GPT Image 2 prompts

Goal: Create a refined Traditional Chinese Medicine wellness event poster about moxibustion and meridian activation, with an elegant classical Chinese aesthetic and warm beige medicinal atmosphere.

Canvas: Horizontal 16:9 poster, 1366×768 style composition, soft cream parchment background with mist, pale ink-wash mountains and a tiny pine tree at the lower left, subtle golden-brown gradients and thin decorative divider lines.

Layout: Left third contains a large vertical Chinese headline with exactly 4 oversized calligraphy characters, arranged in two stacked columns: {argument name="main headline" default="温经活络"}. Add tiny pinyin annotations near the characters, reading “wēn”, “jīng”, and “huó”. Beside the headline, add a slim vertical explanatory Chinese text column about warming meridians, regulating qi and blood, dispelling cold and dampness, and nourishing the body; include a small red seal reading “艾灸”. Center-right shows a close-up photographic back/shoulder therapy scene: a person seen from behind with neck and shoulder exposed, a lit moxa stick angled diagonally toward the upper shoulder, glowing ember and white smoke curling upward. Mark exactly 3 acupuncture points on the shoulder/back with small red dots and labels: “大椎穴”, “肩胛穴”, and “膏肓穴”. Across the lower middle, place a Chinese subtitle separated by thin lines: {argument name="subtitle" default="温经散寒 · 通络止痛 · 调和气血 · 提升体感"}.

Bottom information band: Add exactly 4 rounded rectangular service cards with orange line icons and Chinese headings. Card 1 has a flame icon, heading {argument name="card one heading" default="温经活络"}, description “温通经络，改善寒凝”, and two pill tags: “适合人群” and small text about cold hands/feet, shoulder-neck fatigue, cold-damp constitution, qi-blood deficiency, low immunity, and high pressure. Card 2 has a seated therapy icon, heading “背部调理”, description “疏通经络，缓解酸胀”, and tags “建议频次” plus “每周1–2次，循序渐进，效果更佳”. Card 3 has a steam/waves icon, heading “艾烟闻气”, description “艾草温润，放松身心”, and tags “调理时长” plus “60分钟/次，深度放松，温和舒适”. Card 4 has a clipboard icon, heading {argument name="card four heading" default="体质方案"}, description “辨体定制，长期调养”, and tags “温馨提示” plus “调理后多喝温水，忌食生冷，注意保暖”.

Visual style: High-end TCM product poster, minimal and spacious, sepia-orange typography, traditional calligraphy mixed with clean Song-style Chinese body text, translucent rounded cards, soft shadows, no harsh contrast. Use realistic skin texture and smoke for the therapy photo area, blended into the parchment background.

Constraints: Keep the design calm, premium, and educational; preserve exactly 4 bottom cards, exactly 3 acupuncture-point labels, and exactly 4 main headline characters. Avoid extra logos, QR codes, watermarks, clutter, or modern neon colors.

Create an ultra-realistic cinematic vertical portrait photo of {argument name="subject" default="a stunning 20-year-old East Asian woman"} surfing in {argument name="location" default="Sanya"} at sunset, captured from a very low water-level angle close to the front of the surfboard. She is leaning forward on the board with both arms in the water, laughing brightly with eyes nearly closed, radiating joy and confidence. Her dark wet hair is loosely tied back with messy strands around her face, and she wears {argument name="outfit" default="a soaked short-sleeve white collared knit top over dark bottoms"} with a small necklace visible. The surfboard nose dominates the lower right foreground, cutting diagonally toward the viewer, while splashing seawater and frozen droplets sparkle around her arms and torso. Behind her is a dramatic tropical sunset with a glowing orange sun near the horizon, golden reflections across the ocean, distant dark hills, and a moody sky blending deep blue, purple-gray clouds, peach, and fiery orange light. Use professional photography realism, intense focus on her smiling face, shallow depth of field, water droplets frozen in mid-air, high dynamic range, natural skin texture, wet fabric detail, cinematic contrast, golden rim light, and an 85mm lens look. Frame as {argument name="aspect ratio" default="vertical 9:16"}; avoid text, logos, extra people, or artificial beauty-filter smoothness.

Goal: Create a high-energy football match poster in a green-and-white tactical blueprint style for {argument name="event name" default="FIFA World Cup 2026"}, combining a top-down soccer field diagram, bold Chinese typography, and dynamic ball motion.

