GPT Image 2 prompts

Ultra-realistic ethereal fantasy portrait of a breathtaking young woman with delicate porcelain skin, soft grey-blue eyes, and natural rosy lips. She gazes gently toward the viewer with a serene, dreamy expression, her fingertips lightly touching her chin. Wispy ash-brown hair flows softly in the breeze, styled in a loose romantic updo adorned with pastel blush roses, shimmering crystal ornaments, delicate feathers, and intricate floral accessories. She wears elegant dangling crystal earrings and a translucent, flowing gown made of sheer iridescent fabric embroidered with tiny sparkling flowers.

The scene is bathed in soft diffused morning light, creating a luminous glow around her face and shoulders. Surrounded by floating butterflies, sparkling dust particles, translucent petals, and dreamy floral textures, the background blends pastel lavender, pearl white, blush pink, and silver tones. Cinematic fine-art photography, fairycore aesthetic, enchanted garden atmosphere, magical realism, ultra-detailed skin texture, soft focus highlights, volumetric lighting, bokeh, masterpiece quality, highly detailed, 8K resolution, delicate feminine beauty, romantic fantasy artwork, elegant composition, dreamy color grading, soft glow, celestial ambiance.

Create an ultra-realistic premium 1864 Victorian daguerreotype portrait using the uploaded photo strictly as the identity reference of the SAME woman. ABSOLUTE IDENTITY PRESERVATION — HIGHEST PRIORITY Preserve EXACTLY: - facial structure - bone anatomy - proportions - eye shape and spacing - nose shape - lips and smile structure - jawline and cheekbones - skin texture and pores - eyebrows - hairline - unique facial details and asymmetry - overall natural likeness and recognizability Do NOT beautify, redesign, replace, or reinterpret the face. Do NOT alter ethnicity, age, proportions, or identity. The final image must look unmistakably like the original person. STYLE & ERA: Authentic 1864 Victorian daguerreotype portrait photography. The woman appears posed in a historically accurate antique studio setting, captured with the realism and imperfections of early photographic processes from the mid-19th century. HAIR & STYLING: Long natural dark wavy hair with soft realistic texture and historically appropriate Victorian styling. Elegant, subtle, timeless appearance. WARDROBE: An off-the-shoulder cream lace corset gown with delicate Victorian details and realistic fabric texture. A simple antique gold locket necklace. COMPOSITION: Centered bust or half-body portrait with graceful posture and direct gaze toward the camera. Framed inside an antique oval vignette. SURFACE & MATERIAL DETAILS: Printed on aged yellowed parchment paper with authentic tea stains, faded edges, paper texture, dust particles, oxidation marks, scratches, subtle cracks, silvering artifacts, and realistic daguerreotype aging imperfections. TEXT: Elegant handwritten cursive calligraphy at the bottom reading: “LEO 1804” LIGHTING: Soft natural studio lighting with realistic falloff, subtle shadows, true-to-life skin rendering, and historically accurate exposure characteristics. Natural tonal transitions with no artificial beauty filter effect. COLOR & TONE: Sepia-toned faded monochromatic palette with authentic antique photographic chemistry aesthetics. Rich historical atmosphere, cinematic depth, hauntingly beautiful mood. REALISM & QUALITY: Ultra-realistic 8K masterwork. Hyper-detailed premium photography quality. Looks like a genuine professionally preserved Victorian daguerreotype scanned in ultra-high resolution from a museum archive. Natural skin texture, realistic eyes, realistic fabric, realistic lighting, physically accurate details. Shot as if captured by a world-class professional portrait photographer using an ultra-high-end medium format camera while preserving authentic 1864 photographic character.

A highly detailed monochromatic blue watercolor and ink portrait illustration of {argument name="subject" default="Bill Gates"}, centered in a vertical poster composition. The artwork features a realistic, dignified head-and-shoulders portrait with fine pencil sketch linework, delicate watercolor washes, and subtle paper texture. {argument name="subject" default="Bill Gates"} gazes slightly upward with a thoughtful, visionary expression, wearing rectangular glasses, a crisp white shirt, and a tailored blazer.
The background is an elegant collage of innovation, technology, education, and philanthropy themes rendered in soft blue watercolor splashes and technical sketch elements. Surrounding details include a world map with interconnected nodes, circuit board patterns, scientific diagrams, an open book, a microscope, architectural sketches, stacks of books labeled “Knowledge,” “Innovation,” “Education,” and “Impact,” and modern research facilities. Faint handwritten notes, geometric constructions, and data-network graphics are integrated throughout the composition.
Large refined serif typography at the top displays “{argument name="name" default="BILL GATES"},” accompanied by an inspirational quote in classic editorial layout. The entire poster is executed in a sophisticated blue-and-white palette, combining watercolor painting, architectural drawing, scientific illustration, and luxury magazine cover aesthetics. Ultra-clean composition, museum-quality artwork, soft atmospheric splashes, intricate details, premium paper texture, intellectual and inspirational mood, editorial biography poster, award-winning design, 8K resolution, vertical 3:4 aspect ratio.
Style Keywords: watercolor illustration, ink sketch, monochromatic blue palette, editorial poster, visionary leader portrait, scientific collage, philanthropy theme, educational artwork, luxury print design, architectural line art, highly detailed, elegant typography, premium textured paper.

Create a clean educational {argument name="expression grid" default="FACS Action Unit expression grid"} featuring a {argument name="subject" default="realistic adult female character"}. Minimal studio lighting, {argument name="background" default="neutral white background"}, professional facial anatomy reference sheet aesthetic, realistic skin texture, consistent

Generate four DIY baby toy concept sheet images using the theme '{argument name="theme" default="plastic bottle"}'.

Important:
Generate the four images as separate images. Depict only one toy idea per image. Do not include multiple ideas in one image. Grid, collage, 2x2 splits, or comparison table layouts are prohibited.

