GPT Image 2 prompts

Based on the clear single-person portrait reference image I uploaded, create a set of 8 'freediving underwater dream art portraits', all in 3:4 vertical composition, as independent images. 1 | Character Consistency: Must strictly preserve the original facial features, face shape, eyes, nose, lips, skin color, hairline, age, and overall identity of the reference person. No face swapping, no turning into a stranger, no 'influencer face', no anime style, no plastic-looking face. Skin should retain realistic texture and a high-end photographic feel; body proportions must be realistic and natural with slender limbs and correct lines for neck, back, waist, hips, legs, and feet. 2 | Overall Style: High-budget, realistic, professional, dreamy, ethereal, and breathtaking freediving underwater art masterpiece. Combines a high-fashion editorial feel with documentary-level ocean realism. Emphasize freediving professionalism, body control, buoyancy, and the coexistence of strength and elegance. Not an ordinary tourist photo, not cheap mermaid cosplay, not low-quality boudoir style. 3 | Professional Freediving Gear: Must feature professional freediving equipment, including: low-volume freediving mask, long carbon fiber fins, slim weight belt, dive computer, and minimal use of a nose clip if needed. No oxygen tanks, BCDs, or scuba gear; the set must be clearly identified as freediving. 4 | Clothing and Styling: High-end backless and high-cut freediving fashion styling, balancing professionalism, fashion, and luxury. Includes: backless one-piece high-cut swimsuits, fitted fashion dive wear, in colors like obsidian black, pearl white, sea salt gray, mist blue, deep sea blue, pale gold, or cool gray. Minimal use of floating veils or ribbons is allowed but must not be distracting. Must look like international fashion underwater photography, not cheap swimwear or e-commerce style. 5 | Scene and Marine Life: Clear, realistic, and magnificent underwater world. Water in deep blue, turquoise, or cyan with realistic sunbeams, particles, light refraction, and spatial depth. Backgrounds include sand patterns, reefs, coral, trenches, and open deep sea. Marine life should be realistic and elegant, focusing on: dancing with sharks (reef, lemon, or nurse sharks), bait balls/fish vortexes, and optionally sea turtles, rays, or tropical fish. No deformed creatures or aquarium-like fake settings. 6 | Action and Camera Requirements: Professional freediving movements: diving, hovering, gliding, rotating, stretching, and ascending. Varied camera angles and shots: top-down, low-angle, side-view, front, close-up, medium, wide, ultra-wide, backlit, and sub-surface perspectives. 7 | Mood and Temperament: Quiet, focused, brave, dreamy, elegant, and mysterious. Expressions should be calm and immersive, not necessarily looking at the camera. 8 | Negative Requirements: No AI look, no plastic skin, no extra limbs, no deformed hands or feet, no incorrect body structures, no wrong fin structures, no scuba gear, no low-quality filters, no watermarks. 01 | Sub-surface descent opening; 02 | Lateral glide through coral; 03 | Dancing with sharks (wide shot); 04 | Dancing with sharks (medium shot relationship); 05 | Fish storm vortex (top-down); 06 | Close-up backlit portrait; 07 | Ascending toward the light; 08 | Deep sea dream finale (wide master shot).

Use my uploaded face image as the primary identity reference. Preserve my exact facial identity, face shape, hairstyle, hair texture, beard style, skin tone, eye shape, eyebrows, and all facial details with maximum accuracy. Do not change my face.

Create a cinematic fantasy editorial portrait of me standing inside an {argument name="location" default="ancient abandoned library"}. I am wearing an elegant loose {argument name="clothing" default="white knit sweater and white pants"}, barefoot, with a calm mysterious expression. My body is slightly turned while I look upward, reaching one hand toward floating books. Hundreds of old books and pages are suspended in the air around me as if frozen by magic.

Warm golden sunlight streams through large old windows behind me, creating dramatic volumetric light rays, dust particles, and a dreamy atmosphere. The background is filled with towering bookshelves, aged wooden floors, scattered papers, and floating open books. Ultra-realistic photography style, cinematic lighting, shallow depth of field, high detail, luxury magazine aesthetic, 8K resolution.

Keep my face as the main focus and make the scene look like a professional movie still.