Canvas: Horizontal 16:9 sports poster, deep grass-green background with subtle paper grain, scuffed print texture, and white ink distressing. Use a full-bleed soccer pitch viewed from above as the main composition, with thin white technical linework, corner flags, goal boxes, center circle arcs, stadium seating outlines, measurement marks, dotted rules, arrows, crosses, tiny tactical symbols, and a thin white border.

Layout: Put an enormous distressed white Chinese headline {argument name="main headline" default="开幕之夜"} across the center, occupying most of the poster width. Add a realistic black-white-red soccer ball on the lower-right side of the headline, flying diagonally from right to left with bright white and red motion streaks, paint splatter, and speed particles crossing the field. Keep the headline layered above the field markings but partly intersected by the ball trail for depth.

Top information row: Include exactly 3 labeled information modules: left module with a stopwatch icon and {argument name="date and time" default="6月14日 20:00"} plus small English label “KICK-OFF”; center module with a simplified World Cup trophy icon and text “WORLD CUP 2026”; right module with a location pin icon and {argument name="venue" default="国家体育场"} plus small English label “NATIONAL STADIUM”. Along the very top border add small uppercase microtext: “FIFA WORLD CUP 2026”, “THE WORLD IS OURS”, and “FOR THE GAME. FOR THE WORLD.” separated by dotted lines and plus symbols.

Bottom information row: Include exactly 3 main callouts below the headline: left callout with a clipboard tactics icon and text “A组首轮” with small English “GROUP A · ROUND 1”; center callout with a monitor/play icon and text {argument name="booking text" default="直播预约已开启"} with small English “LIVE STREAM BOOKING OPEN”; bottom-center scoreboard badge reading “MATCH 01”. Add small footer labels and icons such as “TEAM SPIRIT”, “TACTICS PLAN”, “CONTROL THE GAME”, and “PASSION & GLORY”.

Visual style: Bold modern Chinese block display font, very heavy weight, rough stamped edges, cracked grunge texture, slight off-white ink. Background should feel like a tactical coaching board mixed with a stadium blueprint. Use only green, white, black, and small red accents from the soccer ball trail. High contrast, crisp vector linework, energetic sports advertising aesthetic.

Constraints: Use exactly one soccer ball, one central oversized Chinese headline, three top modules, three bottom callouts, and one match badge. Keep all text white, avoid extra players or people, avoid photorealistic stadium crowds, no logos other than generic World Cup-style trophy iconography, no watermark.

Create a vertical cinematic action poster featuring {argument name="character description" default="an adult Asian female swordswoman with cool beautiful features, translucent fair skin, subtle makeup, long black wet-looking hair with messy strands blown by high-speed wind, and calm sharp eyes looking directly at the viewer"}. She is posed in a close-up three-quarter stance, gripping a Japanese katana horizontally across the frame at chest and mouth level, the blade running from the left edge toward the right edge and passing directly in front of her face. Her hands are clenched around the hilt on the left, the ornate guard is visible, and the polished blade has a wavy hamon pattern. Add bright white blade aura, smoky air trails, thin slash streaks, sparks, glassy particles, and motion distortion around the sword so it feels like the cut is already slicing through the air. Dress her in {argument name="outfit" default="a loose slightly wrinkled white shirt with rolled sleeves, a gray knit cardigan tied at the waist, a dark gray-black plaid pleated skirt, black over-knee socks, and black academy-style leather shoes"}; show two black Japanese long swords crossed and strapped on her back, with black scabbards and dangling cords visible over her shoulder. Use a pale gray studio background with subtle grunge scratches, ink smears, wind lines, and faint debris, high-fashion editorial lighting, sharp photorealistic detail, glossy skin highlights, and a cold silver-white color palette. Make the composition intense and dynamic, with her face centered in the upper half and the sword dominating the middle. Add exactly 4 visible futuristic typography elements: small top-left label “BLADE MODE”, small left label beneath it “CUT THROUGH”, huge distressed black glitch title across the lower third “SHADOW SLASH”, and tiny bottom-left label “PHASE ATTACK”. The main title should be bold, stretched, fragmented, and partially transparent with horizontal scanline breaks, while the small labels use thin sci-fi technical lettering. No extra characters, no watermark, no logo, no additional readable text.

Goal: Create a wide comparison graphic showing four generated anime-style close-up portraits of the same young woman, comparing prompt-writing AI tiers.

Canvas: 1200×675 landscape image with a dark charcoal/black background, like a social media comparison card. Use a clean editorial layout with a narrow title band at the top and four equal vertical portrait panels underneath.

