiting, browsing, reacting, hiding, laughing, or holding a prop [WARDROBE_STYLE] — casual student streetwear, oversized shirt, hoodie, tote bag, loose trousers, jacket, headphones, sneakers, or art-school layers --- NEGATIVE PROMPT --- watermark, username, creator ID, platform logo, app mark, QR code, clean vector poster, fully illustrated scene, polished advertising layout, luxury editorial shoot, sterile studio, perfect typography, perfect sticker sheet, subtle doodles, empty background, corporate mascot, identifiable celebrity, real public figure, tiny unreadable text

A high-quality {argument name="style" default="Korean-style"} commercial photograph, in the style of a premium sports magazine cover.
[Composition and Pose]
Half-body portrait (from mid-torso upward), slightly angled frontal view. The {argument name="subject" default="male"} subject stands confidently, looking slightly downward toward the camera with a strong, {argument name="expression" default="determined"} expression.

A beautiful {argument name="ethnicity" default="young Asian woman"} with long straight black hair, soft doe eyes, natural beauty makeup with subtle blush and glossy lips, wearing a {argument name="clothing" default="white baseball cap, white fitted crop top, and white wide-leg casual trousers"}, layered delicate silver necklaces, standing in a softly blurred night event venue with {argument name="lighting" default="warm ambient lighting"}, candid portrait style, elegant and effortless street style, shallow depth of field, cinematic lighting, 8K, hyperrealistic --ar 2:3 --style raw --q 2

SYSTEM: Render the input as a luxury historical civilization stratigraphy poster. Do not hardcode dates unless inevitable. Infer the territorial expansion phases, economic foundation pillars, military innovation cycles, cultural assimilation patterns, and the geological layers of decline.  SEMANTIC SOLVE: EMPIRE_AUTOPSY =    (INFER(territorial_architecture FROM founding_core + expansion_vectors + frontier_fortifications + trade_route_control + vassal_networks) ::5) +    (INFER(economic_pillars FROM agricultural_base + mineral_resources + taxation_system + currency_standard + labor_organization) ::4) +    (INFER(military_innovation FROM weapon_technology + tactical_doctrine + logistics_chain + fortification_engineering + naval_capacity) ::4) +    (INFER(decline_mechanics FROM succession_crises + economic_inflation + frontier_pressure + internal_rebellion + environmental_stress) ::3) -    (generic timeline infographics + cartoon maps + cluttered textbook layouts + stock-photo ruins + cheap educational posters) ::-4  COMPOSITION: One central empire visualized as a multi-layered archaeological cross-section. The bottom layer shows the founding settlement, middle layers reveal territorial expansion through colored strata, and the top layer displays the fragmentation pattern. Surrounding the core are floating callouts showing key battles, economic indicators, and dynastic transitions mapped as physical artifacts. Use callout lines like an archaeological dig site report crossed with a military campaign map.  STYLE DNA: Ancient Roman Tabula Peutingeriana ::0.30 archaeological stratigraphy diagram ::0.25 vintage military campaign map ::0.20 museum exhibit infographic ::0.15 aged parchment with stain texture ::0.10  OUTPUT: Warm sepia or deep terracotta background luxury history poster, elegant classical serif + handwritten annotation typography, restrained callouts, hyper-realistic map textures, refined aging effects, premium negative space.  NEGATIVE: no holograms, no glowing elements, no VR/AR overlays, no modern digital maps, no cartoon illustrations, no cluttered timelines, no stock-photo emperors, no watermark, no anachronistic elements.

Use the uploaded image as the ONLY facial identity reference. Absolute face preservation mode. Strict face lock. Maintain 95–100% facial similarity and preserve the exact eyes, eyelids, eyebrows, nose shape, nostrils, lips, jawline, cheekbones, forehead, skin tone, ethnicity, age appearance, facial proportions, beauty marks, and natural skin texture from the uploaded face reference. Do not redesign, beautify, alter facial structure, blend identities, or generate a different person.