This is not a commercial baby toy design; it is a DIY toy made from household materials. Do not make it look like a mass-produced product, a luxury toy with special molds, or industrial design renders.

Specific to 'plastic bottles': Use actual empty bottles. Show the original bottle shape, transparent body, existing caps, and signs of homemade construction like masking tape. It should look like something made with materials from a home or dollar store.

The four images should show toys with different play styles:
1st: Sound-focused DIY toy
2nd: Rolling/chasing DIY toy
3rd: Seeing/observing DIY toy
4th: Touching/sensory DIY toy

Each sheet must include: Main visual, baby playing scene, materials list, 3-step assembly diagram, structural diagram, safety diagrams (sealing/fixing), play style icons, and developmental benefits.

Safety: Target age 6–18 months. Must show parental supervision. Small materials must be sealed inside. No loose small parts on the exterior. No choking hazards, sharp parts, or long strings. No scenes of babies biting/licking.

Visual Style: White background, natural light, recipe-sheet style. Realistic and reproducible. Cute, clean, and friendly. Minimal text, centered on diagrams and icons.

Negative Prompt: commercial baby toy, premium product design, luxury product rendering, silicone molded toy, custom molded parts, factory-made toy, overmolded rubber, special cap, one-piece molded design, BPA-free product label, food-grade silicone label, industrial design board, polished manufactured product, baby chewing toy, baby biting toy, grid layout, collage layout, 2x2 layout, multiple ideas in one image

FACIAL REFERENCE: Use provided photos as absolute highest priority.
The viewer must instantly recognize the subject without any doubt.
Lock in: exact hairstyle with natural volume, precise facial
proportions, eye shape and color, natural beard shadow,

FACIAL REFERENCE: Reference photos are the absolute supreme
priority. The subject's face is photorealistic and completely
specific to the references. Lock in: exact dark voluminous
hair style, real facial proportions, authentic eye shape,
natural light

Create a cinematic widescreen sci-fi dystopian landscape in the style of a serious film still. A lone {argument name="female character" default="short-haired woman in a dark teal long dress or tunic with trousers"} walks across a vast muddy, barren plain, holding a small dark case or tool bag at her side. In the immediate foreground, place exactly 1 motionless human body lying horizontally on the wet ground, mostly in shadow. In the left midground, include exactly 1 pale vintage sedan parked or slowly driving across the flats, angled slightly away. The background should show a hazy futuristic industrial city with exactly 2 large elevated bridge or viaduct structures spanning from the left and right sides toward the skyline, plus bulky cylindrical and rectangular towers softened by smog. Use a muted teal-gray and dusty beige color palette, low contrast, atmospheric haze, wet reflective patches in the mud, and a bleak post-apocalyptic mood. Composition: 16:9 aspect ratio, wide establishing shot, horizon high in the frame, the woman positioned slightly left of center, the car farther left, the skyline distant and desaturated. Lighting should be overcast and backlit through polluted air, with realistic film grain, anamorphic cinematic framing, and no visible text, logos, or UI elements.

Goal: Create a premium square social-media showcase poster about turning everyday objects into tactile 3D fabric art, using {argument name="headline text" default="PROMPT SHARE"} as the main caption and {argument name="signature text" default="Samann_AI"} as the creator tag.

Canvas: Square 1:1 composition, clean white background, rounded-corner gallery layout. Use a soft editorial product-design look with realistic cloth, burlap, felt, wool, tweed, denim, suede, leather, and brushed copper accents. Lighting is soft and directional with realistic shadows, giving every element a raised bas-relief, handcrafted textile sculpture feel.

Layout: Arrange exactly 4 large rounded-square image panels in a 2 by 2 grid with narrow white gutters. Each panel has a muted sage-green or olive-gray backdrop and contains a centered 3D fabric collage object. Add a bottom branding strip beneath the grid: on the left, large condensed blue-gray uppercase text reading {argument name="headline text" default="PROMPT SHARE"}; directly below or overlapping it, italic gray script text reading "chat GPT 2.0". On the right, place a long rounded light-gray pill containing spaced blue-gray letters reading {argument name="signature text" default="Samann_AI"}, plus a small circular profile avatar at the far right.

Four panel contents, exactly 4:
1. Top-left panel: a stylized male business portrait built from layered fabric rectangles and rods. The figure wears a navy suit jacket, white shirt, and red woven tie. Gray-beige sculpted cloth hair sits above the head. The face area is intentionally covered by a flat square brown fabric patch, like an anonymized placeholder. Surround the portrait with stacked tan, brown, navy, and charcoal textile blocks, thin copper rods, and small round brass disks.
2. Top-right panel: a circular BMW-inspired emblem made entirely of fabric and leather-like materials. Show a black outer ring with raised cream letters "B", "M", and "W" spaced around the top. In the center, include the classic quartered circle in beige and dark blue fabric. Surround it with rectangular burlap blocks, dark vertical panels, thin copper rods, small brass knobs, and layered geometric backing shapes.
3. Bottom-left panel: a cozy modern sofa scene rendered as a fabric wall sculpture. Show one beige upholstered couch with rounded arms, exactly 2 pillows on it, a slim dark wooden shelf or tabletop line beneath the seat, and many stacked textile rectangles behind and below it. Use tan, cream, brown, taupe, and gray fabrics, plus a large rounded brown semicircle on the left, small bead-like decorations, brass disks, and thin copper rods.
4. Bottom-right panel: a second stylized portrait, this time an older person with voluminous curly white-gray yarn hair and a gray tweed coat or jacket. The face is intentionally hidden by a flat rectangular taupe-brown fabric patch. Build the background from dense layered squares and rectangles of brown, beige, gray, and charcoal textiles, with small brass circles and vertical copper pins.

