GPT Image 2 prompts

Generate a single-page technical specification sheet, A3 portrait, aspect ratio 1:1.4, black line art on white background with Pantone chips.

Style: LX-DR-01, evening gown, 1970s Halston bias drape combined with 2025 circular design principles.

Layout:
  left: front flat and back flat at 1:6 scale
  center: detail views at 1:2 scale (cowl neck, bishop sleeve cuff, detachable train attachment, side slit)
  right: bill of materials, measurement chart, title block

1970s elements:
  - bias-cut silhouette with cowl neckline depth 18 cm from HPS
  - long bishop sleeves gathered into 6 cm cuff height with covered button closure
  - empire seam positioned 5 cm below bust line
  - floor length with 20 cm side slit at left leg
  - bias drape at hip

2025 elements:
  - shell 19 momme 100 percent recycled silk charmeuse with blockchain traceability, Pantone 19-5408 TCX Deep Emerald
  - lining 100 percent cupro bemberg Pantone 11-0602 TCX Gardenia
  - detachable train length 80 cm with hidden magnetic closure at center back waist
  - 3D-knitted internal bodice support at bust
  - zero-waste pattern layout noted at 92 percent efficiency
  - NFC authenticity chip encased in right side seam
  - printed QR code care label

Measurements size FR 38:
  center back length 150 cm from nape to hem, bust 1/2 43 cm, waist 1/2 34 cm, hip 1/2 46 cm,
  sleeve length 62 cm from shoulder point, sleeve cuff circumference 18 cm, sweep 1/2 75 cm, slit height 20 cm

Construction:
  French seams throughout body using 301 lockstitch at 14 spi, seam allowance 1 cm
  hand-rolled hem 3 mm at bottom and train edge
  invisible zipper center back length 55 cm
  bar tack at slit top

Materials and color:
  thread 100 percent silk size 120 to match shell
  hardware: two covered buttons 12 mm diameter one per cuff, six pairs neodymium magnets 8 mm diameter for train

Labels:
  main label center back neck, care label left side seam with text "Dry clean only. 100 percent recycled silk." in English and French and Italian

Render annotations in 8pt sans-serif horizontal with leader lines, no shading.

Generate a full-length editorial image, aspect ratio 3:4, of a female model age 24 height 178 cm measurements 86-61-88 cm.

Garment exact specification LX-DR-01:
  - bias-cut evening gown in 19 momme 100 percent recycled silk charmeuse with blockchain traceability Pantone 19-5408 TCX Deep Emerald, lining 100 percent cupro bemberg Pantone 11-0602 TCX Gardenia
  - cowl neckline depth 18 cm from HPS, empire seam 5 cm below bust line
  - long bishop sleeves gathered into cuff height 6 cm with two covered buttons 12 mm diameter one per cuff, sleeve length 62 cm from shoulder point, cuff circumference 18 cm
  - floor length center back 150 cm from nape to hem, sweep 1/2 75 cm, 20 cm side slit at left leg with bar tack at slit top
  - detachable train length 80 cm attached at center back waist with hidden magnetic closure using six pairs neodymium magnets 8 mm diameter
  - internal 3D-knitted bodice support at bust, French seams throughout with 301 lockstitch at 14 spi seam allowance 1 cm, hand-rolled hem 3 mm at bottom and train edge, invisible zipper 55 cm center back
  - zero-waste pattern layout 92 percent efficiency, NFC authenticity chip in right side seam, printed QR care label left side seam, main label center back neck, thread 100 percent silk size 120

Styling: paired with 1970s satin strappy heels in champagne, 2025 recycled gold ear cuffs, hair in low sleek chignon.

Pose: standing in contrapposto, left leg forward to show side slit, right hand lightly holding train to reveal magnetic attachment point, gaze to camera, professional runway posture.

Environment: 1970s marble staircase with brass railing, updated with LED cove lighting, evening ambient light.

{argument name="character" default="White Bone Demon"} transformation video cover, {argument name="transformation scene" default="a village girl instantly transforms into a glamorous queen"}, with special effect {argument name="visual effect" default="surrounded by flames"}.