Black-and-white fashion casting contact sheet of a woman with {argument name="hair style" default="short wet curls"} glistening under studio lights, arranged in a clean 2x2 grid of four close portrait frames against a {argument name="background" default="seamless black"} backdrop, wearing a {argument name="outfit" default="silk robe-inspired jacket"} and no visible jewelry. Each frame shows a different expression and angle: haunting direct gaze, relaxed smile, introspective downward look, and profile portrait with dramatic shadow sculpting across the face. Soft studio lighting, crisp monochrome contrast, natural skin texture, visible facial details, clean plain backdrop, subtle film grain, high-end editorial test shoot, minimal styling, intimate camera distance, professional portrait photography, aspect ratio 4:5.

Use the uploaded image as the exact facial identity reference. Preserve facial structure, jawline, hairstyle, skin texture, facial proportions, expression realism, and overall likeness with maximum identity accuracy.

A breathtaking {argument name="setting" default="celestial observatory"} floating high above an endless ocean of clouds under a star-filled night sky. A handsome rugged explorer sits confidently in an antique wooden chair, holding a steaming metal coffee cup, his face perfectly matching the uploaded reference image. He wears a luxurious {argument name="outerwear" default="dark leather trench coat"} layered over weathered explorer clothing, creating a timeless adventurer aesthetic.

Surrounding him is an extraordinary collection of massive brass telescopes, antique astrolabes, celestial navigation instruments, illuminated star charts, ancient books, mechanical clocks, glowing lanterns, intricate clockwork devices, and vintage scientific equipment. Several adorable {argument name="animals" default="tabby cats"} wearing miniature aviator goggles and explorer gear roam the observatory, curiously inspecting maps and instruments.

In the background, an immense floating city of interconnected observatory towers rises above the clouds, connected by elegant bridges and illuminated by thousands of warm golden lights. A giant crescent moon hangs in the sky while countless stars sparkle across the heavens. Dramatic cloud formations surround the structures, creating a majestic dreamlike atmosphere.

Ultra-realistic cinematic photography, masterpiece composition, fantasy realism, hyper-detailed environment, atmospheric storytelling, volumetric moonlight, warm lantern glow, realistic reflections, polished brass surfaces, detailed leather textures, intricate mechanical craftsmanship, cinematic depth, floating dust particles, shallow depth of field, creamy bokeh, HDR lighting, global illumination, ray-traced reflections, professional color grading, dramatic perspective, luxury fantasy aesthetic, award-winning photography, ultra-sharp focus, photorealistic, 8K, breathtaking visual narrative.

Keywords: floating observatory, sky kingdom, celestial architecture, brass telescope, explorer aesthetic, fantasy realism, cinematic atmosphere, cloud city, steampunk luxury, warm golden lights, volumetric lighting, ultra-detailed, masterpiece, best quality, photorealistic, epic scale, environmental storytelling.

Negative Prompt: blurry, low quality, cartoon, anime, oversaturated colors, distorted face, extra limbs, duplicate objects, unrealistic anatomy, bad hands, low detail, watermark, text, logo, cropped subject, out of frame, plastic skin, CGI look.

Aspect Ratio: 9:16

Style Strength: Ultra Realistic • Cinematic • Fantasy Realism • Editorial Photography • Award-Winning Concept Art • Hyper Detailed 8K

An ultra-realistic, cinematic portrait of a {argument name="subject" default="beautiful young woman"} riding a {argument name="bicycle style" default="vintage black bicycle"} through an autumn park. She is wearing an elegant, {argument name="outfit" default="dark brown Victorian-style dress"} with detailed lace trim, a high ruffle collar, a black bow tie, and a matching wide-brimmed vintage hat. Her long dark hair is caught in a gentle breeze, flowing dynamically behind her. The background features brilliant, golden-yellow ginkgo trees with autumn leaves softly falling around her, creating a warm, sun-drenched atmosphere. Captured on a 85mm lens, shallow depth of field, soft natural golden hour lighting, DSLR quality, 8k resolution, highly detailed textures, masterpiece.

