GPT Image 2 prompts

Create a square, retro behind-the-scenes film still of {argument name="character name" default="a young East Asian woman warrior"} sitting in a low squat on worn stone steps in an old rustic courtyard or period village set. She has {argument name="hair color" default="messy black hair"} tied loosely up with flyaway strands, a pale tired face, minimal makeup, and a defiant deadpan expression while staring directly into the camera. She wears a dirty all-black ancient martial-arts rogue costume with exactly 5 visible outfit components: layered wrap robe, textured leather-like shoulder vest, dark forearm wraps, loose trousers, and worn black boots. Pose her with her left hand raised toward the camera giving the middle finger and her right hand holding exactly 1 half-eaten steamed bun or meat bun. Place exactly 1 crumpled brown paper food bag on the ground beside her right boot. The background should show blurred stone steps, beige plaster walls, dark wooden posts, and rustic set dressing, suggesting a wuxia or historical drama filming location during a break. Use muted sepia-brown color grading, soft natural daylight, shallow depth of field, realistic skin texture, candid snapshot framing, slight motion imperfection, and an aged photographic print look with visible film grain, dust specks, scratches, faded contrast, worn white rounded borders, and creased corners. No text, no watermark, no modern objects except the food bag, and keep the image gritty, natural, and unpolished rather than glossy.

Create a vertical cinematic portrait photo of a young East Asian woman in a romantic ancient Chinese hanfu style, sitting barefoot on a stone railing beside a lotus pond at sunset. She wears a flowing translucent pale pink silk hanfu with layered gauze sleeves, delicate floral embroidery, a soft peach-pink inner bodice, and long draping fabric pooling around her legs. Her {argument name="hair color" default="black"} hair is styled loosely up with wispy bangs and small pink-and-white floral hair ornaments and ribbons. She smiles gently at the camera with a soft, dreamy expression while holding one large pink lotus flower on a long green stem in both hands. The setting is a classical Chinese garden pavilion corridor: dark wooden doorway and pillar on the right, gray stone wall and railing on the left, a pond filled with green lotus leaves and scattered blossoms behind her, distant pavilion silhouette and trees across the water. Warm golden sunset light glows low on the horizon, reflecting across the pond and backlighting the sheer fabric, creating a hazy nostalgic atmosphere. Include vertical Chinese calligraphy on the right wooden pillar reading {argument name="pillar calligraphy" default="荷风送香气"}. Use an analog film look with visible grain, slight softness, warm faded colors, subtle lens flare, gentle overexposure near the sun, and a small orange film date stamp near the lower left edge reading {argument name="date stamp" default="05 21 '99"}. Composition should be full-body seated, centered slightly right, portrait orientation 9:16, intimate and elegant, with no modern objects, no extra people, no watermark, and no distorted hands or feet.

Using REFERENCE_0 as the character base, generate a 16:9 surrounding-environment 3x3 view sheet that keeps the person’s identity, outfit, pose energy, and body proportions consistent across all angles.

Goal: Show the same subject from multiple camera directions in a coherent photo studio environment, as if captured around the person.

Canvas and layout: Create one wide horizontal 16:9 image with a dark header bar and a clean 3x3 grid of exactly 9 panels separated by thin white borders. Add the centered Japanese title {argument name="title text" default="周辺景観 3x3ビュー"} at the top.

Required 9 panels and labels: Place a black rounded label with white Japanese text in the upper-left of each panel. Use exactly these 9 labeled views, arranged left to right, top to bottom: 1) 「左後方」, 2) 「背面」, 3) 「右後方」, 4) 「左側面」, 5) 「正面」, 6) 「右側面」, 7) 「左前方」, 8) 「俯瞰」, 9) 「右前方」.

Scene transformation: Expand the original reference into a realistic behind-the-scenes studio setup. Keep the beige seamless paper backdrop from the reference, but reveal the surrounding studio: light stands, softbox lights, camera on tripod, gray concrete floor, a potted plant, a step ladder, a rolling equipment cart with a laptop/monitor, cables, and cardboard boxes. These objects should appear consistently from different angles according to the viewpoint.

Subject: Keep the same man from the reference wearing the same black cap, black T-shirt with the white {argument name="shirt text" default="20A"} lettering, black pants, and black sneakers. Maintain his crouched energetic stance and expressive hands, but rotate him naturally for each labeled viewpoint. The front-view panels should show his face; the rear and side panels should correctly hide or reveal facial features based on angle.

Style: Photorealistic, high-resolution studio photography, wide-angle documentary look, consistent lighting, realistic perspective, no cartoon styling.

Constraints: The 9 views must be spatially consistent, especially front, back, left, and right. Do not add extra panels, extra labels, watermarks, or unrelated text.

Goal: Create a hyper-realistic futuristic medical engineering visualization of an exploded transparent artificial human heart, presented like a premium bioengineering product render with technical callouts.

Canvas: 4:3 horizontal image, dark laboratory background, glossy black reflective floor, cinematic rim lighting, shallow depth of field, cool blue highlights and warm amber internal glow. The heart floats centered above the floor with a soft reflection beneath it.

Main subject: A transparent biomechanical anatomical heart labeled as {argument name="device concept" default="next-generation bioengineered artificial heart"}. The central heart body is made of clear glass-like biocompatible polymer, with visible internal tubes, valves, microfluidic channels, circuit traces, miniature mechanical pumps, red and blue glowing vascular pathways, and amber illuminated electronic modules. Use realistic refraction, caustics, chrome screws, polished metal rings, and black carbon-fiber structural components.

Exploded layout: Show exactly 13 major separated component groups around the central heart: 1 central transparent heart chamber; 2 left outer clear curved shell; 3 left black carbon-fiber rib panel; 4 left inner transparent curved support layer; 5 upper-left aorta connector ring assembly; 6 upper central transparent aorta and vessel towers; 7 upper-right pulmonary artery connector ring assembly; 8 right-side cylindrical port and interface rings; 9 right black carbon-fiber rib panel; 10 right transparent circuit-interface module with amber glow; 11 right outer clear curved shell; 12 far-right small transparent cap/lens; 13 multiple small floating screws and bolts aligned around the exploded parts. Keep all parts suspended in precise horizontal layers, as if disassembled for inspection.