Use the uploaded image as the only {argument name="identity reference" default="identity reference"} and transform the subject into a character rendered in the official {argument name="anime series" default="Demon Slayer: Kimetsu no Yaiba"} anime TV series style. The visual execution must strictly match the Demon Slayer aesthetic, featuring clean and crisp linework, high quality cel shading, and saturated yet balanced colors with a polished TV series anime finish. Do not include any manga style sketching, cross hatching, grayscale, or black and white rendering. The subject's identity must be fully preserved, including their clothing design, hairstyle and hair color, accessories, skin tone, facial expression, and pose, with no replacements or removals of any items.
The existing outfit should be subtly stylized with Demon Slayer Corps inspired detailing, such as sharper fabric folds, slightly reinforced layered textures, and clean anime fabric finishes, while keeping the original outfit clearly recognizable. The character must follow accurate Demon Slayer proportions, including a lean and agile body type, natural anime anatomy, expressive anime eyes with strong highlights, stylized brows, and simplified anime nose and mouth.
The composition is critical and must show the character fully from head to toe, with feet, legs, torso, arms, and head all visible. The framing should not crop any part of the body, and the camera distance must be wide enough to capture the entire figure, leaving a small amount of empty space above the head and below the feet so the character fits comfortably within the frame. The output must be a single full body illustration with the character centered against a {argument name="background" default="plain white background only"}, with no ground plane, shadows, scenery, props, weapons, objects, effects, aura, particles, motion lines, or environmental details. The final image should contain only the character, and must not include cropped framing, multiple characters, multi panel outputs, other anime styles such as Dragon Ball, Naruto, or One Piece, photorealism, painterly or textured rendering, or any background elements or objects.

Please generate a realistic close-up of hand nail art, with the theme inspiration: {argument name="theme" default="jellyfish"}.

The main subject is a close-up of a meticulously cared-for female hand. The nails feature high-end gel designs, centered around {argument name="theme" default="jellyfish"}, presenting an elegant, clean, refined visual effect characteristic of fashion magazines. Emphasize the detail of the nail art, the glossy texture, and the consistency with the theme's atmosphere.

Composition and Subject Requirements:
The focus is on the hand and nail close-up. Finger gestures are natural and elegant, conveying a sense of luxury. Each nail is designed around the theme, using combinations like hand-painted elements, solid bases with glitter/foil accents, and varied but cohesive styles. Designs should include premium base gel, delicate painting, and matching decorations like sequins or metallic foil. The style must be professional and sophisticated, not cheap or juvenile.

Lighting and Photography:
Soft natural light, no harsh shadows, shallow depth of field, and sharp focus on nail details and gloss. High-end beauty editorial style, professional commercial photography, realistic skin textures, and 4K ultra HD clarity.

Background:
A soft, blurred background that echoes the {argument name="theme" default="jellyfish"} without distracting from the main subject.

Overall Style:
Photorealistic, glossy gel finish, elegant, and clean. Avoid cartoon/illustration styles, distorted fingers, or low-quality textures.

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

Using the {argument name="brand name" default="[BRAND NAME]"}, create a highly detailed cinematic poster:

“HOW {argument name="brand name" default="[BRAND NAME]"} TAKES OVER A CITY”

GOAL:
Show the brand expanding across an entire city ecosystem until it becomes impossible to ignore.

This must feel like:
global campaign rollout + futuristic city domination + luxury advertising ecosystem.

The brand should appear integrated naturally into the city, not random logo placement.

CORE RULE:

Everything must be visually inspired by the brand identity of “{argument name="brand name" default="[BRAND NAME]"}”:

brand colors

typography style

design language

visual energy

aesthetic mood

The city itself should begin to feel transformed by the brand identity.

MAIN STRUCTURE:

Vertical 4:5 poster
Dense multi layered composition
Huge cinematic city scene mixed with many smaller detail panels

TOP SECTION:

Include:

“{argument name="brand name" default="[BRAND NAME]"}” logo or wordmark

Campaign title

Short {argument name="slogan" default="slogan (max 5 words)"}

MAIN HERO SCENE:

Show a massive cinematic city environment where “{argument name="brand name" default="[BRAND NAME]"}” has become dominant.