Using the uploaded image and without changing face create a stunning cinematic lifestyle portrait of the {argument name="subject" default="woman"} seated confidently at sunset in an {argument name="setting" default="urban desert-inspired setting"}, preserving her exact facial features, expression, hairstyle, body proportions, and identity with absolute accuracy. Transform the scene into a premium fashion editorial where she sits on a sleek {argument name="car detail" default="custom sports car wheel"} beside a high-end modified vehicle, surrounded by warm golden-hour light and dramatic cotton-candy skies painted in shades of peach, orange, and violet. Enhance her colorful retro-inspired jacket, denim jeans, and sneakers with rich textures and realistic fabric detail while maintaining the original styling. Add subtle wind movement through her hair, glowing rim light outlining her silhouette, and beautiful bokeh from distant city lights and palm trees. Capture the scene with luxury automotive photography aesthetics, shallow depth of field, ultra-realistic skin texture, cinematic color grading, HDR lighting, razor-sharp focus on the subject, and an atmosphere of ambition, confidence, and youthful freedom. Photorealistic 8K resolution, Vogue x Porsche campaign quality, award-winning fashion photography, dynamic composition, emotional storytelling, and a powerful modern dreamer aesthetic.

Create a high-end Korean fashion lookbook featuring the same {argument name="subject" default="Japanese woman"} across all six looks, with strong identity consistency throughout the entire image. Her facial features, age, ethnicity, hairstyle, body proportions, and overall presence should remain unchanged in every panel, making it unmistakably the same person wearing different outfits.\n\nArrange the composition as a clean and elegant 2×3 grid with six equal-sized full-body panels. Maintain consistent model scale, clear spacing between panels, and balanced editorial symmetry. At the top center, include refined luxury serif typography reading: '{argument name="title" default="LOOKBOOK — The Edit • Vol. 01"}'.\n\nEach panel should showcase a different outfit and natural fashion pose:\n\n1. Beige oversized blazer, white top, wide-leg trousers, and heels.\n2. Cropped leather jacket, fitted top, cargo pants, and luxury sneakers.\n3. Black fitted evening dress with sophisticated accessories.\n4. White oversized shirt paired with denim shorts.\n5. Cream knit sweater with a pleated midi skirt.\n6. Monochrome tailored jacket and matching skirt set.\n\nGive each look a unique standing pose and subtle change in expression while maintaining the same model identity and luxury campaign energy. The body language should feel confident, polished, and suitable for a premium department-store fashion catalog.\n\nUse a seamless warm ivory-beige cyclorama studio background with no props or furniture. Lighting should be soft, diffused, and inspired by Vogue Korea editorials, creating realistic shadows, natural skin texture, visible pores, and beautifully rendered fabric details. The final image should feel like ultra-premium commercial fashion photography with HDR quality, photorealistic detail, and 8K-level sharpness.\n\nPrioritize strict identity preservation, full-body visibility in every panel, clear panel separation, consistent proportions, and a sophisticated luxury presentation.\n\nAvoid: identity changes, different faces, face swapping, cropped figures, overlapping subjects, extra limbs, anatomical errors, cluttered backgrounds, props, furniture, logos, watermarks, anime styling, CGI, illustrations, overly smooth plastic skin, blur, or low-resolution output.

Create a realistic full-body portrait of a {argument name="age" default="25-year-old"} adult {argument name="character traits" default="East Asian woman"} with a cute and soft appearance, standing in front of a light wood door indoors. She has fair skin, a sweet rounded face, large brown eyes, soft glossy lips, rosy shy cheeks, and a gentle, bashful smile. She has {argument name="hair color" default="light greyish-brown"} shoulder-length hair with soft curls at the ends, straight bangs, and loose strands framing her face. She is wearing cute heart-shaped earrings. Body Type: Soft plus-size/curvy figure with natural realistic proportions, a rounded soft shape, wide hips, thick thighs, and a warm, approachable look. Make her look healthy, cute, natural, and real, not exaggerated. Pose: Standing facing the camera, full body visible from head to toe. Her knees are slightly turned inward in a shy, cute pose. One hand lightly holds the edge of her cardigan, and the other is naturally placed near her shorts pocket. Shoulders are slightly relaxed, with bashful yet modest body language. Clothing: A white fitted camisole fully covering the chest and waist, a light grey cardigan, light grey casual shorts with drawstrings, and black knee-high socks. The clothes are opaque, comfortable, casual, and realistic, with natural folds, fabric stretch, and soft indentations around the waist and thighs. Scene: A minimalist apartment entrance or bedroom doorway, with a light wood door behind her, clean floors, soft indoor lighting, warm neutral tones, and a quiet, private home atmosphere. Full-body vertical composition, 9:16 aspect ratio. Photography Style: Realistic smartphone photo, natural skin texture, soft indoor light, slightly handheld perspective, high detail, a lifelike cozy home portrait, realistic hair strands, authentic fabric folds, and natural body proportions. Avoid anime style, illustrations, CGI looks, underage appearance, thin body types, exaggerated fantasy proportions, explicit nudity, transparent clothing, overly sexual poses, distorted hands, extra fingers, bad anatomy, plastic skin, over-smoothed faces, and low resolution.