Visual style: Ultra-detailed 3D textile craft, realistic woven fiber texture, layered collage depth, warm earthy palette with navy-blue accents, handmade luxury mood, crisp edges, rounded panel corners, subtle ambient occlusion, high-end Behance-style design presentation.

Constraints: Keep exactly 4 main image panels and exactly 1 bottom branding strip. Do not add extra panels. Make all objects look physically made from fabric and thread rather than flat digital illustration. Preserve the visible text "BMW", "chat GPT 2.0", the headline, and the creator tag. Avoid photoreal human skin; portraits should remain abstract textile constructions with the face covered by plain fabric patches.

Goal: Create a vertical encyclopedia-style atlas card for {argument name="animal topic" default="British Shorthair cat"}, designed like a polished Chinese science-popularization infographic.

Canvas: Portrait 3:4 layout, warm ivory background, soft sage-green borders, rounded modular cards, subtle shadows, clean editorial spacing, high-resolution print quality.

Header: At the top left include a tiny book icon and the small atlas label in Chinese inside this parameter: {argument name="atlas label" default="百科科普图鉴 CAT BREED ATLAS"}. Add a large dark-green Chinese headline using this parameter: {argument name="headline text" default="英国短毛猫"}. Under it place the English subtitle “British Shorthair” and a short Chinese descriptor line. Add three small pill tags with icons below the subtitle.

Main subject: Center-right dominant realistic studio photo of a British Shorthair cat: plush blue-gray coat, round face, dense fur, copper-gold eyes, upright ears, sitting front-facing with tail curled to the right. Background behind the cat is softly blurred home interior in beige tones.

Layout: Use exactly 9 labeled information modules, each with a green circular letter badge: A through I. Arrange A, B, C, D, E around the upper and middle area, F, G, H beneath them, and I as a full-width bottom feature strip. Each module is a rounded rectangle with pale cream fill and thin green outline.

Section count and contents: Include exactly these 9 modules: A Basic profile, B Appearance features, C Personality / behavior, D Feeding and care, E Risk and precautions, F Suitable owners, G Pros and cons comparison, H Quick rating, I Top 5 keywords. Use concise Chinese body copy and tiny icons, but keep the layout readable and not overcrowded.

Detail callouts: Add exactly 5 circular zoom-in detail callouts connected to the cat by thin leader lines: 1 large round eyes, 2 plush coat texture, 3 thick rounded tail, 4 dense chest / neck fur, 5 round paws / compact body feature. Each callout should contain a cropped macro image and a short Chinese label with a paragraph-length note.

Visible repeated elements: In section A include exactly 4 small coat-color cat head icons in a row. In section E include exactly 5 warning rows with red triangular warning icons. In section F include exactly 3 owner-suitability rows with green check icons and exactly 2 unsuitable rows with orange X icons. In section G include exactly 2 comparison columns, one green “advantages” column and one orange “disadvantages” column. In section H include exactly 6 rating rows, each with blue star ratings. In section I include exactly 5 keyword cards, numbered 1 to 5, each with a small icon: smiley face, male symbol, fish, small circular symbol, and heart.

Bottom photo strip: Along the bottom right, place exactly 4 small rounded photos of the cat showing lifestyle / posture scenes: sitting front view, standing side view, lying relaxed, curled up sleeping in a bed. Add small Chinese captions under each photo.

Visual style: Professional encyclopedia card, soft educational design, muted green and beige palette, realistic animal photography mixed with vector UI icons, thin dividers, rounded corners, compact but clean Chinese typography. Avoid neon colors, clutter, watermark, QR codes, or extra decorative text.

Goal: Create a vertical modular encyclopedia card about {argument name="animal breed" default="Golden Retriever"}, designed like a polished Chinese science-popularization pet profile poster.

Canvas: Tall portrait infographic, about 9:16 aspect ratio, warm off-white background, soft beige and cream rounded cards, subtle shadows, plenty of white space, clean grid alignment, high-resolution print-ready look.

Main layout: Place a large realistic full-body golden retriever in the center-left, standing and facing forward, happy open mouth, visible tongue, fluffy golden coat, natural paws, soft studio lighting. At the top-left, use a small brand/tag pill, then a large bold Chinese headline for the breed name and a short subtitle meaning “gentle and friendly family companion.” Add the English text “GOLDEN RETRIEVER” below the subtitle in small uppercase letters.

Top area elements: Include exactly 3 compact information blocks around the title and dog: 1 small breed-origin/specification card on the left with multiple short rows of stats, 1 small popularity or rating card at the top-right with star rating, and 1 small introduction card below the left stats with brief descriptive text.

Right detail column: Create exactly 5 stacked rounded detail cards, each with a circular close-up photo and short Chinese explanatory text. The 5 close-ups are: 1 eye/face detail, 2 ear or muzzle detail, 3 coat/fur detail, 4 tail or body coat detail, 5 paw pad detail. Keep these cards aligned vertically on the right side of the dog.

Middle and lower modules: Use a neat dashboard of exactly 10 rounded rectangular content cards below and around the dog. The 10 cards are: 1 physical characteristics card with a small side-view dog silhouette and bullet text, 2 personality traits card with five small line icons, 3 ability/rating card with exactly 5 rows of star ratings, 4 lifestyle habits card with four icon bullets, 5 care advice card with multiple short bullet rows, 6 safety or warning card with triangle alert icons, 7 strengths and weaknesses comparison card split into green “advantages” and red “disadvantages” areas, 8 suitable owners card with five icon bullets, 9 Top 5 key facts card with exactly five numbered orange circles, 10 small bonus profile/knowledge card if needed to balance the grid.

Visual style: Professional encyclopedia-card aesthetic, clean Chinese editorial layout, soft warm palette of cream, beige, pale gold, muted orange, light brown, and gentle green/red accents. Use thin line icons, circular photo crops, small gold star ratings, rounded corners, subtle drop shadows, and consistent margins. The central dog should be the most visually dominant element; all text should look dense but organized.