Include:

giant digital billboards

skyscraper LED screens

branded taxis

buses

train wraps

storefronts

public transport stations

branded cafés

interactive street ads

branded drones or futuristic media systems

The city should feel alive and overloaded with coordinated branding.

CITY ENVIRONMENT:

Create:

realistic streets

crowds of people

traffic

nightlife

urban lighting

weather atmosphere

The branding should blend naturally into the environment.

MICRO SCENES:

Around the main scene, include many smaller panels showing:

people using branded products

social media campaign visuals

branded pop up events

digital ads on phones

branded shopping bags

influencer campaigns

café branding

event installations

giant building projections

ADVERTISEMENT SYSTEM:

OUTDOOR:

billboards

LED towers

transit ads

street posters

DIGITAL:

mobile ads

social media ads

app UI

interactive screens

PHYSICAL:

packaging

shopping bags

uniforms

vehicles

VISUAL DETAILS:

Add:

glowing reflections

rain reflections on streets

holographic screens

animated signage

crowds interacting with ads

city lights matching the brand palette

COLOR SYSTEM:

Strictly derived from the brand identity colors.

Use:

lighting

reflections

atmosphere

ad screens

to reinforce the brand identity.

STYLE:

cinematic urban realism

futuristic advertising aesthetic

dense but readable

luxury campaign presentation

Inspired by:

Tokyo/Shibuya LED overload

Times Square campaigns

cyberpunk city advertising

luxury fashion campaign rollouts

DEPTH:

40 to 80 visual elements

layered foreground/midground/background

multiple scales of detail

IMPORTANT RULES:

no random logo spam

branding must feel intelligently integrated

every ad placement should feel int

A stunningly beautiful {argument name="subject" default="young Asian woman"} with a K-pop idol aesthetic, featuring a soft V-shaped face and pale porcelain skin. She has long, {argument name="hair color" default="platinum white hair"} with airy see-through bangs and loose, wind-blown strands creating a natural messy look. Her eyes are large, crystal blue (resembling contact lenses), looking directly at the viewer with a pure and innocent expression. She is wearing a tight-fitting {argument name="outfit" default="grey ribbed knit camisole"} that highlights her figure, paired with a matching grey skater skirt. The background is a soft, out-of-focus park scene with creamy bokeh. The lighting is diffused and airy, creating a dreamy, soft-focus photorealistic portrait, 8k resolution, detailed texture.

Please generate a high-quality "Urban Line Atlas Poster."

This image is designed based on the visual language of a "monochrome urban line-drawing travel poster," featuring a complete composition with a large top title, mid-view street depth, core landmarks, flanking buildings, foreground figures, and a bottom signature block.

[Theme City]
City Name: {argument name="city" default="Yichun"}
English Main Title: {argument name="English title" default="FOREST LINE ATLAS / YICHUN CHRONICLE"}
Chinese Title: Yichun · City of Forests
Country/Region: Heilongjiang Province, China
Core Landmarks: Lesser Khingan Mountains forest, Korean pine forest sea, Wuying National Forest Park observation tower
Street Scene: Small forest town streets, residential intersections with wooden buildings, forestry city atmosphere
Emotional Theme: Commuting moments in the morning mist
Main Color Tone: {argument name="main color" default="dark green + dark grey line drawing"}
Aspect Ratio: 3:4 vertical