Technical annotations: Add thin white/gray leader lines and small uppercase sci-fi typography. Include exactly 7 visible callout labels: 1 “AORTA” with subtext “HIGH-FLOW / BIOCOMPATIBLE / POLYMER”; 2 “RIGHT ATRIUM” with subtext “MICRO-FLUIDIC / CHANNELS”; 3 “CARBON-FIBER RIB STRUCTURE” with subtext “ULTRA-LIGHT / HIGH-STRENGTH”; 4 “VENTRICLE” with subtext “BIO-MIMETIC / PUMP CHAMBER”; 5 “PULMONARY ARTERY” with subtext “PRESSURE REGULATING / VALVE”; 6 “BIO-CIRCUIT INTERFACE” with subtext “NEURAL SYNC / WIRELESS POWER / REAL-TIME MONITORING”; 7 “MICRO-FLUID CHANNELS” with subtext “NANO-SCALE / SELF-CLEANING / FLOW OPTIMIZATION”. Place these labels around the heart without covering key details.

Footer text: In the bottom left, small spaced lettering reads {argument name="left footer text" default="HUMANITY\nA STRONGER TOMORROW"}. In the bottom right, add a tiny schematic line icon and text reading {argument name="right footer text" default="BIOENGINEERING\nFOR A HEALTHIER WORLD"}.

Visual style: Ultra-detailed photorealistic 3D render, luxury industrial design, transparent medical device, carbon fiber texture, chrome hardware, glass refraction, glowing red and blue tubes, amber circuit light, high contrast, clean futuristic interface typography, technical diagram aesthetic, 8k quality.

Constraints: No people, no hands, no blood or gore, no cartoon styling, no messy background, no extra large titles. Keep the anatomy recognizable as a human heart while emphasizing transparent futuristic engineering.

Create an ultra-detailed vertical anime fantasy illustration of exactly one elegant female samurai princess, {argument name="character name" default="unnamed"}, shown in dramatic side profile at extreme close range. She has long flowing {argument name="hair color" default="pink"} hair, delicate green eyes, glossy lips, and an intense calm expression. She wears an extravagant layered {argument name="kimono color palette" default="red, black, purple, white, and gold"} floral kimono covered in sakura, chrysanthemum, brocade, tassels, beads, gold embroidery, lacquer ornaments, and ornate hairpins. Composition: a low-angle cinematic perspective with exactly one katana dominating the foreground, held diagonally from the bottom edge toward her hand near the upper center; the blade is mirror-polished and glowing, while the scabbard or spine is richly decorated with gold filigree, red flowers, black lacquer, and jewel-like details. Her near hand reaches toward the viewer beside the blade, with long manicured nails. Background: an impossibly ornate night festival palace city filled with golden lanterns, black-and-gold folding screens, sakura blossoms, hanging ornaments, bridges, and sparkling particles; include exactly one full moon, exactly one coiling golden dragon near the moon, exactly one white crane flying on the right, exactly one tiger motif on a folding screen at lower left, and three visible multi-tier pagoda towers. Style: high-resolution gpt-image-2 anime key visual, maximalist Japanese fantasy, luminous gold highlights, pink petals drifting, prismatic lens flares, sharp face detail, shallow depth of field on the foreground sword, hyper-intricate textures, jewel-like lighting, no text, no watermark.

Goal: Create a polished anime character redesign presentation sheet for a blue-themed schoolgirl mascot, combining a large central illustration with labeled design-analysis panels around it.

Canvas: Square 1:1 white presentation board, clean editorial layout, soft pale-blue accents, thin light-gray divider lines, high-resolution anime art, crisp but hand-drawn feel.

Main subject: A cute teenage anime girl named {argument name="character name" default="Blue Ribbon"} sitting in the center foreground in a shy, appealing pose, one hand against her cheek and legs folded forward. She has long vivid {argument name="hair color" default="blue"} twin-tail hair with loose flyaway strands, glossy blue eyes, soft blush, small open smile, and delicate facial features. Outfit is a navy-blue sailor school uniform with white trim, oversized pale-blue bow at the chest, pleated skirt with white frills, thigh-high blue socks with white stripes, and black loafers with small blue ribbon details. Use a monochrome blue palette with subtle skin tones and painterly highlights.

Layout: Place the large finished redesign illustration slightly left of center, occupying most of the canvas. Surround it with structured reference panels: top-left original design panel, left-column issue cards, top-right expression set, right-middle turnaround, right-lower detail closeups, and bottom-right color palette. Add small blue heart doodles and handwritten Japanese-style notes as decorative accents.

Sections and exact counted elements:
- Top-left panel: 1 original chibi-style blue character drawing, labeled in Japanese “元のデザイン” with small English subtitle “Original”.
- Left column issues: exactly 3 numbered issue cards labeled 1, 2, and 3. Card 1 shows a tiny face sketch and says the eye expression is weak. Card 2 shows a rear/silhouette sketch and says the silhouette is monotonous. Card 3 shows a clothing/pose sketch and says the pose and outfit lack charm.
- Top-right expressions: exactly 3 rounded-rectangle face closeups labeled “表情集 Expressions”: 1 smiling with closed eyes, 1 surprised with wide eyes and small open mouth, 1 sleepy or pouting sideways glance.
- Right-middle turnaround: exactly 3 full-body views labeled “三面図 Turnaround”: Front, Side, Back, showing the same outfit and long twin-tail hair from each angle.
- Right-lower details: exactly 3 square closeup panels labeled “ディテール Details”: 1 eye detail, 1 chest ribbon detail, 1 skirt frill detail.
- Bottom-right color palette: exactly 5 circular color swatches: deep blue, royal blue, light blue, very pale blue, and pale skin-pink.

Visible text style: Use neat Japanese headings with small English subtitles, including “Original”, “Issues”, “Expressions”, “Turnaround”, “Details”, and “Color Palette”. Include handwritten Japanese catchphrases around the character, such as {argument name="main handwritten note" default="もっと好きになれるわたしに"}, plus bottom notes meaning “I have loved you since then” and “Unchanged blue, cuter.” Add small English footer text: “CHARACTER REDESIGN / SAME BLUE, A BRIGHTER ME.”

Visual style: Modern Japanese anime illustration, bright glossy eyes, soft cel shading blended with watercolor-like gradients, delicate blue linework, subtle paper texture, clean UI-like character sheet design, balanced negative space, no heavy background. Make the final design look more charming, expressive, and polished than the simple original sketch.

Constraints: Keep the exact panel counts above, maintain the blue-and-white visual identity, avoid extra characters, avoid clutter, no watermark, no photorealism, no 3D rendering.

Goal: Create a clean 3D chibi fighting animation sprite sheet for a cute mascot character, suitable for making a looping GIF.

Canvas: Square 1:1 canvas, about 1200 x 1200 px, with a transparent-background preview shown as a light gray checkerboard. Arrange the sprites in an evenly spaced 4 columns by 4 rows grid, for exactly 16 separate animation frames. Leave generous padding around each character and keep all frames at a consistent scale.