Text content: Use Chinese UI-style labels and short Chinese paragraphs throughout, with the main title set to {argument name="headline text" default="金毛犬"}. Use a warm subtitle set to {argument name="subtitle text" default="温暖友善的家庭伴侣"}. Keep text legible-looking but do not overcrowd; use placeholder-like short Chinese bullet lines where tiny text is not readable.

Customization: Set the overall palette to {argument name="color palette" default="warm cream, beige, golden brown, muted orange"} and the illustration/photo style to {argument name="visual style" default="realistic pet photography combined with clean infographic design"}.

Constraints: Exactly one main dog, exactly 5 right-column close-up detail cards, exactly 10 lower/middle information cards, exactly 5 rows in the star-rating card, exactly 5 numbered items in the Top 5 card. No people, no watermark, no messy background, no extra animals.

Goal: Create a clean brand exploration board for a fictional SaaS landing-page builder called {argument name="brand name" default="Neito"}, showing the same website and brand identity system repeated in exactly 4 color directions.

Canvas: Wide 16:9 white canvas, arranged as a 2×2 grid of four complete brand/website mockup panels with generous white space. Each panel is nearly identical except for the accent color and soft blurred gradient decorations.

Layout: Each of the 4 panels contains exactly 2 main areas: a large landing-page mockup on the left and a narrow brand-guide column on the right. The four accent themes are: 1 blue, 2 green, 3 orange, 4 purple.

Landing-page mockup details: At the top left place the logo mark as two forward slashes followed by “Neito” with a small star/spark symbol. Across the top navigation place exactly 5 items: Product, Templates, Pricing, About, and a black rounded “Get started” button. Center a large editorial serif headline: “Words become landing pages.” Below it add small gray supporting copy about turning ideas and content into beautiful ready-to-share landing pages, then a black pill button reading “Start for free”. Under the button place exactly 3 small feature chips with icons: “No code”, “Quick setup”, and “Always in sync”. Behind the hero headline use a huge airy blurred circular gradient ring in the panel’s accent color.

Bottom website cards: Under each hero section place exactly 3 rectangular content cards with thin borders and rounded corners. Card 1 reads “Build pages from your ideas.” with a small black pill button “Try Neito”. Card 2 reads “Launching soon.” and includes a small rounded email/input pill plus a black “Notify me” button and a thin abstract circle-line graphic. Card 3 is numbered “01” and reads “Type your thoughts.” with small descriptive text and a large cropped blurred accent-color ring in the lower right.

Brand-guide column: Add a small heading “BRAND IDEA” and a short Japanese-style paragraph block in small black text. Below it create exactly 4 labeled sections separated by thin gray horizontal rules: COLOR PALETTE, TYPOGRAPHY, VISUAL LANGUAGE, and a sample card. In COLOR PALETTE show exactly 2 circular swatches labeled Main and Accent; the Main swatch is near-white and the Accent swatch matches the panel color, with a small hex-code label beside each. In TYPOGRAPHY show exactly 2 type specimens: a large serif “Aa” labeled for headlines and a smaller sans-serif “Aa” labeled for body text. In VISUAL LANGUAGE show exactly 4 circular tokens labeled Flow, Connection, Simplicity, and Flex, with one or more using the accent color as a blurred ring. The final sample card reads “From thought to landing page.” with small subtext and a cropped blurred accent-color ring.

Visual style: Minimal premium web-design presentation, monochrome typography, lots of white space, thin light-gray dividers, subtle shadows, soft frosted gradients, editorial serif headline paired with tiny modern sans-serif UI labels. Keep everything crisp like a Figma brand exploration screenshot. Use {argument name="headline text" default="Words become landing pages."} as the main repeated hero headline, {argument name="accent palette" default="blue, green, orange, purple"} for the four color variants, and {argument name="canvas ratio" default="16:9"}.

Constraints: Show exactly 4 panels, exactly 3 bottom website cards per panel, exactly 2 brand areas per panel, exactly 2 color swatches per brand guide, exactly 2 typography specimens per brand guide, and exactly 4 visual-language tokens per brand guide. Do not add photos, people, device frames, dark backgrounds, or extra panels.

Using REFERENCE_0 as the base image, transform the scene into a polished Japanese anime illustration while preserving the same vertical composition, seated side pose, table placement, mug-holding gesture, open book, chair, jars, and bookshelf arrangement. Keep the face anonymized with the same centered square blur/censor block.

Main transformation: Replace the real person with an elegant anime girl, keeping the pose and hand positions, but change her into a stylized character with {argument name="hair color" default="short glossy blue hair with turquoise tips"} and a refined dark idol-style outfit. The outfit should include exactly 5 prominent visible costume elements: a sleeveless navy dress, black lace shoulder trim, a white high-collar front panel, a black ribbon bow at the neck, and a white apron-like side panel with gold edging.

Style and lighting: Convert the photo into clean high-quality anime line art with smooth cel shading, warm beige tones, crisp highlights, and a calm cozy reading atmosphere. Add strong afternoon sunlight entering from the left, forming diagonal light-and-shadow bands on the wall.

Background: Keep the simple room, wooden desk, open book, white mug, glass jar, small wooden box, chair, and right-side bookshelf, but render them as anime background art with warm wood tones and neatly simplified details. The bookshelf should remain on the right edge and contain multiple neutral-colored books without readable text.

Constraints: Do not change the camera angle or crop. Do not add extra characters. Avoid photorealism. Keep the anonymous square face block visible and centered over the face.