[Overall Positioning]
This is a work that serves as a travel souvenir poster, an urban visual guide, an architectural line manuscript, and an editorial design cover. The image needs to present Yichun's unique temperament as a "city in the forest," with both urban recognition and a sense of real street life—not a standard travel poster or a single landmark illustration, but a visual archive piece of collectible value.
[Composition Design]
The top is a strong title area using the large English title "FOREST LINE ATLAS," supplemented by the subtitle "YICHUN CHRONICLE." A numbering system (e.g., No. CN-HLJ-YC-01) and brief coordinate info can be added, with a modern and design-oriented layout.
The middle is the urban street主体, using a deep street perspective composition with a slightly low angle, the road extending into the distance and gradually entering the forest area.
The core landmark is the forest observation tower or the silhouette of the Lesser Khingan forest in the distance, clear but not overly prominent, naturally blending into the street scene.
Add local architectural details on both sides: wooden houses, forestry bureau-style buildings, residential blocks, small shops, bus stops, streetlights, utility poles, simple road signs, etc., forming high-density urban information.
The foreground includes lifestyle elements: pedestrians in autumn/winter clothing, people on bicycles, residents with shopping bags, small vehicles, and fallen leaves on the ground to enhance the seasonal atmosphere.
A separate signature area at the bottom is designed as a city archive label structure, containing "YICHUN / HEILONGJIANG / CHINA" and a simplified city info module or map index-style layout.
[Text System]
The top main title is clear and eye-catching. Keep only 2–4 clear readable texts in the image (e.g., "Yichun Station," "Forestry Road," "Bus Stop"), with other signs using abstracted Chinese strokes or symbolic characters to avoid gibberish. The overall text system should be unified, restrained, and integrated into the design structure.
[Visual Style]
The overall style uses monochrome or a very limited two-color line drawing on an off-white or warm white paper texture. Lines are precise, clean, and layered, with the feel of an architectural pen manuscript and screen print texture. Foreground lines are slightly thicker, background lines thinner, with figures and objects being simple but recognizable. The style is restrained and sophisticated.

[Innovation Requirements]
Incorporate design variations in: title naming system, top layout structure, slightly low-angle street perspective, fusion composition of forest and city, bottom archive-style information module, and expression of autumn fallen leaf atmosphere.
[Avoidance Matters]
No ordinary travel poster style, no watercolor, impasto, or cartoon/chibi styles, no cyberpunk, no high-saturation colors, and no cheap souvenir style.

Please generate an independent, complete, high-definition, clear-lined vertical urban line-drawing poster with a sense of collectability and series identity.

Please generate a 2×2 four-grid collage photo set based on the photo uploaded by the user, with the theme of "French Vintage Oil Painting Girl Soft Boudoir Photography."

Strictly retain the subject's real identity features, including face shape, facial proportions, eyes, nose, lips, skin tone, perceived age, hairstyle basics, and overall temperament. Do not turn the person into a stranger. Extend the body naturally from the facial features; do not make it too thin, keep a slight soft fleshiness.

## Overall Style
The overall set is in the French vintage oil painting girl soft boudoir style, featuring very obvious hazy soft focus, a dreamlike atmosphere, divine halos, and a soft oil-painting-like texture. The temperament should be quiet, soft, fragile, poetic, and lazy, with a touch of slight melancholy and feminine vitality, like a girl deity statue wrapped in flowers, white gauze, and warm light.
It is not ordinary boudoir photography, not ordinary wedding photos, and not ordinary Korean-style studio shots, but rather: a vintage oil painting girl portrait under warm golden twilight light, featuring soft boudoir body lines, a dreamlike atmosphere, and a sense of goddess-like divinity.

## Makeup and Hairstyle
The makeup should be a translucent soft-mist base, warm cream skin tone, low-saturation nude pink or red-brown eye makeup, natural eyelashes, and bean paste nude pink or rose nude lips. The overall makeup should be soft, not heavy, and not sharp.
The hairstyle is French vintage long hair, which can include loose braids, a voluminous crown, natural wisps, slight curls, and a slightly messy airy feel. The hair should not be too tidy, but should have soft, loose, oil-painting-like layers.

## Clothing
Clothing consists of cream-white, ivory, or off-white French vintage lace long dresses, off-the-shoulder tulle dresses, and vintage lace dresses. It should feature lace, light gauze, semi-transparent cuffs, soft folds, and layered skirts. The overall look is sacred, soft, and feminine, like a combination of a vintage nightgown and a light wedding dress feel, but avoid looking too much like modern commercial wedding photography.