Character: A toy-like glossy 3D chibi mascot with a peach-colored round onion/droplet-shaped head and a small pointed top, tiny smiling mouth, rosy cheeks, oversized black sunglasses, and the black number {argument name="forehead number" default="29"} printed on the upper forehead. The body is small and heroic, wearing a bright {argument name="suit color" default="blue"} long-sleeve bodysuit with a white belt stripe, a red scarf/capelet at the neck, a small white circuit-like emblem on the chest, red gloves or rounded fists, and shiny red boots. Use soft studio lighting, smooth plastic/vinyl toy material, rounded proportions, and a cheerful confident expression.

Sprite actions: Depict exactly 16 frames of a simple battle motion sequence, all facing mostly forward with slight 3/4 turns. Count and show these frames in order from left to right, top to bottom: 1 neutral guard stance with both fists raised; 2 wider guard stance; 3 right-arm straight punch beginning; 4 right-arm punch fully extended; 5 reset guard stance; 6 guard stance leaning slightly; 7 long straight punch across the body; 8 return to guard; 9 left knee lift preparation; 10 higher knee lift; 11 side kick fully extended to the right; 12 second extended side kick frame; 13 knee lift recovery; 14 neutral guard stance; 15 neutral guard stance variation; 16 final neutral guard stance.

Visual style: High-quality cute 3D render, Nintendo-like mascot toy aesthetic, crisp edges, saturated primary colors, soft shadows, consistent camera angle, consistent character size, no motion blur, no outlines except natural 3D shading.

Constraints: No text labels other than the forehead number. Do not add weapons, backgrounds, UI elements, frame borders, watermarks, or extra characters. Keep the checkerboard transparency preview visible behind every frame.

Using REFERENCE_0 as the architectural source, reorganize it into a cleaner, more focused Japanese technical drawing sheet for the corridor window assembly. Keep the same subject and hand-drafted architectural illustration style, but simplify the page and redraw the construction details as an inferred measured detail rather than a broad explanatory board.

Goal: Create a detailed drawing titled {argument name="main title" default="廊下 窓まわり詳細図"} with subtitle {argument name="subtitle" default="(断面図・展開図・部分詳細図)"}. Treat all dimensions as estimates from the reference/photo and note that the unit is {argument name="unit note" default="mm"}.

Layout: Use exactly 4 drawing areas: 1 large A-A section detail on the left, 1 interior elevation view at the upper right, and 2 small partial detail diagrams at the lower right. Make the left section the dominant focus and include a small key plan/section marker at the top of it.

Left main section: Redraw the corridor window section at S=1/10 with clearer construction layers and orange dimension lines. Show exterior planting and exterior ground on the left, the tall wood sash/glass opening in the center, and the interior corridor side on the right. Include labels for exterior side, interior side, ceiling, steel column, wood sash, double glass, brick sill/window base, brick exterior wall, steel angle/bracket, floor build-up, existing RC slab/foundation, GL ±0 and FL ±0. Emphasize estimated dimensions including opening height about 2100, window base about 300, brick rise 180, brick ledge width 350, exterior passage about 600, wall 100, sill/bench 350, and corridor width about 1200.

Upper-right drawing: Create an interior elevation view labeled {argument name="elevation label" default="内観立面図（廊下側） S=1/50"}. Show the corridor-side window wall with repeated vertical wooden frames, glass panes, brick/bench base, steel columns, exterior planting visible through the glass, and orange A-A section markers.

Lower-right details: Add exactly 2 partial detail diagrams labeled a and b: a. upper/head detail showing ceiling, PB substrate/paint finish, wood sash, and double glass; b. lower/sill detail showing brick finish, mortar substrate, brick support, and steel angle plus bracket, with orange dimensions 350, about 300, and 180.

Style and constraints: Maintain the reference’s clean Japanese architectural presentation: thin black linework, light gray hatching, muted wood and brick tones, orange dimensions, white background, precise leader lines, and a professional measured-drawing look. Use Japanese technical labels where shown, but avoid overcrowding. Add a bottom caution note {argument name="caution note" default="※寸法は写真からの推定です。実施設計時は現地調査のうえ決定してください。"}. Do not add unrelated photos, people in the interior, decorative icons, or extra drawing panels.

Goal: Create a cinematic VTuber teaser movie storyboard contact sheet featuring {argument name="character name" default="an original cyber idol VTuber"} as the only character, shown across exactly 9 anime cuts in a 3 by 3 grid.

Canvas: Wide 16:9 image, clean white gutters separating all panels, high-resolution anime key visual style. Use a neon cyber stage at dusk/night with glossy reflective floors, dark vertical skyscraper-like light columns, diagonal cyan and hot-pink light beams, floating pink shard confetti, and a dreamy pink-blue sky. Palette is electric cyan, pastel blue, magenta, black, and white.

Character design: A beautiful young anime idol with very long flowing {argument name="hair color" default="sky-blue hair with vivid pink streaks"}, pink-violet eyes, pale skin, and a confident emotional stage presence. Outfit: black high-neck cropped top, sheer black mesh sleeves with crisscross straps, white corset bodice with black lacing, black shorts, thigh straps/garters, and white lace-up platform boots. Hair should stream dramatically in almost every panel as if moved by wind or performance motion.

Layout and exact 9 panels:
1. Top-left: full-body front stage introduction, character standing centered in a strong idol stance before a large pale circular backlight, reflective floor, neon columns framing her.
2. Top-center: close-up singing into a black handheld microphone, mouth open mid-vocal, intense eyes, hair crossing the frame, diagonal neon lights behind.
3. Top-right: profile close-up from the shoulders up, looking upward with a quiet emotional expression, long hair flowing to the right against dark blue and pink light beams.
4. Middle-left: dramatic low-angle dynamic shot, character reaching one open hand toward the viewer, foreshortened arm and fingers, corset and shorts prominent, energetic stage background.
5. Middle-center: rear full-body shot, character facing away toward a huge glowing lavender ring hovering above the stage, symmetrical neon columns and pink clouds reflected on the floor.
6. Middle-right: three-quarter medium shot, character touching or lifting one hand near her hair/head, confident gaze toward camera, mesh sleeve and corset clearly visible.
7. Bottom-left: rear three-quarter medium shot with character glancing back over her shoulder, long hair sweeping across the panel, pink sunset clouds and silhouetted rails/columns in the background.
8. Bottom-center: extreme close-up of the eyes, pink-violet irises sharply detailed, lower face partly hidden behind a sheer black sleeve or gloved arm, hair strands across the forehead.
9. Bottom-right: another dynamic low-angle reaching shot, open hand extended toward the viewer with fingers spread, mouth slightly open as if calling out, hair and ribbons of light arcing around her.