Using REFERENCE_0 as the pose and composition base, transform the seated woman into a polished anime-style character portrait. Keep the face anonymized with a large soft rectangular blur block, but change the framing from a side-view café/table scene into a close-up front-facing bust portrait. Replace the original realistic indoor room, table, mug, book, and chair with a bright leafy indoor garden/window background. Restyle the character with {argument name="hair color" default="pale icy blue"} short bob hair, and dress her in an ornate {argument name="outfit style" default="black, white, and silver gothic idol-style sleeveless blouse"} with lace shoulder details, ribbon tie, decorative buttons, and structured paneling. Use crisp high-quality Japanese anime illustration style with clean line art, soft lighting, vivid greenery, shallow depth of field, and a vertical smartphone composition. Preserve the calm seated mood from the reference while making the final image look like a finished character illustration rather than a photo.

Create a full-body vertical fashion illustration of an elegant young woman in an ornate Japanese festival kimono, standing in a graceful three-quarter back pose and looking over her shoulder, with her face intentionally left featureless as a smooth soft beige blank mask. The character wears {argument name="kimono colors" default="deep emerald green, teal, ivory, and metallic gold"} with lavish botanical patterns, gold filigree, and white blossom motifs. Show exactly 6 main costume elements: an elaborate floral hair ornament with teal ribbons and dangling gold chains, a structured high collar, a layered wide-sleeved kimono robe, a large sculptural obi bow at the back, translucent gauze-like trailing panels, and a long floor-sweeping train. The hairstyle is glossy dark hair pinned up in an ornate bun with loose wisps. The kimono should feel luxurious and ceremonial, with oversized sleeves, crisp folded fabric edges, transparent layered fabric, gold tassels, and embroidered flowers. Surround her with a delicate botanical atmosphere: floating teal and gold leaves, pale cream flowers, and thin branches growing from the lower left and lower right edges. Use {argument name="art style" default="high-detail anime fashion illustration with watercolor softness and Art Nouveau elegance"}, refined linework, luminous cream background, warm golden lighting, and a vertical 2:3 composition. Emphasize intricate textile detail, flowing fabric, graceful posture, and a dreamy early-morning festival mood. No text, no watermark, no modern accessories, no background buildings.

Create a highly detailed vertical anime key visual of {argument name="character name" default="a petite fairy-like biker girl"} riding a futuristic yellow hoverbike through the sky above a city. Use an extreme dynamic low-angle perspective with fisheye / tiny-planet distortion: the motorcycle and rider rush toward the viewer, the front wheel huge in the lower left foreground, the rider’s heavy boot thrust toward the camera in the lower center, and the city below curves into a circular vortex-like panorama. The character is a young anime girl with short tousled dark brown hair, golden eyes, a focused determined expression, and light freckles. She wears an oversized orange hooded jacket with long fluttering fabric strips, large brass-and-teal aviator goggles on her hood, black gloves, dark shorts, thigh-high stockings, and bulky mechanical boots with cyan glowing accents. Give her translucent iridescent insect/fairy wings, exactly two large wing pairs visible behind her, shimmering with rainbow colors and thin dark veins. The hoverbike is a compact cyberpunk sport motorcycle in glossy yellow-orange armor with exposed black mechanical parts, a clear tinted windscreen, cyan glowing headlights, cyan energy rings in the wheels, and small turquoise light details throughout. The background is a bright blue sky filled with scattered white clouds, warm sunlight from the left, and a dense modern coastal city far below wrapping around the circular horizon; include hints of ocean near the horizon. Visual style: crisp polished Japanese anime illustration, sharp linework, rich mechanical detail, cinematic lighting, high saturation, sparkling highlights, motion energy, no blur on the main character, ultra-clean rendering suitable for GPT image generation. Constraints: no text, no logo, no watermark, keep the character and bike sharply defined, emphasize clarity and beautiful detail for the rider and motorcycle.

Goal: Create an extremely dense anime fan-art collage sheet celebrating one cute schoolgirl character, like a chaotic “favorite character appreciation” page covered in doodles, hearts, sparkles, handwritten praise, and many repeated poses of the same girl.

Canvas: Vertical 3:4 illustration, white paper background, edge-to-edge composition with no empty space. Use a warm hand-drawn manga sketch style, soft watercolor markers, brown ink outlines, pink accent lettering, and energetic scribbled annotations.

Subject details: The repeated character is {argument name="character name" default="a cheerful short-haired schoolgirl"}. She has a fluffy short brown bob with bangs, small gold hairpins, large amber-brown eyes, rosy cheeks, and a lively innocent expression. Outfit is a beige oversized cardigan over a white lace-collar blouse, blue ribbon tie, blue plaid pleated skirt, white knee socks, and brown loafers. Keep the color palette consistent: beige, navy blue, warm brown, cream, gold, pink, and black sketch lines.

Layout and counted elements: Include exactly 27 visible depictions of the same character plus 2 extra detail panels. Arrange them as overlapping sticker-like cutouts with white borders and diagonal angles. The 27 character depictions are: 1 very large cropped face at upper left with a rectangular soft brown blur covering the face; 1 small excited upper-center bust with open mouth; 1 small shy upper-left-center bust with blush; 1 upper-center pointing pose with a soft brown blurred face; 1 large seated pose at upper right with a soft brown blurred face; 1 laughing upper-middle-right bust; 1 tiny angry chibi on the left; 1 tiny chibi inside a thought bubble; 1 very large central torso pose reaching toward the viewer with a soft brown blurred face; 1 central peace-sign pose with bright eyes; 1 middle-right clasped-hands pose with a soft brown blurred face; 1 right-middle kneeling full-body pose; 1 tiny right-side chibi shouting; 1 small lower-left head with a soft brown blurred face; 1 lower-left chibi yelling; 1 lower-left crying chibi; 1 lower-center full-body pose reaching an open palm toward the viewer; 1 tiny chibi beside that hand; 1 lower-right standing bashful full-body pose; 1 bottom-left happy chibi; 1 bottom-center crying chibi; 1 bottom-center excited chibi; 1 bottom-center full-body standing pose with a soft brown blurred face; 1 bottom-right tiny chibi; 1 bottom-right cropped large face with a soft brown blurred face; 1 small flustered chibi near the lower-middle notes; and 1 tiny smiling chibi near the lower-right margin. The 2 extra detail panels are: 1 close-up pair of sparkling amber eyes on the left side and 1 small lower-left inset showing a hand or folded-paper detail.