## White Gauze and Flowers
The scene must contain large areas of light and transparent white gauze, naturally draped over the back of chairs, table edges, the floor, and around the person, creating a misty sense of envelopment, floating, and airy layers.
Floral materials should mainly be light pink roses, pink-white bouquets, white lilies, or soft light-colored flowers. They can be held in the person's arms, placed on the table, or scattered on the floor to enhance the soft beauty, vitality, and goddess atmosphere.

## Scene
The scene is a minimalist vintage studio space with dark gray-brown, dark brown, or smoke-brown background cloth and floor cloth. Accompanied by a vintage round chair, a small round table, a lace tablecloth, a white vase, and a small amount of floral material. The scene is not complex but should have a soft, enclosed, dream-like small stage feel.

## Lighting and Texture
This is the most important part:
Please use very obvious warm golden side lighting or side backlighting, like sunset sunlight shining into an old room. The light should be soft but directional, making the skin, white gauze, and flower edges appear with a distinct glow, halo, bloom, and slight hazy flare.
The entire set must be hazier, softer focused, and lower contrast than ordinary photography:
- soft focus
- diffusion filter
- bloom
- halation
- dreamy blur
- hazy glow
- film grain
- oil painting softness
The image should look like it was taken through a thin veil; details should not be too sharp, not high-definition or rigid, and not too clean. The overall presentation should have obvious mist, halos, and soft oil painting textures.

## Emotion and Expression
The subject should not just be "quiet and pretty" but should have a sense of emotion and immersion. Please emphasize: fragility, laziness, thoughtfulness, slight melancholy, immersion, self-perception, quiet blooming, and divinity lit by light.
Expressions can include: eyes closed, eyes half-closed, looking to the side, head slightly bowed, head tilted up toward the light, looking back with a gaze, or an accidental downcast look. Reduce the posed feel of ordinary front-facing shots and increase the feeling of emotional moments captured by a photographer.

## Four-Grid Format
Generate one 2×2 four-grid collage, where each grid is a 3:4 vertical composition. All four images must belong to the same photo set: same person, same makeup system, same clothing system, same scene system, same lighting language, and same post-processing texture.

## Consistent Elements
The following elements remain unified across the four images: same girl, same French vintage oil painting soft boudoir style, same cream-white lace/gauze clothing system, same warm golden hazy side backlighting, same dark gray-brown vintage background space, same prop system of gauze, flowers, vintage chair, and small table, and the same soft focus, oil painting, film grain, and dream halo texture.

## Allowed Variations
To avoid the four images looking too similar, naturally vary the following: composition distance, camera height, poses (sitting/leaning/lying/turning/tilting head), the way the flowers are held, the flow of the white gauze, expressions and gaze direction, hand movements, and the relationship between the person and the chair, table, or floor.

## Lens Pool Rules
Randomly prioritize 4 shots from the following pool, ensuring at least 2 core shots are included:
### Core Shots (Choose at least 2)
1. Half-body portrait sitting and holding flowers
2. Emotional medium shot leaning against the chair back or table edge
### Other Shots (Choose to fill the rest)
- Tilting head up toward the light
- Looking back with an exposed back
- Lying on the floor
- High-angle lying shot
- Close-up side profile with closed eyes
- Full-body shot wrapped in white gauze
- Sitting in a slightly curled-up position
- Standing or leaning by the flower table
Please let the four images vary in lens choice while overall remaining like four moments from the same dreamlike oil painting photo set.

## Key Objectives
The whole set must be hazier, more glowing, and more emotional than ordinary photography. Prioritize: enshrouding soft light, warm golden dreaminess, feminine divinity and vitality, the sense of being wrapped in gauze and flowers, stronger emotional flow, and more obvious oil painting soft focus.

## Avoid
Avoid high-definition sharpness; avoid ordinary warm-toned studio feels; avoid modern wedding photo vibes; avoid Korean-style commercial studio looks; avoid plastic-like skin retouching; avoid being too clean and clear; avoid influencer-style photography; avoid cheap "fairy" styles; avoid messy backgrounds; avoid watermarks and logos.