{argument name="title" default="The Escaped Boy and a Cup in the Desert"}

A boy who escaped from a white ship is taking a short breath at a {argument name="location" default="desert cafe"}.

The shopkeeper asks nothing and simply brews a cup quietly.

"I'm the one who can use it best..."

Before that murmur, the cat manager seems to have found a way to become friends with the {argument name="robot" default="round robot"}.

[Information Input Field]

Character Name: [{argument name="character name" default=" "}] (Required)

Additional Notes: [{argument name="additional notes" default="optional"}] (Enter if there is anything you want to tell the AI)

-----------------

[Image Generation Instructions]

Based on the character in attached image 1, dress them using the devil transformation set in attached image 2.

Read the character's appearance, atmosphere, age, expression, equipment, and personality traits from the input character name and attached image 1. Automatically devise a background, situation, posing, expression, composition, lighting, and a short phrase within the image.

Maintain the original character's hairstyle, hair color, eye color, physique, facial features, age, atmosphere, and character personality.

Reflect the design elements of the devil transformation set so they blend naturally with the original character. It should not just be layering a costume, but an integrated design showing the original character transformed into a devil.

Pay particular attention to the placement of horns and tails, depicting them to look anatomically and visually natural. Horns should sprout naturally from the head and not look like they are roughly placed on top of the hair. The base of the horns should be at the top of the head, near the forehead, or around the temples in a reasonable position, with balanced symmetry, orientation, length, thickness, and a proper sense of connection. The tail should sprout naturally from the lower back as an extension of the tailbone, and not from unnatural positions like the abdomen, chest, legs, or arms. Adjust the horns and tail to look natural according to the pose and costume, ensuring no awkward connections, floating bases, nonsensical numbers, unnatural curves, or strange growth patterns.

Construct a new background and situation as a scene that strongly conveys the character's charm, without relying on standard compositions. Make the character's expression, costume, pose, gaze, and gestures the main focus, rather than just flashy backgrounds.

Treat the expression as a high-priority item. Do not use expressionless, straight, flatly calm, or standard 'cool' faces. Create an impressive expression that looks like multiple emotions are layered, rather than a single simple emotion.

Add variety to the expression using eyebrows, eye direction, pupil highlights, corners of the mouth, cheeks, face angle, and neck tilt. Make it a natural expression that is slightly broken rather than perfectly symmetrical. Create an expression that makes viewers want to imagine the character's inner thoughts.

For posing, prioritize showing the entire costume while creating a dynamic composition that conveys the character's personality and the atmosphere after transformation. Do not use a pose where the character is just standing. Emphasize character traits using weapons, wings, horns, tails, chains, accessories, hand gestures, and gaze direction.

Utilize the eccentricity of the devil transformation set but do not crush the original character's individuality. Ensure the original character does not look like a different person.

In the top-left or bottom-right of the screen, automatically generate a title with decorations, with the first line being the '[Character Name]' and the second line being the title/nickname gained this time, placed at the same font size. The text should be a size that does not interfere with the image.

Only if there is content in the additional notes of the information field, reflect that content in the generated image.

Style:
- High-quality 2D animation
- Language: Japanese
- Genre: Dark fantasy leaning, but do not make the background too dark.
- Delicate costume design
- Character-focused
- Impressive expressions
- Dramatic lighting

Aspect Ratio: {argument name="aspect ratio" default="3:4"}

I transform the {argument name="subject" default="people in my photographs"} into adorable, {argument name="style" default="fashion-inspired caricature sketches"}. They feature exaggerated, cute body language, adorable proportions, expressive hands, dramatic expressions, awkward yet graceful postures, satirical fashion sense, messy ink lines, rough watercolors, a sketchbook-like style, and illustrations cut from vintage magazines on worn beige paper. I surround the figures with handwritten English notes, arrows, amusing observations, doodles, and charming {argument name="details" default="personality analyses"} gleaned from the photographs.