Text content: Fill the page with handwritten Japanese-style fan notes, exclamation marks, arrows, hearts, and sparkles. The two biggest pink phrases should be {argument name="main fan phrase" default="大好き!!!!!"} across the lower-left center and {argument name="divine praise phrase" default="神!!"} on the lower-right. Add many smaller handwritten annotations such as cute, precious, angelic, too lovable, best, adorable, and emotional reactions, mostly in black, pink, and purple ink. Text should look hand-drawn, uneven, and integrated into the artwork.

Visual style: Energetic shoujo manga illustration, expressive linework, soft cel shading, watercolor texture, sticker collage layering, exaggerated hands in the foreground, cute chibi variations, blush marks, hearts, stars, speech bubbles, speed lines, and tiny sound-effect scribbles. Keep the background white but visually packed with annotations.

Constraints: No realistic photography, no 3D rendering, no clean corporate layout, no empty background. Preserve the dense fan-collage feeling and the repeated-character count. Several larger faces should be covered by smooth rectangular brown blur blocks as part of the final image.

Create a highly detailed fantasy patisserie product shot of a single round whole cake themed after {argument name="character name" default="Nozomu"}, presented as if an original anime character has been transformed into an official celebration cake. The cake is one large cylindrical tier on a white porcelain plate with gold trim, decorated in pale icy blue, lavender, cream, pearl white, and metallic gold. Include exactly two character depictions: one three-dimensional anime-girl cake topper emerging from the top center, and one oval framed portrait on the front side of the cake. The character has {argument name="hair color" default="long pale blonde hair"}, a blue-lavender witch or magician aesthetic, an oversized ornate wide-brim hat covered with blue and pink flowers, gold filigree, and a corset-style dark blue dress with puffed off-shoulder sleeves; keep the face softly hidden or covered by a plain skin-toned rectangular mask so the image reads as a customizable character-template cake. On the cake top, surround the topper with dense piped buttercream rosettes, shell swirls, pearl sugar beads, glossy blue gem candies, crystal-like ice shards, crescent-moon glass decorations, and many small flowers in blue, white, lavender, and blush pink. At the front top center place an ornate cream plaque with raised gold baroque trim and the visible Japanese name text {argument name="plaque text" default="のぞむ"} in elegant brown-gold lettering. On the side of the cake, create pale blue gradient icing with sparkling sugar speckles, gold scrollwork, dangling blue jewel accents, white piped borders, and an elaborate oval gold frame containing the second portrait of the same character, also with the face covered by a skin-toned rectangular mask. Include exactly three warm lanterns in the background: one blurred tall lantern on the left, one sharp antique streetlamp-style lantern just behind the cake on the right, and one blurred hanging lantern in the upper right. Set the scene in a dreamy nighttime glass-garden or magical patisserie environment with soft bokeh lights, reflections, floating petals, shallow depth of field, luxurious macro detail, luminous highlights, and a romantic fairytale atmosphere. Use a vertical 4:5 composition, centered cake, photorealistic 3D anime illustration quality, ultra intricate frosting texture, sparkling jewels, and no extra words besides the plaque text.

Goal: Create an information-dense anime character appreciation sheet filled edge-to-edge with expressive drawings of one adorable gothic-lolita girl, as if made by an obsessed fan. The overall feeling is cute overload, frantic admiration, and high-detail character design notes.

Canvas: Vertical 3:4 portrait canvas, white paper background, dense collage layout with no borders. Use a hand-drawn sketchbook aesthetic with watercolor/marker rendering, ink outlines, soft gray shading, pale pink hearts, blue accent hearts, sparkles, arrows, exclamation marks, and handwritten Japanese praise notes scattered everywhere.

Main character: A young anime girl named {argument name="character name" default="unnamed gothic-lolita girl"} with {argument name="hair color" default="glossy black"} long wavy twin-tail or half-up hair, thick bangs, soft blush, large pinkish eyes, and a delicate doll-like face. Her face should be intentionally left as a soft blurred/obscured oval in several larger portraits, while smaller detail drawings may show expressive eyes and mouth. She wears a black frilly gothic-lolita dress with pale blue ribbon accents, lace trim, puff sleeves, layered skirt, heart charm details, white frilled socks, and chunky black Mary Jane platform shoes. Main accent color is {argument name="accent ribbon color" default="powder blue"}. Mood is overwhelmingly affectionate and cute.

Layout: Place exactly 20 discrete character/detail illustrations around the canvas, overlapping slightly but readable. Count and include these 20 visible elements: 1 central full-body standing pose holding an open white folding fan; 2 large upper-left bust portrait with blurred face; 3 small top-center head-and-shoulders portrait with blurred face; 4 top-right close-up of elegant hands with black lace cuffs and pale nails; 5 upper-right shy bust portrait with fingers near mouth and blurred face; 6 upper-right close-up of one sparkling pink eye; 7 left-middle chibi sitting while hugging a pink plush toy; 8 left-middle side-profile bust showing long hair and blue bow; 9 lower-left close-up of one pink eye; 10 lower-left close-up of a raised hand gesture with cuff; 11 lower-center small bust focused on hair bow and chest ribbon with blurred face; 12 bottom-left large cropped bust portrait with blurred face; 13 bottom-middle large cropped bust with hand near cheek and blurred face; 14 right-middle close-up of the bodice ribbon and waist bow; 15 small right-middle crying chibi head; 16 small right-lower crying chibi head with open mouth; 17 lower-right large shy bust with both hands covering cheeks and blurred face; 18 lower-center-right kneeling or seated half-body pose with hand near lips, visible face, looking bashful; 19 bottom-center back-view hair and outfit detail showing the rear hair ribbons and dress back; 20 bottom-right tiny blushing chibi face with tears.