Recreate the exact scene shown in the reference image: a candid urban subway platform environment with a train on the left side, tiled station walls, cinematic shallow depth of field, natural station lighting, soft background blur, realistic motion atmosphere. Subject looking back over her shoulder toward the camera with a warm genuine smile. {argument name="hair color" default="Long dark hair"} visible from beneath a light pink {argument name="headwear" default="New York Yankees baseball cap"}, small gold hoop earrings, oversized black outfit, black shoulder bag. Natural candid street-photography aesthetic, authentic lifestyle moment, realistic skin texture, ultra-photorealistic DSLR quality, shallow depth of field, soft bokeh, cinematic color grading, highly detailed, 8K resolution.

Important: Keep everything exactly the same as the original image (pose, camera angle, subway platform, cap, clothing, lighting, composition, background, expression, and environment). Replace only the face with the uploaded face reference while maintaining perfect identity preservation and realistic integration

PLAYER={argument name="player name" default="[player name]"} Ultra-realistic sports photography, documentary realism, premium commercial campaign style, immersive 4:5 vertical composition. The 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.

Daytime {argument name="name" default="Yamada"} cashier version, 3:4 vertical image. Create a realistic live-action portrait of an adult young Japanese woman, around 24 years old. She has fair skin, soft delicate facial features, a gentle oval face, calm dark eyes, natural light makeup, reddish hair, straight blunt bangs, and long side locks framing both sides of her face. Her expression is gentle, polite, slightly shy, and quietly mature, like a reliable supermarket cashier with a warm customer-service smile.

Scene: daytime inside a Japanese supermarket checkout area. She is standing behind or beside the checkout counter, facing the camera, with a polite gentle smile. The background has blurred product shelves, checkout counter details, and clean supermarket lighting.

Outfit: {argument name="outfit" default="Japanese supermarket employee uniform"}. Deep red headscarf covering the back of her hair while still showing her straight bangs and red side locks, pale green or beige striped short-sleeve work shirt, deep red apron, black flared work pants. Add a small rectangular employee name badge pinned on the upper chest or apron with readable Japanese text “{argument name="badge text" default="山田"}”. The badge should be realistic, small, and clear. The only readable text in the image should be “{argument name="badge text" default="山田"}”.

Style: realistic live-action Japanese drama still, 3:4 vertical portrait, waist-up or three-quarter body, natural indoor fluorescent supermarket lighting, shallow depth of field, muted realistic colors, natural skin texture, 35mm lens look, high detail.

{argument name="scene" default="The sunset slowly sinks toward the sea level, the glow spreads over the sky, dyeing the floating clouds into gold and red brocade, the vast sea surface floats with shimmering light, every surging wave holds up the jumping afterglow."} The atmosphere is close to a real beach. Capture a natural dynamic pose, {argument name="action" default="one hand lightly covering part of the face, playful wink, with a light smile"}, the expression is smart and lively. Adopt high-end fashion magazine shooting techniques, natural and soft environmental side lighting, truly restore skin texture and pore details, hair layers are natural and have a sense of drape; {argument name="clothing" default="jumpsuit"} shows natural folds, stress deformation and fabric luster, fashion documentary photography, ultra-realistic picture quality, bright colors, clean composition, prominent subject, 9:16 vertical full-body screen

Generate a medium information density infographic poster suitable for social media sharing, with the theme of "{argument name="book title" default="Book Title"}".

[Overall Style]
{argument name="style" default="Modern minimalist + light retro hand-drawn style"}, background using soft light colors ({argument name="background color" default="cream #FAF7F2"} or light gray blue #F5F8FA), lots of white space, quiet, informative, and readable for a long time.

[Core Requirements]
First judge the main type of the book (cognitive growth, business wealth, psychological emotion, novel story), and then naturally organize the content and focus accordingly. Do not use fixed template titles, do not force the structure, act like a person who truly understands the book sharing it.