A vertical A4 poster in the style of a high-end Japanese business magazine cover. At the center is a character representing the '{argument name="personality type" default="ENTJ COMMANDER"}'. An image of an '{argument name="character description" default="intelligent Japanese man in his 40s"}' who blends the roles of a tax accountant, entrepreneur, and AI instructor. He holds a tablet in one hand, surrounded by motifs of AI networks, data flows, real estate, seminar stages, and educational content. The background features a futuristic city and a digital workspace. At the top, the catchphrase '{argument name="catchphrase" default="Pioneer of the AI Era"}'. On the left and right are organized infographics. The content includes 'Strengths: Strategic thinking, execution, learning speed, systematization, communication' and 'Notes: Over-committing, dispersion, easily bored'. At the bottom is a radar chart showing E60 I40 / N85 S15 / T70 F30 / J60 P40. Sophisticated typography, luxurious feel, full color, quality of a magazine feature cover, with a focus on information design, in a composition where the personality can be understood at a glance.

[Information Input Section]

Character Name: [{argument name="character name" default=" "}] *Required

Supplementary Information: [{argument name="supplementary info" default=" "}] *Enter any specific details you want to tell the AI

---------------------

Using the attached image as the primary reference, carefully analyze the appearance, hairstyle, hair color, eyes, outfit, accessories, expression, motifs, atmosphere, occupation, and world-view of the character named [character name]. Draw them in a high-quality 2D animation style without losing the character's essence from the original image.

The theme is a 'Gambling Aptitude Research Report'.
The image should be a fixed 3:4 vertical orientation.
The composition should be a luxurious, flashy, and playful 'diagnostic poster and research report' style design.

Important:
The character must be positioned prominently in the center of the screen as the main subject, primarily in a bust-up size.
Ensure a composition where the face, expression, hair, and upper body clothing are clearly visible, and the character does not become too small.
Utilize the space around and above/below to place diagnostic items, scores, comment sections, and observation notes in a large, easy-to-read, and visually appealing manner.
Create a layout that balances 'character charm' with the 'fun of a diagnostic image'.

Do not fix the background or decorative gambling motifs; change them every time.
Candidate motifs include roulette, cards, poker chips, slot machines, horse racing, cycling races, boat races, dice, coin tosses, bingo, lotteries, casino tables, etc.
Randomly select 2 to 4 types from these candidates and reflect them in the background, decorations, analysis panels, and prop direction in a way that fits the character's world-view.
Do not lean toward the same motifs every time, especially do not fixate only on roulette and poker.
The selected motifs should blend naturally into the screen and be treated as part of the information design.

Please include the following elements in Japanese, in an easy-to-read and high-design manner:

[Required Elements]
1. Title
- 'Gambling Aptitude Research Report'
- Add character titles or aliases as necessary.

2. Large Numerical Score
- Gambling Aptitude: 0-100%
- Make it a large, prominent display.
- Add a short classification label if necessary (e.g., 'High-risk high-return type', 'Intuition-based type', 'Seems steady but dangerous type', etc.).

3. Gambler Type Name
- Give it an impressive and interesting type name that reflects the character's personality.
- It can be cool or humorous, as long as it fits the character.

4. Aptitude Parameters
- Intuition
- Bluffing ability
- Risk tolerance
- Ability to walk away
- Money management skills
Display each with numerical values or simple visual meters for easy reading.

5. Observation Notes
- Up to 2 items.
- Short sentences that convey the character's personality or gambling tendencies.
- A bit comical or sharp is fine.

6. General Evaluation Comment
- 1 to 2 memorable sentences.
- Make it a comical and sharp entertainment diagnosis.
- Example: 'Can see the winning path, but once on a roll, the wallet lightens in an instant.'
- However, do not include content that encourages actual gambling.

[Optional Extra Elements]
- Character's gambling tendencies (e.g., 'Strong in reading opponents', 'Suited for short battles', 'Weak against rare effects', etc., 2-4 items).
- A 'Warning!!' or 'One-point advice from the researcher' section.
- Playful elements like small stamps, crests, mini-icons, or chibi-style points.

The content should maintain the look of a serious research report while keeping the overall tone as a 'fun fictional diagnosis'. Do not make it too stiff or boring; ensure it's something people would laugh at, enjoy looking at, and want to save.

Important:
- The character must be the protagonist.
- Maintain the facial features, outfit, colors, motifs, and atmosphere of the original image.
- Use backgrounds and decorations that match the character's world-view.
- Use readable font sizes.
- Avoid overly detailed long texts, tiny annotations, garbled characters, or cluttered information.
- Avoid changing the person, significantly altering the outfit, changing the age, excessive exposure, low quality, or adding unnecessary people.
- If the original image has weapons or characteristic items, they can be reflected naturally within a range that reinforces the character, but do not let them overshadow the 'Gambling Aptitude Diagnosis'.
- If [supplementary info] is provided, prioritize reflecting its content.