Handwritten text content: Add many casual Japanese handwritten fan notes, keeping them decorative and energetic. Include a huge top-left headline reading {argument name="headline text" default="なのかが 世界一かわいい...!!!"}. Also include scattered notes similar to: 「存在が…尊い…!!」, 「この手が好きすぎる…」, 「天使だよ…」, 「ビジュ強すぎ問題」, 「大好き」, 「全身かわいすぎて無理…!!!」, 「笑顔…最高!!!」, 「守ってあげたい…!!!」, 「横顔も好きすぎる!!!」, 「リボンもフリルも完璧すぎる…」, 「ハートの金具が天才的…!!!」, 「もうほんとに好きすぎるっ!!!」, 「だいすき!!!」, 「尊すぎて語彙力消失!!!」, 「可愛さ大洪水…!!!」, 「なのかしか勝たん!!!」. Write them loosely by hand in black ink with pink highlights, arrows, and hearts.

Visual style: Highly polished anime illustration mixed with rough handwritten annotation energy; detailed shiny hair strands, lace, ruffles, bows, and accessories; soft blush tones; expressive chibi faces; pink and blue hearts throughout. The page should feel packed, chaotic, affectionate, and cute, but still balanced around the central full-body figure.

Constraints: No photo realism, no 3D render, no clean corporate layout, no empty background, no watermark. Keep the collage dense and count exactly 20 illustrated elements.

Goal: Create a square anime-style social media illustration about an AI image generation course, showing a business presentation scene with a large Japanese text slide.

Canvas: 1:1 square image, clean polished digital anime illustration, medium-wide shot in a meeting room or classroom. Warm beige walls, wooden conference table in the foreground, subtle line art and soft shading.

Main subject: On the left side, draw one adult Japanese businesswoman presenter, waist-up to thigh-up, wearing a navy blazer, white button-up shirt, and gray skirt or trousers. She has dark brown hair tied in a low ponytail. Her face is intentionally covered by a rectangular soft blur/mosaic privacy block. She stands confidently with one hand lowered and the other arm extended toward the presentation screen.

Foreground objects: On the wooden table, place exactly 2 objects: one clear glass of water near the center-left and one open dark laptop angled on the right.

Presentation screen: On the right side, place a large white projection screen with a black frame and stand. The screen contains a bold Japanese headline in thick black type, centered and stacked. Use this exact text: {argument name="slide headline text" default="独学のAI画像生成が\n「なんとなく」で終わっていた私が、\n10万文字の「ChatGPT画像生成の教科書」\nで全部見直した話"}. Make the text readable and dominant.

Screen graphics: Include exactly 7 decorative/content elements on the slide: 1 yellow lightbulb icon at the top left, 1 framed landscape picture icon near the top center, 1 green ChatGPT logo icon at the top right with 2 small sparkle marks beside it, 1 small confused chibi woman in a blue thought cloud at the lower left with a question mark, 1 blue arrow pointing right in the lower center, 1 green booklet/tablet labeled {argument name="book title" default="ChatGPT\n画像生成の\n教科書"}, and 1 small confident chibi woman at the lower right raising one finger with a red exclamation mark beside her. Both chibi faces should also have soft rectangular privacy blur blocks.

Style: Japanese business manga / anime explainer thumbnail style, crisp ink outlines, soft gradients, professional yet approachable, high contrast readable slide typography, muted corporate colors with small yellow sparkle accents. The overall composition should resemble an X/Twitter article thumbnail about learning practical AI image generation.

Constraints: Use exactly 1 main presenter, exactly 2 table objects, exactly 7 slide graphic elements as listed, and 3 blurred faces total. Do not add extra people, extra charts, watermarks, logos outside the slide, or additional text beyond the specified Japanese slide and booklet labels.

Create a cinematic widescreen sports poster for {argument name="event title" default="FIFA WORLD CUP 26"}. Show a packed night stadium viewed from field level with dramatic curved architecture, thousands of glowing spectators, red and gold pyrotechnic streaks, bright stadium floodlights, and a dusk-to-night sky over a New York–style skyline with One World Trade Center centered in the background. In the middle foreground, place a male soccer player seen from behind, wearing a white kit with gold trim and the number {argument name="player number" default="10"} on his back, standing with feet apart and both arms raised triumphantly toward a floating golden World Cup trophy above him. Put a detailed soccer ball on the grass in the lower left foreground. Add exactly three large waving national flags on tall poles across the upper stadium: United States flag on the left, Canada flag left of center, and Mexico flag on the right. Use ultra-detailed photorealistic fantasy sports advertising style, HDR contrast, golden rim lighting, lens flares, sparks, confetti, epic scale, sharp focus, and a red-gold-blue color palette. At the bottom, add bold metallic 3D gold typography reading {argument name="headline text" default="FIFA WORLD CUP 26"}; beneath it add smaller text reading "USA CANADA MEXICO" with USA in red, CANADA in white, MEXICO in green, separated by small gold dots; below that add the date line {argument name="date line" default="JUNE 11 - JULY 19"}. Keep the composition symmetrical and heroic, with the trophy, player, skyline tower, and headline aligned on the vertical center. No extra logos, no watermark, no additional text.

Using REFERENCE_0 as the visual style and layout inspiration, create a single focused carousel slide for {argument name="AI tool" default="Claude Opus 4.8"} instead of the six-tool overview. Keep the same premium black background, subtle technical grid texture, glossy 3D icon rendering, dramatic studio lighting, elegant high-contrast serif typography, neon accent color, handwritten underline flourish, and bottom social-handle placement.