[Content Structure Suggestions (Dynamic Combination)]
- Opening: Book Title + Author + an insightful sentence (human-like)
- Core Understanding: What is the core the book truly wants to express? What realistic problem does it respond to?
- Most valuable parts (key modules): Choose models, levers, emotional mechanisms, character conflicts, etc., according to the book type
- Key structure or framework (visual core)
- Impressive examples/clips (with a sense of imagery)
- Memorable quote area (white space treatment)
- Actionable changes/suggestions (3-5 items, specific and executable)

[Visual Requirements]
- Adaptive grid layout (2-3 columns or free distribution)
- White cards + rounded corners + slight shadows
- Each module is accompanied by simple hand-drawn style illustrations
- Clear title hierarchy, appropriate emphasis on important information
- Restrained and elegant color scheme (main color selected according to book type: tech blue, growth green, psychological warm brown, etc.)

[Language Style]
Like a person talking about a book: with understanding, screening, and a sense of realization. Avoid templated expressions like "This book talks about..." or "The core idea is...".

Generate an information-clear, high-end design book summary poster suitable for sharing on Xiaohongshu/Weibo/Twitter.

{argument name="character" default="An adult female cheerleader"} standing {argument name="background" default="near the sidelines of the World Cup stadium"}, wearing {argument name="clothing" default="neat and designed cheerleader support clothes"}, holding a football and a small flag, the breeze blowing hair and corners of clothes, the posture is natural and full of motion, the background is bright stadium lawn, giant stadium, grandstand lights and cheering audience, sunlight or stadium ceiling lights illuminate the character's silhouette, the overall picture is clean and powerful, with motion brand advertising texture, hot-blooded, professional, youthful, and intense international competition atmosphere

An {argument name="person" default="adult female soccer babe"} wearing a {argument name="outfit" default="modified slim-fit jersey in national team colors"}. The top is slightly short, paired with a high-waisted sports skirt. The look is fashionable, fresh, and possesses moderate feminine charm. She has a warm smile and a natural, relaxed posture while holding a soccer ball. The background is a blurred World Cup stadium with cheering spectators. The lighting is bright and transparent. The image has both a sense of sports and fashion photography, high-definition and realistic, youthful, eye-catching, and not overexposed.

Cinematic corporate portrait, ultra-realistic, same person as in the reference image. Preserve accurate facial features, skin tone, and identity characteristics. Style: Classic professional corporate portrait. The subject is wearing a {argument name="outfit" default="dark custom suit, crisp white shirt, and understated tie"}. Soft directional studio lighting with gentle shadows to add depth. Minimalist background with blurred {argument name="background colors" default="neutral tones (dark gray or beige gradient)"}. Shot using a full-frame camera, {argument name="lens" default="85mm lens"}, shallow depth of field, eyes sharply focused, natural skin texture, and high dynamic range. Confident, calm, and approachable expression. Ultra-detailed, 8K quality, realistic color grading.

An {argument name="person" default="adult female soccer babe"} standing in front of a World Cup stadium stand, with a confident and brilliant smile, waving a {argument name="accessory" default="support scarf in national team colors"}, wearing fashionable supporting clothes with soccer elements. The overall look is youthful, healthy, and full of vitality. Her long hair is gently raised in the wind, and her posture is natural and powerful. The background is a brightly lit modern soccer stadium, spectator stands, and green turf. There are streamers and a slight smoky atmosphere in the air. The image is full of the heated atmosphere of a major international football event, commercial sports poster style, high-definition realistic photography quality, vivid colors, and strong dynamism.

An {argument name="person" default="adult female World Cup soccer babe"} excitedly raising her hands during a victory celebration, wearing form-fitting but not overly revealing fashionable support clothing in {argument name="colors" default="national team representative colors"}. A short jacket paired with a slim sports skirt makes her look passionate, healthy, and infectious. Her hair is flying, her expression is excited and real. The background features falling confetti, a score screen, stadium lights, and a cheering crowd. The overall image looks like a post-match victory poster, with heated emotions, high-definition realistic photography.