Top text: Add a large white Japanese headline across the top: 「実験③：女性の顔アップ。同じ画像AI（GPT Image 2）で、プロンプトを“書いたAI”だけが違う4枚」. Beneath it, add a smaller gray Japanese subtitle: 「変えたのは『プロンプトを書くClaudeのモデル階層』だけ（左＝最も安い → 右＝最上位）」.

Layout: Four side-by-side portrait columns, exactly 4 panels, separated by thin black gutters. Each panel has a dark header strip with small white text. The 4 visible panel labels from left to right are: 1) 「haiku｜最も安い」, 2) 「sonnet」, 3) 「opus」, 4) 「fable｜最上位」.

Subject: In every panel, show the same {argument name="character description" default="young woman with long dark hair, delicate features, large emotional eyes, soft lips, and a wistful expression"} in a close-up or bust portrait. She is outdoors in warm golden-hour sunlight with blue sky, soft clouds, distant greenery or mountains, and wind-blown hair. The emotional tone should be tender, cinematic, and slightly melancholic, with teary or reflective eyes.

Panel-specific portrait styles: Panel 1, “haiku”, should look more semi-realistic and decorative, with painterly anime illustration rendering, heavier details, ornate earrings, a braid, warm sunset shadows, and a slightly less clean cel-animation face. Panel 2, “sonnet”, should look cleaner and more theatrical-anime-like, with a soft cel-shaded face, long dark hair tied loosely back, bright rim light, delicate tearful eyes, and a pale blouse. Panel 3, “opus”, should be polished anime illustration with strong cinematic golden backlight, loose dark hair, refined eyes, warm blush, and a close shoulder-turned pose. Panel 4, “fable”, should be the cleanest flat theatrical anime cel style, simpler facial features, long loose hair, sleeveless light top, blue sky and clouds behind her, and an innocent upward-looking expression with subtle tears.

Visual style: High-quality Japanese anime portrait comparison, cinematic lighting, soft lens glow, natural wind, expressive eyes, delicate linework, warm highlights, dark UI-like framing, crisp readable text. Keep the four portraits related but visibly different in rendering quality and style.

Constraints: Use exactly 4 portrait panels and exactly the 4 listed labels. Do not add extra panels, logos, watermarks, or additional explanatory text. Preserve the Japanese text exactly as specified.

Goal: Create a wide social-media comparison graphic showing four AI-generated anime landscape wallpapers side by side, comparing prompt-writing model levels while keeping the same composition and image model.

Canvas: Wide horizontal 2:1 image, approximately 1200×620 px, dark charcoal background. Add a black header band across the top and four tall vertical image panels beneath it, separated by narrow black gutters.

Header text: Large Japanese title at the top left: 「同じ構図・同じ画像AI（GPT Image 2）。プロンプトを“書いたAI”だけが違う4枚」. Smaller gray subtitle underneath: 「変えたのは『プロンプトを書くClaudeのモデル階層』だけ（左＝最も安い → 右＝最上位）」.

Layout: Use exactly 4 vertical panels, equal width, arranged left to right. Each panel has a dark label strip at the top with white text. The four labels are exactly: 1) 「haiku｜最も安い」, 2) 「sonnet」, 3) 「opus」, 4) 「fable｜最上位」.

Panel imagery: All four panels depict the same basic anime cinematic composition: a tiny solitary human figure on a grassy mountain ridge, overlooking a vast valley, distant layered mountains, dramatic clouds, and golden sunlight. Make each panel beautiful but subtly different in art direction.

Panel 1, haiku: Warm sunset fantasy landscape, orange and pink sky, sun near the horizon, purple-blue mountain layers, mist in the valleys, wildflowers and grass in the foreground, one tiny dark standing figure centered near the lower third, highly polished but conventional.

Panel 2, sonnet: More restrained natural lighting and atmospheric perspective, huge textured cloud mass in a blue-and-gold sky, soft glowing valley haze, one tiny standing figure at the end of a ridge path, realistic depth and gentle warm light.

Panel 3, opus: Stronger cinematic composition with a sweeping cloud formation, low sun at the right horizon, glowing lake or river in the valley, one tiny figure on a grassy ridge near the bottom center, refined color balance, crisp foreground grasses, natural aerial depth.

Panel 4, fable: Brighter anime film still feeling, turquoise-blue sky, towering cumulus clouds, sunbeams breaking through, lush green valley and hills, one small but clearly animated person in a white dress and sunhat moving or turning on the grassy slope, clothing and pose readable despite small scale.