Please transform the entire image into a single {argument name="style" default="flat decorative folk illustration"} with {argument name="elements" default="swirling elements"}. Use a {argument name="color palette" default="bold and playful color palette"}, different from the original image. Simplify all details into clean flat shapes with a hand-drawn, slightly imperfect feel, as if drawn on white paper. The overall style should look cute, childish, and imaginative.

Live stream shopping screenshot, host holding {argument name="product" default="lipstick"} for color testing, bullet chat filled with "{argument name="comment" default="Order placed"}"

Create a character sheet for the character only. Make the character's appearance natural with a {argument name="head to body ratio" default="7.5-8.0"} head-to-body ratio.

{argument name="character name" default="Miku Nakano"} prize figure

Create a comical scientific explanation illustration. The verification theme is "{argument name="validation theme" default="How many banana peels are needed to make a car slip?"}"

A vertical A4 manga-style infographic that seriously verifies absurd questions with science and calculations. White background, thick black lines, and emphasis colors of red, blue, and yellow. A design reminiscent of retro educational manga, science reference books, or scientific columns. Use a professor character, surprised people, speech bubbles, hand-drawn arrows, red pen annotations, and chalkboard-style formula boxes. The overall atmosphere should be 'using physics to verify a silly question with full effort,' making it fun, energetic, and not too serious.

Structure:

1. 
Depict the theme's situation in an exaggerated and ridiculous way that is understood at a glance. The characters are trying their best, but the appearance is quite reckless. The professor interjects with a comment from the side.

2. Thinking with Real Numbers
Display 3 to 4 boxes of important values related to the theme, such as speed, time, distance, weight, force, temperature, or energy. Include comparison charts or timelines to clearly show the gap between 'human senses' and the 'scale of the phenomenon.'

3. Formula Corner
Draw necessary conditions or comparison formulas largely in chalkboard-style frames. Example: 'Required Ability > Human Limit' 'Reaction Time < Time Phenomenon Occurs' 'Required Force/Speed/Durability ≫ Normal Human.' Formulas don't need to be strictly accurate; keep them intuitive and easy to understand.

4. Is Realization Impossible?
Explain problem points using three small panels. Example: ① Too fast / Too heavy / Too hot ② Precision is too severe ③ Even if successful, the body won't hold out / Surroundings will be in trouble. Include comical failure illustrations, onomatopoeia, and speech bubbles.

5. What If It Was Realized?
Make this part important. Illustrate what superhuman or speculative scientific abilities would be needed to actually succeed. For example, super reflexes, future prediction, abnormal grip strength, ultra-high-speed vision, physical ability to ignore air resistance, or a skeleton that can withstand impact. If necessary, use scientifically impossible numbers like 'reaction speed 0.0001s,' 'grip strength 3000kg,' or 'dynamic vision dozens of times that of a fighter pilot.' Furthermore, depict what happens the moment it is realized. Example: an impact that blows an arm off, sinking into the ground, shockwaves in the surroundings, while the person themselves remains calm. Bring out the 'if successful, they are no longer human' feeling in a fun way.

6. Conclusion
Write the conclusion largely in a red highlighted frame. 'Conclusion: XX is almost impossible for a normal human!' or 'Even if realized, it would be at a superhuman, cyborg, or speculative science level!' Summarize it in a way that states it's impossible for real humans but achievable with speculative scientific abilities.

Design Specifications: Manga paneling + science illustrations + professor's commentary. Use thick headings, red circles, emphasis lines, hand-drawn notes, and speech bubbles frequently. Formulas and numerical values should be large and easy to read, conveying the message intuitively even to children. Finish it like a page from an interesting science reading that is absurd yet strangely convincing.

Please create a design in the style of Japanese TV show telops (captions). Ensure all text used is in Japanese and clearly legible.

Font Styles:
- Main Heading: Thick, legible rounded Gothic similar to LINE Seed JP Heavy.
- Emphasized Telop: Ultra-thick Gothic typical of variety shows, with white outlines, red text, and yellow shadows.
- Supplementary Text: A professional typeface suitable for news programs, similar to BIZ UDP Gothic.
- Handwritten Memo: Rough script style like Tanuki Oil Magic marker.