A high-fashion commercial poster of an adult female soccer fan wearing a haute couture cheerleading outfit infused with soccer elements, such as {argument name="outfit" default="a national team color-coded short jacket, slim-fit sports skirt, or a sleek modified jersey design"}. She has a confident posture and a determined gaze, set against an abstract background of {argument name="background elements" default="World Cup stadium lights, goal net structures, giant soccer balls, and a branded stage set"}. The image combines the aesthetic of a fashion magazine with a sporting event promotion, featuring a mix of warm and cool lighting, high-quality textures, and a high-definition realistic style.

I would like to see a {argument name="style" default="hand-drawn style"} illustration of a {argument name="subject" default="cat-eared girl in a white dress at an old street corner"} that is extremely rough and could be interpreted as an abstract painting. Can you generate a sharp digital illustration that looks like an oil painting in the style of Midjourney?

I would like to see a hand-drawn style illustration of {argument name="subject" default="a cat-eared girl and a gentle cat"} wearing a {argument name="clothing" default="white dress"}, with extremely rough brushstrokes that could be seen as an abstract painting. Can you generate a {argument name="style" default="digital illustration that looks like a bold Midjourney-style oil painting"}?

You are a specialized pixel art director converting anime, game, and manga characters into MapleStory-style player avatar sprites.

[Tasks]
1. Analyze the character's physical characteristics:
- Hairstyle
- Hair color
- Eye color
- Representative outfit
- Accessories
- Weapon
- Symbolic elements
- Representative colors

2. Reinterpret these features into the MapleStory player character structure while maintaining the above characteristics.

3. Ensure it looks like an actual MapleStory player character sprite rather than generic pixel art.
────────────────────
[Character Structure Specifications]
- Head + Hair: 60-65% of the total
- Torso: 15-18%
- Legs: 20-25%
- 2 to 2.3 heads tall
- 35-45 degree quarter-view
- Angle facing forward so both eyes are visible
- Mostly flat 2D sprite face
- Eyes placed at the bottom of the face
- Wide spacing between eyes
- Wide, flat eyes
- Large pupils
- Pupil highlights included
- Prohibition of nose representation
- Very small mouth
- Soft cheeks
- Prohibition of neck representation (head directly connected to torso)
- Outfit begins immediately below the chin

Hairstyle:
- Core of the character's identity
- Volume larger than the head itself
- Silhouette priority
- Minimal expression of hair strands
- Highlight at the top

Body:
- Very small torso
- Short arms
- Short legs
- Simple cylindrical limbs
- Minimal joint expression

Equipment Layer Structure:
Hair / Hat / Face Accessory / Top / Bottom
Shoes / Cape / Weapon

Pose:
- Standing pose (Character selection screen style)
- No jumping
- No combat poses
- No action effects
────────────────────
[Pixel Rendering Specifications]
Goal: Atmosphere of an enlarged 32-64px game sprite

- Low-resolution game sprite
- Blocky pixel structure
- Grid-aligned pixel art
- Visible pixel blocks
- Clean 1px outline
- Limited color palette
- Flat color shading
- Minimal shading
- No dithering
- No smooth gradients
- No airbrush effects

[Forbidden Styles]
Pokemon-style / Terraria-style / Stardew Valley-style
Realistic / 3D rendering / Illustration-style
Anime illustration / Digital painting
HD pixel art / Smooth gradients
────────────────────
[Output Conditions]
- Canvas size: 1080x1080
- Background: Pure white (#FFFFFF)
- Single character only
- Full body display
- Centered placement
- Occupies approximately 75% of canvas height
- No text, logos, or watermarks
- No background objects
- No ground
────────────────────
Generate the character as an actual MapleStory player avatar sprite. It should look like the result of converting the attached image into a MapleStory style, not just a generic pixel art character.