Visual style: Polished modern Japanese anime, VTuber teaser MV concept art, crisp linework, cel shading with soft gradients, cinematic camera angles, dramatic foreshortening, glossy reflections, high contrast neon lighting, emotional and energetic. No captions, no logos, no watermark, no extra characters. Keep the grid exactly 9 panels and maintain consistent character design across every cut.

Goal: Create a clean 4x4 pixel-art sprite sheet of {argument name="character name" default="a stylish street girl"} skateboarding, suitable for animating into a GIF.

Canvas: Square canvas, white or transparent background, no borders, no text, no watermark. Arrange exactly 16 separate full-body sprites in an evenly spaced 4 columns by 4 rows grid, each sprite isolated with consistent scale and generous white space.

Visual style: High-quality retro pixel art with crisp blocky edges, limited palette, soft cel shading, cute chibi proportions, expressive face, dynamic hair motion, and a video-game character sprite-sheet feel.

Character details: A young streetwear girl with {argument name="hair color" default="long messy blonde hair"} blowing with motion, small blue eyes, peach skin, silver hoop earrings, chunky silver chain necklace, oversized dark olive bomber jacket, black cropped top, baggy black pants, and white sneakers. Keep her proportions and outfit consistent across all frames.

Skateboard details: A black skateboard with a tan wooden edge, small beige wheels, and pixelated trucks. The board angle changes with the trick, but the design remains consistent.

Sprite count and poses: Include exactly 16 sprites, ordered left to right, top to bottom:
1. Standing upright on the skateboard, rolling forward, hair streaming left.
2. Beginning to crouch on the board, knees bent, one hand low.
3. Deep crouch preparing for a trick, body leaning forward.
4. Very low crouch, face focused, one hand near the board.
5. Rolling with board slightly tilted upward, crouched and balanced.
6. Popping the board upward, front foot lifting, hair flying.
7. Midair trick with board angled diagonally, arms out for balance.
8. Higher midair trick, board steeply angled, tongue sticking out playfully.
9. Peak jump pose, arms wide, tongue out, board crossing diagonally beneath her.
10. Descending from the trick, board nearly level, arms extended.
11. Landing crouch, board angled slightly, focused expression.
12. Stabilizing after landing, low stance with arms out.
13. Low rolling landing with small motion marks beside the wheels.
14. Rising from crouch while rolling forward, side profile.
15. Upright cruising pose, side profile, relaxed expression.
16. Rear-view cruising pose, showing her back and flowing hair.

Constraints: Keep all 16 sprites the same character and scale, preserve a tidy grid layout, avoid extra props, avoid background scenery, and make the sheet readable as an animation sequence.

Using the provided reference image as the base, expand it into a wider vertical lifestyle product photo. Keep the golden retriever being sprayed and the hand steadying its body, but zoom out to show the dog sitting happily in profile with its face visible. Add a second human hand holding a blue-and-white trigger spray bottle aimed at the dog, with a fine mist visibly spraying onto the fur. Set the scene in a warm sunlit forest campsite during golden hour, adding exactly 3 major new background elements: 1 olive-green canvas tent on the right, 1 small camping table behind the dog, and 1 softly blurred forest of tall trees. Make the product feel like an outdoor pet repellent spray advertisement; put small readable label text on the bottle: {argument name="product label" default="Outdoor Repellent Spray"} and a simple brand mark: {argument name="brand name" default="NIPIP"}. Use photorealistic shallow depth of field, warm backlighting, natural colors, crisp fur detail, and realistic water droplets/mist. Do not crop tightly like the reference; show the dog, both human hands, the spray bottle, and the campsite context clearly.

Create an ultra-detailed vertical cinematic close-up portrait of {argument name="character name" default="the First Decree"}, a stern futuristic noblewoman or judge-warrior shown from the right side with only the left half of her face visible at the frame edge. She has pale porcelain skin, sharp calm eyes with deep red eye makeup, glossy {argument name="hair color" default="black hair with subtle crimson streaks"} tied into a low ponytail with a red band, loose windblown strands crossing her cheek, and a long ornate dangling gold earring made of rings, rods, and spear-like pendants. Her outfit is a luxurious black high-collar bodysuit with fine gold circuit-like filigree, layered beneath polished {argument name="armor color" default="deep crimson lacquered shoulder armor"} engraved with subtle patterns and fitted with elaborate brass-gold circular mechanisms and compass-like ornaments on the chest. The composition is an intimate cropped bust portrait, filling the frame with her ear, jawline, collar, shoulder plate, and chest ornament; the face is partially cut off on the right for dramatic tension. Use a dark futuristic cathedral or ceremonial city setting in the background, with tall blurred architectural panels and narrow bright vertical light slits behind her. Lighting should be high-contrast and elegant: cool shadows, warm rim light glinting through hair strands, tiny red-gold sparks near the hair, glossy specular reflections on red armor and metal jewelry. Style: hyperrealistic anime-inspired concept art, premium sci-fi fantasy fashion, intricate materials, shallow depth of field, dramatic editorial framing, 9:16 vertical aspect ratio, no visible text, no watermark, no extra characters.

Goal: Create a cinematic 3×3 anime key-visual collage featuring {argument name="character name" default="an unnamed silver-haired girl"} across nine emotional story panels, like a music video storyboard or visual novel ending montage.

Canvas: Wide 16:9 image, approximately 1200×675, divided by thick black gutters into exactly 9 rectangular panels arranged in 3 columns and 3 rows. Use a moody dusk-to-night palette with soft bloom, film grain, shallow depth of field, rain/snow-like particles, and high-detail anime rendering.

Main character: A young anime girl with pale skin, messy long {argument name="hair color" default="silver-white"} hair in loose twin tails with black ribbons, golden amber eyes, delicate face, and a melancholic expression. Outfit is dark streetwear: oversized black off-shoulder jacket, black camisole/straps, distressed jeans in some panels, sneakers in full-body views, and headphones in the singing panel. Keep her recognizable in all nine panels.