Design:
Organize several telop samples onto a single sheet, like a collection from a Japanese variety or news show. The background should be gray. Use a bold color palette of red, yellow, blue, and white. Incorporate white outlines, black outlines, drop shadows, and fluorescent highlighter-style decorations. Include small labels like "Free Font" or "Usage Example" near each sample.

Please generate a resume based on the ChatGPT I usually use. Create it while reflecting this ChatGPT's personality. It should be a clear, business-appropriate resume with a touch of humor. Aspect ratio 4:5.

Please transform this illustration into a {argument name="style" default="professional Lo-Fi art style"}. Add warm, slightly muted pastel tones, film grain, soft focus, light fading, and a vignette to strongly express a nostalgic and chill atmosphere. Let soft natural light gently envelop the character, casting warm highlights and shadows on the skin and hair. Finish with the overall relaxed, somewhat nostalgic, and comfortable mood unique to Lo-Fi. Completely maintain the original composition and character.

Please create a photorealistic ita-bag of the character based on the first image. The bag color is '{argument name="favorite color" default="favorite color"}', decorated with ribbons and charms in '{argument name="favorite color" default="favorite color"}'. Inside the bag, include 20 character badges, the character's '{argument name="favorite motif" default="favorite motif"}', and a chibi-style plushie of the character. The scene is a photorealistic natural-style room with the bag placed on a wooden table.

{argument name="name" default="Luce"}
{argument name="personality" default="ESFP Entertainer, Sun Idol"}
Twin tails, energetic pose, {argument name="outfit" default="idol costume"}, light effects

Please create a fictional VTuber using the character in the image. Name: '{argument name="name" default="name"}', Motif: '{argument name="motif" default="motif"}', a VTuber streaming screen based on '{argument name="favorite color" default="favorite color"}' with a 'favorite atmosphere'. A room background based on '{argument name="favorite color" default="favorite color"}' with a 'favorite atmosphere'. Place the character in the center, a comment section on the right, a VTuber logo based on the character's motif in the top right, 'Chat Stream' text in the top left, and a character profile (type, age, likes, skills, catchphrase) in the bottom left. Place the time in the top center. Image size: 16:9.

Please use the uploaded portrait as a reference. Create a 2x2 summer selfie collage with a {argument name="aspect ratio" default="9:16"} vertical aspect ratio, comprising four panels. Ensure that the facial features remain consistent, with soft, youthful features, a warm smile and natural skin texture. In each panel, depict the character outdoors in bright summer sunshine, wearing a straw hat. Have her strike a variety of poses across the four panels: a peace sign selfie, a pose cupping her face with her hands, an OK gesture, and a relaxed close-up selfie. Keep the composition cute, lively and casual. Add colourful, watercolour-style hand-drawn doodles around her: soft outlines around her body and hat, floating soap bubbles, small hearts, simple sun drawings, sparkles, and cute little animal stickers. Ensure the doodles have a playful, hand-drawn feel, like mobile phone photo stickers. Remove all watermarks, logos, AI labels, app labels and signature marks. Keep the decorative handwritten text, but replace it with neat, legible words such as {argument name="sticker labels" default="‘Sheena’, ‘sunny’, ‘cute day’ and ‘summer’"}. Make ‘{argument name="main text" default="happy day"}’ the most prominent text. Ensure the text looks like colourful, hand-drawn sticker lettering. Style: Soft Korean/Japanese-style summer selfie aesthetic, bright yet subtle sunlight, fresh colours, a slight film-like softness, a refreshing outdoor atmosphere, cute social media collages, clean ultra-high-resolution images. Elements to remove: Watermarks, logos, brand marks, AI labels, signatures, QR codes, random text, illegible characters, misspelled ‘happy day’, long captions, cluttered typography, unnecessary people, duplicate faces within a single panel, distorted faces, asymmetrical eyes, plastic-looking skin, excessively smooth skin, blurry faces, lack of detail, unnatural body proportions, distorted hands, extra fingers, missing fingers, fused fingers, twisted arms, severe overexposure, dull colours, severe compression artefacts, dark and underexposed panels.