Ultra photorealistic portrait of a young woman with {argument name="hair style" default="long straight dark brown hair"}, sun-kissed glowing skin, seated at an outdoor cafe table. She's wearing a {argument name="clothing" default="white vintage Swiss dot corset mini dress"} with a sweetheart neckline, ruffled cap sleeves, front lace-up ribbon detailing, fitted bodice, and slightly sheer ruffled hem. Hands raised playfully covering her eyes, head tilted back laughing, red manicured nails, thin bracelet and ring on left hand. Setting: {argument name="location" default="luxury outdoor hotel terrace at Hotel Florence"}, historic yellow building with "HOTEL FLORENCE" signage, lush green mountains in the background, cloudy blue sky, vintage globe street lamps, round glass-top table with two white ceramic coffee cups and a paperback book. Coquette cottagecore soft feminine luxury aesthetic. Natural bright afternoon sunlight, high contrast, sharp shadows on the table, backlighting creates a hair halo effect, warm vibrant color grading. DSLR 85mm portrait lens, f/2.8 shallow depth of field, 1/500s shutter, ISO 100, 8k RAW photo.

Use 100% of the face and body from my reference image. Create a cinematic winter scene with a side face close-up view, inspired by a Minecraft death screen. The person is lying face up in the snow, wearing {argument name="clothing" default="casual winter clothing"}, in a realistic pose, with cool atmospheric lighting, slightly desaturated tones, and visible footprints in the snow. Surround the person with floating or placed pixelated objects in the Minecraft style, including a {argument name="weapon" default="diamond sword"}, bow, pickaxe, emerald, steak, Eye of Ender, music disc, and a mob of spiders. Add a realistic Minecraft "You Died!" overlay on top with pixel font, including "Score: 0," the reticle symbol, and two retro UI buttons labeled "Respawn" and "Title Screen." Naturally blend the realistic photography with pixel-art game elements. Soft shadows, melancholic winter aesthetic, ultra-detailed, vertical composition, DSLR realism, subtle depth of field, high contrast, 9:16 aspect ratio.

Negative Prompt:

"Low quality, blurry face, distorted body, extra limbs, incorrect facial features, cartoonish body, unrealistic"

Create a viral CapCut-style {argument name="nationality" default="South korea"} graffiti image from the uploaded person. Keep the face consistent. Add {argument name="jersey" default="South korea jersey"}, full-body pose, giant hand-painted mural portrait in the background, {argument name="logo" default="South korea logo"}, {argument name="year text" default="South korea 2026 text"}, yellow and green football colors, concrete wall, clean poster composition, realistic 闪 photo foreground, illustrated thelifeafptfiti b--'ground,TikTok

Create a {argument name="aspect" default="vertical"} cinematic tutorial comparison poster showing the same {argument name="subject" default="urban street scene"} repeated in five horizontal panels, each panel demonstrating a different film genre color grading style. The scene is a busy small city street at sunset, with cars, motorcycles, cyclists, shop buildings, electric poles, tangled overhead wires, and warm evening sky. Keep the camera angle exactly the same in every panel, like a fixed street photography shot.

Each horizontal strip should have a different cinematic mood and color tone:

Panel 1: SUSPENSE - cool blue shadows, dark contrast, slightly mysterious atmosphere, tense and moody lighting.
Panel 2: ACTION - warm orange sunset tone, high contrast, energetic cinematic look, dramatic highlights.
Panel 3: CRIME - cold teal blue grading, darker exposure, gritty urban feeling, realistic shadows.
Panel 4: ROMANCE - soft golden warm tone, gentle contrast, nostalgic sunset mood, dreamy atmosphere.
Panel 5: DRAMA - muted purple blue evening tone, emotional cinematic lighting, soft shadows and serious mood.
Add bold white uppercase text on the right side of each panel: SUSPENSE, ACTION, CRIME, ROMANCE, DRAMA. Use clean modern sans serif typography with slight glow or subtle shadow. Separate each panel with thin white horizontal borders. Make it look like a professional filmmaking color grading tutorial thumbnail, cinematic, realistic, high detail, urban street photography, 9:16 vertical composition.
Negative Prompt: cartoon, anime, painting, low quality, blurry, distorted buildings, extra text, wrong spelling, messy typography, overexposed, unrealistic colors, duplicated cars, warped street, bad perspective, random objects, low resolution, watermark