Style: High-quality 2D animation
Composition: Luxurious and fun, research report-style vertical diagnostic poster
Aspect Ratio: 3:4 fixed vertical

Create a vertical 3:4 {argument name="page type" default="product design catalog page"} with a {argument name="background" default="warm neutral paper-like background"}.

Top section — lifestyle hero shot: place the product (use the uploaded image as the exact reference, preserving its form, proportions, materials, and identity without redesign) center-dominant with generous whitespace. Setting is a minimal architectural interior with a textured plaster wall and subtle concrete/stone floor. {argument name="lighting" default="Lighting is natural sunlight angled from the side, soft but casting high-contrast shadows"}. Render in editorial lifestyle photography style, high realism, warm and muted color grading.

Bottom section — technical specification panel laid out in a clean modular grid:
- Bottom left and center: orthographic architectural line drawings showing front view, side view, and three-quarter cutaway/profile view. Lines in muted red or sepia, fine technical weight, with minimal editorial measurement and construction callouts.
- Bottom right: 3-4 material swatch samples derived from the product's actual materials (fabric, leather, metal, wood, or plastic as applicable), in square or rectangular format with small editorial captions.

Typography: minimal editorial style, subtle captions only, no large headlines, soft black or dark brown.

Overall mood: design catalog / product design journal — architectural, premium, calm. No clutter, no bold colors, no heavy branding, no decorative graphics, no perspective distortion in the technical drawings.

Ultra-realistic aesthetic 3-frame portrait collage of a {argument name="subject" default="beautiful young woman"}, soft natural makeup, glowing fair skin, rosy cheeks, glossy pink lips, expressive eyes, {argument name="hair style" default="long wavy brown hair"}, delicate dangling earrings. {argument name="lighting" default="Warm golden-hour sunlight"} streaming through a window, creating soft shadow patterns on a beige wall. Cozy minimal indoor setting with a small green plant in the background. Frame 1: close-up selfie pose, slight duck-face expression, looking directly at the camera.

Cinematic close-up portrait of a {argument name="subject" default="man"} use image for face reference with same hair and light stubble, looking over his shoulder toward the camera with an intense, wearing oval same type of sunglasses but transparent, confident expression. Dramatic {argument name="lighting" default="neon lighting"} with deep blue key light and vivid red rim light creating strong color contrast across the face. Moody {argument name="atmosphere" default="cyberpunk atmosphere"}, shallow depth of field, soft bokeh, dark blurred background, highly detailed skin texture, sharp eyes, professional studio photography, ultra-realistic, high contrast, volumetric lighting, cinematic color grading, 85mm lens, f/1.8, photorealistic, premium magazine portrait, 8K quality.

Create a premium FIFA World Cup poster design featuring a {argument name="player type" default="legendary football superstar"}. Minimalist white background with a giant monochrome close-up portrait occupying most of the canvas. The portrait is divided into several horizontal slices with thin white gaps between them, creating a modern fragmented editorial effect. On the left side, large vertical bold typography spelling the player’s name in {argument name="color scheme" default="national team colors"}. A full-body action pose of the player overlaps the portrait in the foreground, wearing the national team jersey and celebrating confidently. National flag and federation crest placed on the upper right. Elegant autograph signature near the portrait. Soft fog/smoke fading at the bottom. Luxury sports magazine aesthetic, {argument name="branding style" default="Nike-style"} advertising campaign, high contrast black-and-white portrait, sharp facial details, clean composition, premium typography, poster design, sports branding, editorial layout, cinematic lighting, ultra realistic, 8k, professional graphic design, negative space, modern FIFA World Cup promotional artwork.

Style Keywords:
Sports poster, Editorial design, Modern typography, Monochrome portrait, Dynamic athlete pose, White minimal background, Premium branding, High-end advertising, World Cup campaign, Ultra detailed, Photorealistic, Graphic design masterpiece

Draw a {argument name="subject" default="beautiful girl"} in {argument name="style" default="Japanese anime style"}

Generate a {argument name="composition" default="9:16 vertical"}, high-precision ancient Chinese fantasy-style refined portrait of a woman. The main subject is a young adult Eastern woman, approximately {argument name="age" default="20–28"} years old.