{argument name="subject" default="Portrait of a person"} portrayed in a Grayscale Cinematic Drama, inspired by classic film noir. Use strong contrasts and dramatic shadows to create a sense of tension and intrigue, with subtle variations of black and white

16:9, 2x2 grid, for “{argument name="food" default="sauerkraut"}”, “kombucha”, “kimchi”, “sourdough starter”. => fermentation_content = ( isometric_fermentation_station_setup_for_[fermented_food]_with_jars_weights_airlocks * 1.6 ) + ( microbial_timeline_chart_bacteria_yeast_activity_by_day_for_[fermented_food] * 1.5 ) + ( ph_and_temperature_sweet_spot_diagram_specific_to_[fermented_food] * 1.4 ) + ( taste_evolution_flavor_wheel_day_by_day_for_[fermented_food]_from_sweet_to_tangy * 1.3 ) + ( troubleshooting_common_ferment_fails_mold_kahm_yeast_soft_mush_with_photos_for_[fermented_food] * 1.5 ) + ( gut_health_benefits_explainer_linking_[fermented_food]_metabolites_to_microbiome * 1.2 ) + ( recipe_variant_matrix_spicy_herbal_fruity_editions_of_[fermented_food] * 1.1 ) + ( safety_and_sterilization_checklist_for_[fermented_food]_with_myth_busting * 1.0 ) + ( serving_suggestion_plate_scape_integrating_[fermented_food]_into_daily_meals * 0.9 ) - ( fear_based_“dangerous_bacteria”_stock_images / 3.2 ) - ( over_styled_hipster_mason_jars_with_no_instruction / 2.9 ) - ( unclear_burping_or_maintenance_routine / 3.1 )

PYTHON_SCENE_GRAPH :: PARAMETRIC_CITY_RELIEF class Variables: city = "{argument name="city" default="[CITY]"}" city_name_text = "{argument name="city name" default="literal city name from input"}" region_context = "infer country, terrain, climate, culture, urban identity" topography = "infer mountains, rivers, coastlines, plains, islands, deserts, hills" urban_grid = "infer district density, roads, transit corridors, urban pattern" landmarks = "infer landmark_set(city)" signature_core = "infer most symbolic central landmark or public space" style = "luxury 3D cartographic city model" class TerrainSlab: form = "thick raised cutout map base" surface = Variables.topography edges = "engraved title panel, legend, compass, scale, inset regional map" material = "matte stone/plaster/cartographic model material" class CityTypography: text = Variables. city_name_text form = "monumental 3D letters" function = "each letter is an inhabitable building mass" placement = "integrated into city map, not floating" rule = "word must remain readable from aerial view" class UrbanLayer: roads = Variables.urban_grid districts = "infer neighborhoods and density zones" landmarks = Variables.landmarks core = Variables.signature_core labels = "derive place labels from city geography" class Atmosphere: camera = "elevated three-quarter macro" lighting = "soft premium studio daylight" details = "vehicles, clouds, aircraft, trees, people only if appropriate" def render(): return """ Render {argument name="target city" default="[CITY]"} as a raised terrain-map diorama where the city name becomes monumental architecture, surrounded by inferred geography, landmarks, labels, roads, and atlas-style cartographic details. """

Transform only the {argument name="target object" default="food items"} in the image into tiny {argument name="style" default="LEGO-style miniature sculptures"} while keeping everything else completely unchanged and realistic. Preserve the original composition, lighting, camera angle, perspective, background, hands, table, people, utensils, and environment exactly as they are. The food should appear rebuilt from authentic LEGO bricks and pieces: visible studs, blocky geometry, plastic texture, tiny LEGO tiles, rounded brick edges, and miniature construction details. Keep the food instantly recognizable while translating it into a highly detailed LEGO miniature aesthetic. The LEGO food should match the original food’s colors, proportions, toppings, and arrangement. Maintain realistic shadows, reflections, depth of field, and interaction with the environment so the LEGO food feels naturally placed inside the real photograph. The contrast between the realistic world and the giant LEGO food sculptures should feel playful, surreal, and highly believable. Add 3-4 big {argument name="extras" default="emojis"} to the picture.