Visual style: High-end anime background art, painterly but detailed, Makoto Shinkai-inspired light, cinematic atmosphere, soft bloom, elegant color grading. Keep the comparison graphic clean and readable, with thin gutters and consistent panel sizing.

Customizable details: Use {argument name="image model name" default="GPT Image 2"} in the title, compare {argument name="first model label" default="haiku｜最も安い"}, {argument name="second model label" default="sonnet"}, {argument name="third model label" default="opus"}, and {argument name="fourth model label" default="fable｜最上位"}.

Constraints: Include exactly 4 panels and exactly 4 tiny human figures total, one per panel. Do not add logos, watermarks, extra charts, extra icons, or additional captions beyond the header, subtitle, and four panel labels.

Goal: Create a vertical Japanese manga page for Tanabata, romantic comedy tone, showing a misunderstanding about wish strips on a bamboo tree between two stylish anime characters.

Canvas: Tall vertical comic page, approximately 2:3 aspect ratio, with clean white panel gutters and dynamic diagonal manga panel borders. Nighttime interior/exterior balcony setting with a deep blue starry sky and Milky Way visible through large windows. Highly polished modern anime illustration, detailed lighting, glossy shadows, cinematic blue moonlight mixed with warm indoor lamp light.

Characters: Use two recurring characters consistently across all panels. The male lead is {argument name="male character description" default="a slim young man with messy short blue hair, pale skin, multiple earrings, a black glossy hooded jacket over dark streetwear, slightly cool but flustered expression"}. The female lead is {argument name="female character description" default="a petite young woman with long ash-blonde twin-tail hair, red eyes, black gothic lolita dress with lace, black headband, ribbon accessories, pink heart earrings, playful mischievous smile"}.

Layout: Exactly 6 panels. Panel 1 is a wide top panel spanning the full page width. Panels 2 and 3 sit beneath it side by side, with panel 2 narrow on the left and panel 3 larger on the right. Panels 4 and 5 sit below, with panel 4 on the left showing the boy alone and panel 5 larger on the right showing both characters arguing near the bamboo. Panel 6 is a wide full-width chibi gag panel at the bottom with a pastel background.

Panel details:
1. Wide top panel: The blue-haired boy stands by a bamboo branch covered with colorful Tanabata tanzaku wish papers, carefully tying or checking a light-blue strip. Starry night sky fills the left background, bamboo leaves and hanging wish strips fill the right. Include about 7 visible hanging wish strips in blue, yellow, pink, purple, and tan.
2. Middle-left panel: The gothic girl peeks through ornate double doors from a warmly lit room, surprised and curious. Add one vertical speech bubble saying {argument name="girl surprised line" default="あれ？"}.
3. Middle-right panel: Close-up of the girl beside bamboo leaves and tanzaku, smiling teasingly with red eyes bright, one hand near her ear. Add one vertical speech bubble saying {argument name="girl teasing line" default="うちにもユタの笹ってあったんだ？"}. Show about 5 colorful wish strips hanging around her.
4. Lower-left panel: The boy turns back in embarrassment, holding a tan wish strip partly hidden behind him, sweat drop and startled manga mark above his head. No large dialogue needed.
5. Lower-right panel: The boy and girl face each other indoors near the balcony and bamboo. The boy looks defensive with one hand on his hip, the girl smiles with eyes closed, teasing him. Add two speech bubbles: boy says {argument name="boy excuse line" default="物置にあったんだよ！掃除してただけだし！"}; girl says "そかそか".
6. Bottom full-width chibi panel: Simplified cute chibi versions on a pale lavender dotted background. The boy is flustered, blushing and puffing, holding several rectangular tanzaku papers that fly from his hands; the girl giggles with eyes closed and hand near mouth. Add small cartoon steam puffs, sparkle stars, and motion marks. Exactly 5 loose tanzaku papers should be visible in this gag panel.

Text and lettering: Use Japanese manga-style vertical speech bubbles with clean black text on white bubbles. Keep the visible dialogue exactly as specified. Do not add extra narration boxes or sound effects beyond small decorative emotion marks.

Visual style: Premium glossy anime manga rendering, detailed hair strands, delicate faces, high contrast moonlit atmosphere, elegant gothic fashion, romantic Tanabata decorations, expressive comedy beats, crisp white gutters, no watermark, no logo.