Layout and panel content: Use exactly 9 discrete panels.
1. Top-left: close-up from the front, the girl reaches one hand toward the viewer with soft blurred fingers in the foreground; pale sky background. Add Japanese text 「まだ、終わってない。」 with small English subtitle “Not yet. The story goes on.”
2. Top-center: side view of the girl sitting on a rooftop or balcony ledge at sunset, city skyline and antennas in the distance, orange-purple clouds. Add large handwritten Japanese text 「どこまで行けるんだろう」 and a tiny English subtitle suggesting “What a beautiful sunset.”
3. Top-right: extreme close-up of her face lying sideways, amber eye prominent, dark background on the right. Add Japanese text 「きっと、わたしはわたしでいい。」 with English subtitle “It’s okay to be me.”
4. Middle-left: profile bust shot in a rainy neon city at night, head tilted back, eyes closed, hair flowing, glowing skyscraper lights behind. Add Japanese text 「痛みだって歌になる。」 with English subtitle “Even pain can be a song.”
5. Middle-center: intense front close-up, only her eyes and upper face visible above a dark sleeve, abstract black brushstroke graffiti behind her; no large subtitle needed.
6. Middle-right: fragmented mirror scene with her reflected in multiple shards, one hand reaching toward broken glass; cool blue lighting. Add Japanese text 「いくつもの、わたし。」 with English subtitle “So many me. Still me.”
7. Bottom-left: recording studio scene, side view of her singing passionately into a large condenser microphone with pop filter, wearing black over-ear headphones. Add Japanese text 「声になりたかった。」 with English subtitle “I wanted to be a voice.”
8. Bottom-center: rear full-body shot of her walking alone down a graffiti-covered tunnel or underpass toward a bright exit, oversized jacket slipping off shoulders. Add very large white handwritten Japanese text 「また、行ける。」 plus small English subtitle “Go to the next. On to the next.”
9. Bottom-right: warm intimate close-up of her lying on a pillow or bed, looking at the viewer with a soft tired expression. Add Japanese text 「また、どこかで。」 with English subtitle “See you somewhere again.”

Typography: Mix elegant vertical Japanese Mincho-style text and expressive handwritten white brush script. English translations should be small, serif, and placed beneath or near the Japanese. Keep text legible but integrated into the cinematic image.

Visual style: Ultra-detailed modern anime illustration, emotional J-pop album cover aesthetic, soft rim light on hair, sparkling highlights in amber eyes, painterly backgrounds, realistic lighting, dramatic crops, black panel borders, subtle chromatic aberration, no logos or watermarks.

Customizable mood: Overall feeling should be {argument name="emotional tone" default="bittersweet, resilient, and quietly hopeful"}.

Create a warm cinematic anime illustration set inside a traditional Japanese tatami room at dusk. Show exactly three girls: two Japanese girls in kimono inside the room on the left, whispering conspiratorially toward a shoji sliding door, and one blonde blue-eyed foreign maid outside on the right, peeking through the translucent paper door with one finger raised to her lips. The inside girl on the far left has dark hair in a messy bun, wears a muted purple kimono, cups one hand near her mouth, and looks surprised while whispering. The second inside girl is seen mostly from behind, leaning forward in a green kimono with a floral hairpin, pressing her ear toward the wall or door as if listening. The outside maid has {argument name="hair color" default="blonde"} hair, large blue eyes, a frilled white maid headband, a black-and-white maid dress, and a slightly innocent, secretive expression. On the plaster wall above the two kimono girls, add vertical black brush-calligraphy text reading {argument name="wall calligraphy" default="壁に耳あり 障子にメアリー"}. Use a wooden shoji grid door occupying the right half of the image, with soft silhouettes and warm backlight behind the paper panels. Foreground: a low wooden table partially blurred with exactly two blue-and-white teacups, one small dish of sweets, and a blurred flower vase at the lower left. Background: muted hanging scroll and simple ikebana flowers. Style: high-quality Japanese anime key visual, painterly lighting, soft depth of field, warm amber lantern glow from the lower right, detailed wood grain, delicate fabric folds, cozy historical interior, humorous visual pun mood, no modern objects, no watermark.

Using REFERENCE_0 as the base image, perform an extremely subtle product-style edit: change only the visible fingernails on the hand holding the strawberries into a delicate {argument name="nail design" default="pale pink rose-themed manicure with tiny floral nail art and a glossy finish"}. Keep the entire composition, face, pose, dress, bonnet, bouquet, strawberries, roses, wallpaper, lighting, color palette, and the headline text {argument name="headline text" default="PINK EDEN"} as close to the reference as possible. Do not alter the character, background, typography, cropping, or overall mood. The edit should affect exactly 2 visible nail areas: the main thumb nail and the small partially visible fingernail beside it; all other details should remain unchanged.

Create a highly detailed vertical anime fantasy illustration of one ethereal angel princess named {argument name="character name" default="unnamed angel princess"} standing on ornate stone steps in a grand cathedral garden city. She has {argument name="hair color" default="long flowing platinum-blonde hair"}, pale skin, soft blue-gray eyes, a gentle wistful smile, and a delicate youthful face. Show exactly one character with two enormous ivory feathered wings extending behind her, one floral crown with white blossoms and burgundy ribbons, one hand reaching softly toward the viewer, and one hand holding a bouquet of white lilies and small red-white flowers. Dress her in an extravagant {argument name="dress color" default="deep crimson red"} lace bridal gown: layered translucent lace, embroidered floral appliqué, gold filigree accents, long flowing sleeves, a high slit revealing one leg, trailing ribbons, and a dramatic wind-swept train covered in red flowers. Include exactly two ornate high-heeled sandals with flower-and-chain ankle decorations. The setting is a romantic high-fantasy palace balcony filled with roses, lilies, hanging flower garlands, carved gothic arches, banners, distant bridges, spires, mountains, and a shining river below. Use a soft spring daylight palette, cinematic depth of field, foreground flowers blurred, petals floating in the air, intricate fabric texture, delicate feather detail, elegant composition, dreamy atmosphere, ultra-detailed Japanese light novel cover style, graceful and luxurious, no text, no watermark.

Goal: Create a cinematic 3x3 photo collage showing quiet everyday moments from the life of a Japanese high school girl, with pale colors, soft sunlight, and a nostalgic film-still atmosphere.

Canvas: Landscape collage, 1200x900 aspect ratio, divided into exactly 9 equal rectangular panels in a 3 columns by 3 rows grid. Use thin clean white gutters between panels. Overall look should feel like screenshots from an indie coming-of-age movie shot on 35mm film.

Subject: The recurring character is {argument name="main character" default="a Japanese teenage girl with long messy black hair, pale skin, delicate features, and a tired dreamy expression"}. She wears {argument name="school outfit" default="a simple white short-sleeve school blouse with a dark pleated skirt, white socks, and black loafers"}. Her mood is quiet, introspective, sleepy, and slightly melancholic, never posed like a fashion shoot.