Create a cinematic hyper-realistic portrait of a {argument name="character age and gender" default="21-year-old woman"} sitting casually on weathered stone stairs in a quiet park at twilight, beside an old glowing lamp post. She has a soft natural smile, relaxed posture, arms resting on her knees, and slightly messy dark hair styled in a long side braid. Outfit includes exactly 5 visible pieces: a {argument name="shirt color and style" default="pink heavyweight oversized T-shirt"}, black cargo pants, white sneakers, a black smartwatch on one wrist, and minimal natural makeup. Compose her seated on the right side of the frame, with the vintage lamp post on the left emitting warm yellow light; add two small blurred park lights in the background. Use shallow depth of field, soft bokeh trees, warm dusk ambience, realistic skin texture, natural proportions, and cinematic color grading. The stone staircase should look aged, stained, and textured, with a curved side wall leading diagonally through the composition. Shot on a full-frame camera with an 85mm lens look, f/1.8 depth of field, balanced twilight exposure, ultra-detailed photorealism, no text, no watermark.

Create a warm, whimsical photorealistic miniature scene in a cozy pastel study workspace: a tiny cute young South Asian woman, {argument name="character name" default="unnamed girl"}, sits on top of a giant modern smartphone lying flat on a desk, as if the phone is a glossy black platform. She has long wavy dark hair with a side part, delicate drop earrings, and wears an oversized chunky-knit {argument name="sweater color" default="soft blush pink"} sweater dress, white socks, and chunky white sneakers; pose her seated casually with one leg crossed over the other and both hands resting beside her on the phone. Use a dreamy golden-hour aesthetic with sunlight streaming through a large window on the left, soft haze, bokeh, shallow depth of field, and a feminine pink-beige color palette. Surround her with exactly 11 visible workspace objects: 1 rose-gold giant smartphone, 1 open laptop on the right with a pink screen showing motivational script text, 1 pink coffee mug with latte art and a gold handle, 1 pair of pink over-ear headphones in the foreground right, 1 pink spiral notebook in the foreground with gold handwritten words “ideas plans action”, 1 pink pen, 1 pink pencil cup holding pens and pencils, 1 small succulent in a ribbed pot, 1 small desk sign reading “good things take time”, 1 glass vase of airy pale pink flowers behind the girl, and 1 small decorative tray with pastel accessories in the foreground left. Background details include soft shelves with books and pink decor, sheer curtains, and an elegant desk surface. Make the phone enormous relative to the miniature woman, preserve realistic shadows and reflections on the black screen, and keep the overall mood cozy, aesthetic, luxury, cute, and social-media-ready. Avoid extra people, extra text, logos, watermarks, or distorted anatomy.

Goal: Create a polished vertical promotional poster for {argument name="product name" default="GPT Image 2"} explaining that the image model thinks before rendering, using a playful 3D toy-like visual style.

Canvas: 4:5 vertical social media poster, deep cobalt-blue background with a subtle radial glow and soft vignette, high-resolution glossy 3D render aesthetic.

Main visual: Center the top half around a surreal 3D object: a rounded glossy royal-blue cube fused with a shiny pink brain hemisphere protruding from the cube's right side. The cube is angled slightly so the front-left face and top are visible, with soft studio highlights, rounded edges, and a grounded shadow. The brain should look like inflated candy or silicone, with rounded folds, bright pink highlights, and a soft magenta glow where it meets the cube.

Decorative elements: Scatter exactly 34 small floating candy/confetti objects around the poster, avoiding the main text area. Include these visible types: 1 yellow star, 2 curved jellybean shapes, 11 short capsule sprinkles, 11 small spheres/dots, 4 larger glossy balls, 3 tiny round dots, and 2 partial edge sprinkles. Use pink, yellow, mint green, purple, orange, white, and red, with soft shadows and glossy highlights.

Text layout: Use bold white sans-serif typography, centered. Large headline in the middle: “{argument name="headline text" default="It reasons, then renders."}” on two lines, very heavy weight, clean geometric sans serif, slight shadow for contrast. Beneath it, smaller bold subtitle: “{argument name="subtitle text" default="GPT Image 2 plans your image before it makes a pixel."}” on two centered lines.