Create a cinematic dark-fantasy royal scene titled {argument name="title theme" default="The Queen and the Serpent"}: a pale young queen kneels in prayer inside an opulent Gothic cathedral, hands pressed together, eyes gently closed, expression serene and solemn. She has extremely long, voluminous silver-white curly hair cascading to the floor, a delicate gold crown or halo-like filigree headpiece, porcelain skin, and wears an elaborate deep crimson gown with gold embroidery, sheer flowing sleeves, a high slit revealing one bent knee, and a long train spread across a red-and-gold carpet. Beside and slightly behind her is one enormous white serpent, coiled on the polished marble floor, its upper body raised in an elegant S-curve and its head turned toward the queen; the snake has pearlescent ivory scales, golden eyes, subtle gold jewelry around its neck, and a calm guardian-like presence. The setting is a grand medieval cathedral sanctuary with towering pointed arches, tall stained-glass windows glowing with warm backlight, ornate gilded altar structures, carved marble columns, candelabras with lit candles, red royal banners bearing gold crowns and lion-like heraldic emblems, floral arrangements, and reflective marble flooring. Use warm sunlight streaming through stained glass, soft haze, rim lighting around the queen’s hair and serpent scales, rich red and gold color palette, ultra-detailed textures, romantic baroque-gothic elegance, realistic fantasy illustration, vertical composition, full-body view, no modern objects, no text, no watermark.

Create a highly detailed vertical anime fantasy portrait of {argument name="character name" default="a silver-haired maiden"} dressed as Orihime for Tanabata, blending a white maid aesthetic with an ornate celestial kimono. She stands centered from upper thighs upward against a deep midnight-blue Milky Way sky filled with dense stars, two visible shooting stars, and a luminous diagonal galaxy band behind her. The character is a delicate young woman with pale skin, large cool gray-blue eyes, a calm slightly wistful expression, and long {argument name="hair color" default="silver-white"} hair with soft bangs and loose side strands. She wears an elaborate white maid headdress shaped like a ruffled crescent crown, decorated with pearls, silver ornaments, tiny star motifs, and small black ribbon bows at both sides. Her outfit is a layered white and navy kimono-maid dress: white inner kimono collar, dark navy cross-collar accents, broad ruffled shoulder sleeves, translucent flowing outer sleeves, and a dark obi belt with jeweled star brooches, pearl chains, tassels, and ribbon loops. Cover the fabric with intricate silver and gold constellation embroidery, snowflake-like star patterns, fine bead chains, pearl drops, and delicate lace ruffles. Include exactly four prominent hanging tassels: one long white tassel from the central obi ornament, one tassel on the left sleeve edge, one tassel on the right sleeve edge, and one tassel near the lower left drape. Her right hand is raised gently near her collar, while her left hand rests lightly in front of the obi. Use a refined Japanese light-novel illustration style, ethereal lighting, shimmering highlights, crisp linework, soft skin rendering, luminous fabric textures, and an elegant cool palette of {argument name="main palette" default="navy blue, silver, white, and pearl gray"}. No text, no watermark, no modern objects, and keep the composition graceful, symmetrical, and richly ornamented.

Create a high-quality vertical portrait photo of a cute young adult anime-style cosplayer dressed as a Pikachu-inspired character, seated on an elegant vintage pink tufted loveseat in a bright pastel European studio. The subject is {argument name="character type" default="a blonde young adult female cosplayer"} with long wavy twin-tails, soft bangs, large expressive eyes, smooth fair skin, and a gentle smile, looking directly at the camera while holding both hands up to adjust a yellow headband with two upright Pikachu-like ears with black tips. Outfit: a glossy fitted yellow strapless mini-dress with black trim along the sweetheart neckline, a visible front zipper, black side panels, and one black lightning-bolt emblem on the right waist; matching long yellow arm sleeves with black edging; a simple black choker; thigh-high yellow stockings; and black platform Mary Jane heels with ankle straps. Pose: seated upright with knees together and legs extended toward the camera, feet close together, shoulders relaxed, playful and charming but tasteful. Environment: ornate carved wooden loveseat with dusty-rose velvet upholstery and button tufting, pale blue paneled wall with white decorative molding, partial ornate mirror frame on the right, white fireplace mantel on the left, and a vase of soft pink flowers in the background. Lighting and style: realistic fashion cosplay photography, soft diffused natural light, pastel color palette, shallow depth of field, crisp focus on face and costume, flattering skin tones, vertical full-body composition, 85mm lens look, high detail, no text, no watermark, no extra people, no distorted hands or limbs. Use {argument name="main costume color" default="bright yellow"} and {argument name="accent color" default="black"} as the dominant costume colors, with {argument name="sofa color" default="dusty pink"} upholstery and a {argument name="room wall color" default="pale blue"} background.