Panel count and layout: Include exactly 9 discrete panels:
1. Top left: close-up of the girl lying in bed under white sheets in soft morning window light, only her face and messy hair visible, sleepy eyes looking toward camera.
2. Top center: wide scenic view from a hillside town overlooking a sparkling inland sea, rooftops, utility poles, distant islands and mountains under a pale blue sky.
3. Top right: side-profile medium close-up of the girl outdoors on a school rooftop or terrace, wearing a white blouse, face tilted upward into bright sunlight, wind moving her hair, blue sky behind her.
4. Middle left: downward crop of her legs walking on sunlit asphalt, dark pleated skirt hem, white socks, black loafers, long sharp shadow stretching across the road.
5. Middle center: classroom scene with the girl seated by a bright window, elbow on desk, chin resting on her hand, holding a pen over a notebook, other students blurred in the background.
6. Middle right: close-up still life on a wooden school desk: open notebook with small writing, clear mechanical pencil, black school bag, and one small white bunny plush keychain hanging from the bag.
7. Bottom left: close-up of the girl eating ramen in a casual restaurant, holding chopsticks with noodles lifted above a white bowl, warm interior bokeh lights, her eyes glancing sideways.
8. Bottom center: empty bedroom at sunset with unmade bed in the foreground, curtains, window, orange sun low outside, soft dusty golden light filling the room.
9. Bottom right: dark nighttime close-up of the girl lying in bed under blankets, face lit by a smartphone screen, looking tired and absorbed.

Visual style: {argument name="visual style" default="pale pastel colors, low contrast, soft haze, realistic Japanese cinema photography, natural skin texture, slight film grain, gentle lens bloom, imperfect candid framing"}. Use shallow depth of field in character panels, natural backlight, sun flares, and subdued highlights. The overall palette should be airy whites, muted grays, washed blues, warm sunset amber, and soft shadow tones.

Constraints: Keep the same girl recognizable across all panels. Make it photorealistic, not anime or illustration. Avoid text overlays, captions, logos, watermarks, exaggerated makeup, glamour posing, or overly saturated colors. The collage should feel like a coherent diary of one ordinary day, from morning bed to school, town views, food, sunset bedroom, and late-night phone use.

Goal: Create a luxurious square Japanese travel advertisement for an autumn hot spring and gourmet tour sale, aimed at social media promotion.

Canvas: 1:1 square poster, high-resolution, rich warm autumn palette of crimson, gold, amber, cream, and deep brown. Use a premium travel-agency flyer style with layered photo collage panels, diagonal cuts, ornate Japanese paper textures, gold accents, and subtle drop shadows.

Layout: Build a dense collage with exactly 6 main photo areas: 1) upper-left large outdoor rotenburo hot spring with steaming water, natural rocks, a lantern, vivid red maple trees, a lake and mountains in the distance; 2) upper-right close-up of a lavish crab dish in a black tray with steam and side dishes; 3) mid-right close-up of marbled wagyu beef being lifted by chopsticks beside a rice bowl with chestnuts or gingko nuts; 4) lower-left cozy ryokan guest room with tatami, low table, cushions, warm lamps, shoji screens, and autumn scenery outside; 5) lower-right scenic lake and autumn mountain landscape; 6) small background food/seasonal detail panels near the top with mushrooms and vegetables. Separate the panels with clean diagonal cream or gold borders.

Central typography: Place a large cream washi-paper polygon label in the center. On it, use oversized Japanese calligraphy-style headline text reading {argument name="headline text" default="秋の食と温泉旅セール"}. Make “秋” red-to-orange gradient, “食” gold gradient, and the remaining characters dark brown. Add two decorative maple leaves around the headline. Beneath it, add smaller Japanese text: 「秋限定のごほうび旅」 with thin horizontal lines on both sides.

Offer area: Under the central label, create a deep red patterned price banner with gold trim. Show a small rounded box reading {argument name="trip length" default="2泊3日"}. Next to it, display the large price {argument name="price" default="39,800円〜"} in shiny gold serif numerals with white highlights and dark shadow. Add small gold emphasis marks near the price.

Benefits and call-to-action: Below the price, add a cream rounded plaque with fork, knife, and hot spring icons, followed by text 「食事付き・温泉満喫」. At the bottom center, add a glossy red rounded button with gold outline reading {argument name="button text" default="予約受付中 ＞"}.

Additional Japanese copy: Include exactly 5 supporting text blocks around the poster: upper-left white handwritten text 「この秋、心もおなかも満たされる。」 with a thin underline; upper-right vertical text on a cream strip 「旬を味わう、贅沢なひととき」; left vertical text over the ryokan panel 「特別な秋を、旅へ。」; lower-left small white text 「日本の美しい秋に、会いに行こう。」; lower-right white text 「行きたかったあの場所へ。」 above a simple white travel-agency logo mark and tagline 「もっと、心が動く旅を。」 plus tiny English text “TRAVEL for a richer tomorrow”.

Visual style: Photorealistic travel photography blended with polished Japanese advertising design. Emphasize steam rising from the hot spring and hot foods, glossy food textures, crisp autumn foliage, warm indoor lighting, elegant Japanese calligraphy, and premium gold-red sale graphics.

Constraints: Keep all Japanese text legible and placed as described. Use no people. Do not add extra prices, logos, QR codes, or unrelated text. Maintain a balanced luxury-tourism composition with strong central hierarchy.

Create a photorealistic candid campus lifestyle photo of {argument name="main subject" default="a handsome Japanese male university student"} walking through a bright outdoor school corridor with his friend. The composition is vertical 3:4, medium close-up, eye-level, natural daylight, shallow depth of field, soft bokeh background. Show exactly 6 visible people: 2 foreground students and 4 blurred background students. The main subject is centered-left, slim, clean-shaven, with tousled black medium-length hair, soft youthful features, and a gentle smile while looking to the right at his friend. He wears a white raglan T-shirt with black sleeves and a faded gray graphic reading {argument name="shirt text" default="BROOKLYN NEW YORK CITY EST. 1634"}, blue jeans, and a black backpack over one shoulder, holding the strap with his left hand. The friend is partially cropped on the right in profile, wearing {argument name="friend outfit" default="a light gray hoodie and black backpack"}, speaking casually with one hand raised. The corridor has concrete columns, windows, sunlit floor shadows, and a bright academic atmosphere, with students walking in the distance. Style should resemble a realistic Japanese university promotional snapshot, natural skin texture, soft highlights, relaxed friendship, no dramatic posing, no watermark, no extra text beyond the shirt graphic.