Transform the composition into one hero page: place the handle “@juangpt_o1” centered at the top, add a small Claude logo mark and wordmark in the top-right corner, and replace the original headline with a huge two-line title reading “Claude” in orange and “Opus 4.8” in white. Under the title, add one concise Spanish subtitle: {argument name="subtitle text" default="El mejor para programar y para agentes — hasta 1000 subagentes a la vez"}, with “1000 subagentes” emphasized in orange. Add a thin orange hand-drawn underline beneath the subtitle.

Replace the six-icon grid with exactly 1 large central glossy 3D icon: the orange Claude sunburst/asterisk logo, enlarged as the main object, standing on a dark reflective floor with orange glow, rim light, shadow, and subtle volumetric haze. Remove all other tool icons and labels from the reference.

Bottom call to action: replace the original bottom CTA with {argument name="bottom call to action" default="Programa por ti. GO!"}, keeping “GO!” in bright neon green and the rest in white. Keep “@juangpt_o1” centered at the very bottom. Use a vertical 4:5 social media poster format, crisp 4K detail, no extra logos, no added text beyond the specified items.

Goal: Create a bold illustrated travel poster for {argument name="city" default="New York City"} with an anime street-fashion heroine towering over a rainy nighttime Manhattan street, mixing photoreal urban landmarks with comic sticker doodles and punchy typography.

Canvas: Vertical poster, 2:3 aspect ratio, high-resolution, full-bleed. Dark blue dusk sky fading into warm city lights and wet reflective pavement. Energetic youth-culture advertising composition.

Layout: Left third is dominated by a stacked headline in huge condensed uppercase block letters. Center background shows the Empire State Building lit up, with Manhattan skyscrapers behind it. Right half features a full-body anime girl standing in an exaggerated low-angle heroic pose, one foot very large in the foreground. Bottom and side areas are filled with sticker-like NYC icons and doodles.

Main text: Large headline at upper left reads exactly: “YOUR NEW YORK ADVENTURE STARTS HERE”. Use white for “YOUR,” “ADVENTURE,” and “HERE,” yellow for “NEW YORK,” and hot pink for “STARTS.” Below it, handwritten script reads exactly: “Ride the subway. Hit the streets. Taste the city.” with “subway” in cyan and “city” in hot pink. Add small sticker text blocks reading exactly: “NYC IT HITS DIFFERENT,” “NYC ♥ ALL DAY. EVERY DAY. LIVE IT. LOVE IT.,” and “BIG DREAMS BIGGER CITY.” Add a round subway badge near the Empire State Building reading “NEW YORK CITY” around the rim, a large “7” in the center, and “SUBWAY” underneath.

Character details: The main character is one young anime woman, cheerful and confident, with {argument name="hair color" default="bright cyan-blue hair with magenta-pink tips"}, short bob haircut, big expressive eyes, and a small smile. She waves with her raised right hand and holds a steaming hot dog in her left hand. Outfit: oversized yellow-and-black bomber jacket with patches and graffiti marks, black cropped top, black baggy cargo pants with zipper details and pink “Brooklyn” graffiti on one leg, layered necklaces, earrings, rings, and chunky cream high-top sneakers with pink heart and black accents. Jacket includes visible pink “NYC” lettering and yellow graphic patches.

Counted visual elements: Include exactly 1 main anime girl, 1 Empire State Building, 1 steaming hot dog in her hand, 1 oversized foreground hot dog sticker at bottom left, 1 yellow taxi sticker riding a blue wave, 1 small pizza-slice sticker near the upper right, 1 large pizza-slice sticker on the right, 1 Chrysler Building doodle, 1 bearded pigeon with sunglasses and a cap at bottom right, 1 pink heart doodle above the girl, 1 yellow star near the left center, 2 pink star stickers, and 1 pink heart sticker near the “NYC it hits different” label.

Visual style: Hybrid anime illustration and gritty urban poster design; saturated cyan, yellow, hot pink, black, and cream palette; thick black ink outlines; halftone texture; hand-drawn sticker borders; subtle film grain; splattered ink and small bird silhouettes. City background should feel cinematic and semi-realistic, while the character and stickers are crisp comic/anime art.

Atmosphere: Adventurous, playful, iconic, and aspirational, like a modern NYC tourism poster for young travelers. Wet streets reflect taxi lights and skyline glow. Add steam from the hot dog and tiny radiating lines around the heart doodle.

Constraints: Keep all visible text legible and spelled correctly. Do not add logos, watermarks, QR codes, or extra paragraphs of text. Preserve the exact poster hierarchy: headline on left, character on right, Empire State Building centered behind, sticker collage around the bottom and sides.

Create a photorealistic wide-angle travel selfie scene in Paris: an adorable golden retriever dog dressed like a stylish tourist stands in the foreground on a cobblestone plaza in front of the {argument name="landmark" default="Eiffel Tower"}, wearing a black French beret and a beige knit hoodie with drawstrings and a small embroidered monogram. The dog holds a gold iPhone-style smartphone at arm’s length in its right paw, camera lenses visible, as if taking a selfie. Place a large opaque medium-brown square directly over the dog’s face, centered from forehead to muzzle, hiding the facial features while leaving the ears, beret, hoodie, paw, and phone visible. The background shows a busy tourist crowd, outdoor café seating on the left, trees, and the landmark rising dramatically behind the subject, all with realistic depth of field and slight background blur. Use bright daytime lighting, blue sky with scattered fluffy clouds, natural colors, cinematic shallow focus, detailed fur texture, realistic paws, glossy phone reflections, and a polished social-media travel-photo composition. Add a subtle small watermark reading {argument name="watermark text" default="PIERRE_PUP_PARIS"} in the bottom-right corner. No extra text, no cartoon style, no distorted anatomy.