Instruction card: At the lower third, place a rounded off-white command-card panel with a subtle drop shadow, like a copyable prompt box. On the right side, attach a coral-red rounded rectangular button partially behind the card, labeled “COPY” in dark bold text. Inside the card, show exactly 4 monospaced lines in dark charcoal: “PLAN: composition and constraints.”, “SEARCH: facts you need, or SKIP.”, “GENERATE: the brief.”, “VERIFY: check output before showing.”

Footer: Bottom-left, add a tiny white line-art dog/mascot icon followed by the text “{argument name="footer label" default="GPT IMAGE 2 · PLAIN AND SIMPLE"}” in small uppercase white sans serif. Bottom-right, add the progress text “{argument name="progress text" default="1 / 7"}”.

Visual style: Bright modern 3D advertising, glossy plastic materials, candy colors, rounded friendly forms, strong contrast, crisp typography, soft realistic shadows, no photorealistic people, no watermark, no extra text.

Goal: Create a horizontal social-media art poster combining a vintage French melodrama painting with a black poetry-and-credits panel.

Canvas: Wide 16:9 composition, approximately 1024×576. The image is divided into two main columns: a large painted scene on the left taking about 60% of the width, and a flat black text panel on the right taking about 40%. Add a thin dark brown border around the full artwork.

Left scene: Paint an early 20th-century European interior in a muted post-impressionist / Toulouse-Lautrec-inspired oil-painting style with visible brush texture, aged varnish colors, ochre, teal, cream, and black. Show exactly 2 women in a tense dramatic conversation. Woman 1 stands on the left in a loose white blouse and dark skirt, turned toward the other woman, both hands raised in alarm with spread fingers. Woman 2 sits on the right on a bed or cushioned bench, wearing a low-cut black slip dress and black stockings, posed coolly with one elbow bent and hand near her chin, one knee raised, giving the impression of a dangerous femme fatale. Behind them include exactly 3 major background elements: a tall dark wooden mirror, a narrow vertical wooden panel or door, and an iron bedframe. Both women's faces should be obscured by soft rectangular blur blocks, as if anonymized or censored.

Subtitle: At the bottom of the painted scene, add a white italic film subtitle with a thin black shadow reading “{argument name="French subtitle" default="Vous êtes folle!—"}”. Place it centered near the bottom edge of the painting.

Right panel layout: Solid black background. At the top center place one small beige square QR-code-like graphic with exactly 3 visible black corner markers. Below it, center a poem in hand-lettered golden ochre text. Use casual uneven handwritten lettering, not a standard clean font.

Main poem text: “{argument name="poem text" default="you’re insane!\nshouts the distraught woman\nin French—\nI thank goodness for subtitles\nand dangerous femme fatales"}”. Keep the lines centered and stacked, with generous spacing.

Credits text: In the lower middle of the right panel, add smaller golden handwritten text reading: “AI image created by\n{argument name="artist name" default="Michael H. Lester"}\nin collaboration with\nGPT Image 2 Low on\nNightcafestudio\ntwitter: mlhlester”.

Bottom links: Along the bottom of the right panel add two centered golden lines: “{argument name="creator link" default="https://creator.nightcafe.studio/?ru/mhlester"}” and “{argument name="portfolio link" default="https://4-michael-lester.pixels.com/"}”.

Constraints: Use exactly 2 women, exactly 1 subtitle line, exactly 1 QR-code-like square, exactly 1 main poem block, exactly 1 credits block, and exactly 2 bottom link lines. Keep the right panel mostly empty black space, maintain a moody noir theatrical tone, and avoid photorealism.

Create a realistic paparazzi-style street photograph of {argument name="character name" default="an elegant woman"} stepping out of the Four Seasons hotel entrance in a busy city. She is centered full-body, walking confidently from the curb toward the camera, wearing a fitted black short-sleeve wrap dress with a deep V neckline and a high thigh slit, black pointed-toe stiletto pumps, and carrying a small structured black handbag in her left hand. Her face is intentionally anonymized with a soft rectangular blur, while her medium brown shoulder-length hair is visible in natural sunlight. Surround her with exactly 3 hotel security or doorman figures on the left near the doorway and a dense crowd of photographers on the right holding cameras with flashes aimed at her. Include a glossy black luxury sedan parked at the curb on the right foreground with a partial European-style license plate. The background should show the Four Seasons entrance sign, stone architecture, doorway shadows, and bright midday sunlight casting strong realistic shadows on the pavement. Style: high-end celebrity candid photo, editorial realism, 35mm street photography, natural contrast, slight motion energy, no glamour retouching, vertical 3:4 composition.