Goal: Create a vertical social-media comparison graphic showing two cinematic AI-image generations of the same sci-fi scene, emphasizing improved art direction between versions.

Canvas: Portrait 2:3 format, black background, high-contrast cinematic presentation with wide letterboxed image strips. Use a stacked layout with exactly 2 panoramic sci-fi image panels separated by thick black bands.

Layout: Top panel: a wide cinematic space landscape occupying the upper third. Center band below it contains the large white label “{argument name="top label" default="GPT IMAGE 2"}”. Bottom panel: a second wide cinematic space landscape occupying the middle-lower third. Bottom band contains the large white label “{argument name="bottom label" default="GPT IMAGE 2.5"}”. Keep the typography centered, tall condensed sans-serif, all caps, white, clean, and glowing slightly against black.

Scene content in each panel: Show the same composition in both panels: one man viewed from behind, standing on a rocky moon-like surface, looking toward a huge green Earth-like planet filling the right side of the sky. The man is centered-left, wearing a white short-sleeve shirt, black vest, dark pants, and white sneakers. The terrain is gray, cratered, and covered with jagged rocks. The background is deep outer space with stars, nebula clouds, small moons, and scattered asteroids. A bright sun flare appears near the right edge of the large green planet.

Panel differences: The top panel should look slightly darker, rougher, less refined, with a greener color cast and a more chaotic, lower-polish atmosphere. The bottom panel should look more polished, sharper, more balanced, with richer blue space tones, better light direction, clearer composition, crisper rocks, stronger planet detail, and a more cinematic lens flare.

Discrete visible elements to include: exactly 2 image panels; exactly 2 text labels; exactly 1 human figure in each panel, for 2 total human figures; exactly 1 large green planet in each panel, for 2 total large planets; exactly 2 social media icons in each panel, for 4 total icons. In both panels, place the two icons to the left of the man: a red YouTube play-button icon and a pink-purple Instagram camera icon, glowing slightly and floating in space.

Visual style: Ultra-wide cinematic sci-fi concept art, photorealistic digital matte painting, dramatic rim lighting, high detail, strong depth, black letterbox bands, poster-like comparison layout, no watermark, no extra captions beyond the two version labels. Use {argument name="main color palette" default="emerald green planets, blue-black space, gray lunar rocks, white typography"}.

Create a bright vertical anime illustration of two cute girls in a golden wheat field under a vivid blue sky with soft white clouds and distant green mountains. In the foreground, show {argument name="main character" default="a surprised kappa-themed girl"}: a small young anime girl with short dark teal bob hair, huge amber eyes, flushed cheeks, and an open startled mouth. She wears a glossy pale-green hooded raincoat shaped like a kappa, with a ducklike yellow beak on the hood, big cartoon eyes on top, dark teal leaf-like hair decorations, dark buttons, cuff trim, two front pockets, and a small white chest name tag reading {argument name="name tag text" default="かっぱ"}. Add yellow gloves and yellow rain boots, with her hands raised in a frightened pose. Behind her to the right, show {argument name="second character" default="a smiling scarecrow witch girl"}: a slender anime girl with short messy dark brown hair, pink-purple eyes, and a warm smile, wearing a tattered black rustic dress with puff sleeves, rope belt, ribbon bow at the collar, straw cuffs, and a wide frayed straw hat with a dark feather; she waves with one hand like a friendly scarecrow. Include exactly 2 characters: the kappa-costumed girl in front and the scarecrow witch girl behind. Include exactly 3 visible text/emphasis elements: 1 large speech bubble above the scarecrow saying {argument name="speech bubble text" default="こんにちは〜！どこ行きますか?"}, 1 yellow punctuation mark cluster “!?” near the kappa girl’s head, and 1 small label on the raincoat reading “かっぱ”. Use dynamic manga-style linework, expressive faces, high-detail clothing folds, warm sunlight, shallow depth of field with blurred wheat stalks in the extreme foreground, and a cheerful pastoral adventure mood. Make the composition portrait-oriented, with the kappa girl occupying the lower-left foreground and the scarecrow girl in the upper-right midground. Preserve the Japanese text exactly where specified, with clean readable lettering.

Goal: Create a cinematic anime movie-style 3x3 contact sheet featuring {argument name="character name" default="a sweet brown-haired gothic lolita heroine"} in an elegant rose-filled palace interior, with each panel looking like a still frame from the same romantic fantasy film.

Canvas: Wide 16:9 image divided into exactly 9 equal rectangular panels in a 3 columns by 3 rows grid. Use thin white gutters between panels. Each panel has a delicate ornamental film-frame border with tiny rose motifs, corner flourishes, faint guide lines, and small decorative marks, like premium anime key visuals.

Character design: A young anime girl with very long wavy {argument name="hair color" default="warm chestnut brown"} hair, amber-gold eyes, soft blush, and a cheerful gentle expression. She wears a burgundy and cream gothic lolita dress: cropped burgundy jacket or cardigan, ivory ruffled blouse, large red ribbon at the collar with a rose brooch, layered cream lace skirt with burgundy plaid overskirt, bows, ribbons, frills, white lace socks, and dark platform Mary Jane shoes. Add a large burgundy rose-and-ribbon hair ornament with gold chains and dangling clover-like charms. Overall palette: {argument name="color palette" default="warm rose red, ivory cream, gold, soft brown, and glowing peach light"}.

Setting and mood: A luxurious cathedral-like palace hall or conservatory filled with roses, tall arched windows, chandeliers, marble floors, ornate furniture, candles, and golden sunlight. Red rose petals float through the air in every panel. Lighting is dreamy, warm, backlit, shallow depth of field, soft bokeh, highly polished anime illustration quality.

Panel layout, exactly 9 shots:
1. Top-left: full-body establishing shot of the girl standing centered in a grand arched hall, dress fully visible, roses and sunlight around her.
2. Top-center: close-up front portrait, smiling brightly and reaching one hand toward the viewer, hair flowing, petals in foreground.
3. Top-right: side-profile bust shot facing right, ponytail and long hair streaming behind her against a pale city or palace skyline through the windows.
4. Middle-left: rear three-quarter shot from behind, emphasizing the volume and curls of her long hair and burgundy outfit, golden palace interior blurred behind.
5. Middle-center: seated or leaning-at-table portrait, chin resting on hand, gentle smile, books, roses, framed paintings, and candlelit decor behind her.
6. Middle-right: dynamic dancing or twirling shot, eyes happily closed, skirt flaring wide, one arm extended, petals sweeping across the frame.
7. Bottom-left: extreme close-up detail of the hair ornament: burgundy rose, satin ribbon, gold chains, and dangling charms attached to her hair.
8. Bottom-center: close-up detail of her lower legs and shoes walking on glossy marble, white socks with bows, dark platform Mary Janes, scattered rose petals reflected on the floor.
9. Bottom-right: close-up over-the-shoulder portrait, she turns back toward the viewer with a soft smile, roses and warm lights behind her.

Visual style: High-end Japanese anime illustration, movie trailer stills, delicate linework, intricate clothing detail, luminous skin, silky hair rendering, soft cinematic depth of field, warm romantic fantasy atmosphere, elegant and polished like a promotional PV storyboard.

Constraints: Keep the same character, outfit, hairstyle, and accessories consistent across all 9 panels. Do not add dialogue, captions, logos, subtitles, watermarks, or extra characters. Maintain a cohesive film-still sequence rather than nine unrelated illustrations. Use {argument name="render quality" default="ultra-detailed, cinematic, soft-focus anime key visual"}.

Goal: Create a minimalist chat-style image generation interface screenshot showing a rough sketch prompt and a polished generated result.

Canvas: Square-ish light gray/off-white background, about 665×677 px, with lots of empty space. Use a clean modern mobile/web UI aesthetic with rounded cards and subtle shadows.

Layout: Place exactly 3 main UI elements: 1 small sketch image card in the upper right, 1 small rounded text prompt bubble directly below it, and 1 large generated image card in the lower left. Leave the left side mostly blank above the large card.

Upper sketch card: A white rounded rectangle with a childlike black-and-orange doodle. The doodle contains exactly 6 orange star shapes, a tall crooked black tower in the center, 3 small black haunted houses along the bottom, an orange full moon in the upper right, and 1 black witch silhouette flying across the moon. Keep the sketch rough, marker-like, playful, and sparse.

Prompt bubble: A pale peach rounded pill below the sketch card containing the Japanese text {argument name="prompt bubble text" default="闇夜を飛ぶ魔女の画像。"} in small dark brown type.

Generated image card: A large rounded-corner square image in the lower left, cinematic fantasy style. Depict a cozy gothic village at night beneath a deep navy starry sky. Include exactly 1 tall stone clock/bell tower with a pointed dark roof near the left-center, exactly 1 enormous glowing golden full moon near the upper right, and exactly 1 black witch silhouette riding a broom in front of the moon. Include clustered steep-roof houses with warm yellow-lit windows in the foreground, soft glowing stars, dramatic moonlit clouds, and dark treetops framing the bottom. The image should feel like a highly polished transformation of the crude sketch.

UI details: On the lower-left corner inside the large generated image, show the small white Japanese label {argument name="edit label" default="編集"}. On the lower-right corner inside the same image, show a small white share/export icon. No other visible text.

Visual style: High contrast between the simple doodle and the refined result; rounded corners; soft app-interface spacing; no watermark; no extra panels; keep the composition faithful to a generated-image chat result.

Goal: Create a polished square promotional anime key visual for {argument name="product name" default="GPT Image 2.5"}, presenting it as a futuristic creative image-generation service with a cheerful anime ambassador reaching toward the viewer.

Canvas: 1:1 square social-media poster, cinematic neon city background at sunset, high-detail anime illustration blended with sleek SaaS launch-page UI overlays. Use strong depth of field: the character’s outstretched hand is large and slightly motion-blurred in the foreground, her face is sharp, and the city is softly glowing behind her.

Main subject: A cute anime girl in the center-right foreground, about teenage/young adult style, short fluffy {argument name="hair color" default="pale pink-blonde"} bob hair with bangs, one eye winking and one warm reddish-brown eye open, smiling brightly. She wears a futuristic black-and-white techwear jacket with glossy panels, straps, zippers, badges, red ribbon accents, black inner top, and a lanyard badge printed with the OpenAI-like knot logo and “GPT Image 2.5”. Her right arm reaches toward the camera invitingly. Add a small black-and-red hair clip and ribbon on one side.

Companion robot: Place exactly 1 small rounded companion robot at the lower right, white body with cat-ear-like points, glossy black face screen, cyan smiling eyes, and a chest panel containing the OpenAI-like knot logo plus the text “Create Explore Together”.

Branding and headline: At the upper left, place a large glowing OpenAI-like knot icon followed by huge text “GPT” on the first line and “Image 2.5” below it, with “2.5” in a blue-to-pink neon gradient. Under it, add the tagline “More Expressive. More Creative. Now in Your Hands.” Add handwritten Japanese text beneath: 「想像を、もっと自由に。」

Left feature stack: Add exactly 6 translucent dark glassmorphism cards stacked vertically on the left, each with a glowing icon, bold white title, smaller subtitle, and where visible a small preview thumbnail strip. The 6 cards are: 1) “Text to Image” with subtitle “Just type your imagination” and 1 landscape castle thumbnail; 2) “Sketch to Image” with subtitle “Turn your sketches into art” and 2 thumbnails showing a sketch converting to a fantasy landscape with an arrow; 3) “Image Editing” with subtitle “Refine with simple instructions” and 2 anime portrait thumbnails with an arrow; 4) “In-context Creation” with subtitle “Consistent characters & style” and 2 anime character thumbnails; 5) “Faster Generation” with subtitle “Create more, wait less” and a lightning icon; 6) “Higher Detail” with subtitle “Sharper, richer, more real” and 1 flower thumbnail.

Right feature stack: Add exactly 3 translucent dark glassmorphism cards on the right side. The top card is “Multi-style” with subtitle “Anime, Realistic, Illustration...” and exactly 3 thumbnails: anime portrait, realistic landscape, cat photo. The middle card is “Follow Instructions” with subtitle “More accurate, more flexible”. The lower card is “Create on Mobile” with subtitle “@ → Sketch (iOS/Android)”.

Handwritten notes: In the upper right sky, add Japanese handwritten copy: 「好きな世界を、今すぐ、ここから。」 followed by a signature {argument name="signature" default="Chisato"} and a heart. At the bottom right, add handwritten Japanese copy: 「つくろう。あなたの『好き』を。」 and the line “with GPT Image 2.5”. At the bottom left, add a small OpenAI-like knot icon and cursive English text “A more creative tomorrow_”.

Visual style: Premium anime promotional art, vibrant blue, cyan, magenta, and sunset-pink lighting; glowing holographic UI panels; reflective metal and glass; crisp typography; dramatic wide-angle selfie perspective; highly detailed face and clothing; soft bloom, rim light, and lens flare. Keep all UI text legible and aligned, with no extra feature cards beyond the specified counts.