Create a vertical 2:3 anime-style illustration of {argument name="character name" default="a gentle young woman"} standing centered in front of a charming small clothing shop. She has {argument name="hair color" default="very long dark brown hair"}, warm brown eyes, fair skin, a soft shy smile, and a graceful front-facing pose with both gloved hands clasped at her waist. Dress her in an elaborate {argument name="dress colors" default="cream and blush pink"} princess lolita gown: off-shoulder puff sleeves, lace-trimmed neckline, floral choker, fitted lace-up corset bodice, white gloves with lace cuffs, a voluminous layered skirt, ornate cream lace panels, embroidered pink roses near the hem, and exactly 5 prominent pink bows: 1 at the chest, 2 on the sleeves, 1 on the left skirt side, and 1 on the right skirt side. The setting is a sunny European cottage storefront with warm hand-painted textures: a wooden sign above the awning reading {argument name="shop sign text" default="SHOP"}, an orange-and-cream striped scalloped awning, green-framed windows and a green door, and a small hanging sign on the door reading {argument name="door sign text" default="OPEN"}. Include exactly 2 readable signs total: the large shop sign and the door sign. Add cozy details around her: a terracotta pot of white daisies on the left, a small blank chalkboard leaning near the entrance, leafy vines climbing on the right wall, and a colorful flower bed on the right. Use soft golden daylight, delicate line art, pastel colors, romantic shoujo anime atmosphere, rich lace detail, painterly background, and a wholesome boutique fashion theme. Keep the character full-body or nearly full-body, centered, with the skirt filling the lower half of the image; avoid modern logos, extra text, watermarks, or additional characters.

Create a vertical anime light-novel / manga poster cover for {argument name="story title" default="眠気屋の小瓶"}, with a dark midnight-blue fantasy atmosphere and luminous gold typography. Canvas is a tall 2:3 poster, dense cinematic composition, highly detailed Japanese illustration style, dramatic rim lighting, glowing particles, floating manuscript pages, ink splatters, and magical starry effects. At the very top, place the huge ornate Japanese title {argument name="large title text" default="眠気屋の小瓶"} in bright ivory-gold calligraphy with a soft glow, decorative crescent moon, filigree swirls, tiny stars, and an arched celestial ornament above it. Directly below the title, add the smaller subtitle {argument name="subtitle text" default="眠気を売りすぎた漫画家"} in clean pale lettering.

Center of the poster: a transparent glass corked bottle floating upright, filled with a swirling spiral galaxy of blue-white starlight, sparkling dust, and a radiant core. The bottle emits a strong cyan-white magical glow that lights all surrounding characters and papers.

Foreground: a tired young male manga artist with messy black hair, pale skin, blue-gray eyes, and a loose white T-shirt reaches upward toward the glowing bottle with one hand; his expression is exhausted, awed, and desperate. Black ink splashes and manga tools frame the bottom edge.

Right foreground: a beautiful heroine with long flowing dark hair, large blue eyes, and a soft hopeful expression reaches out through drifting pages. She wears a white-and-silver fantasy outfit with delicate armor-like details and a glowing chest highlight.

Upper right: an elegant mysterious shopkeeper, an adult man with long brown hair, sharp tired eyes, and dark clothing, watches calmly from the shadows among shelves of bottles and papers.

Upper left: show the entrance of a strange shop with warm interior light and a sign reading {argument name="shop sign text" default="眠気屋"}; in the doorway, include a small back-view silhouette of a hooded young man facing into the shop.

Background details: include exactly four clusters of manga/manuscript imagery around the central bottle: one lower-left cluster of illustrated manga panels with multiple small characters, one mid-right cluster of fantasy character sketches, one upper-middle cluster of loose pages flying diagonally, and one bottom-edge cluster of inked manuscript sheets. Keep the scene packed but readable, with the glowing bottle as the focal point. Use no English text, no logo, no watermark, and preserve the Japanese text exactly as specified.

Create a vertical manga cover poster for a dark fantasy drama titled {argument name="main title text" default="眠気屋の小瓶"}, with the subtitle {argument name="subtitle text" default="眠気を売りすぎた漫画家"}. The composition shows a tired young male manga artist, {argument name="character name" default="Ren Mashima"}, seated at a cluttered drawing desk late at night inside a mysterious antique shop or studio. He has messy black hair, pale skin, sharp exhausted eyes, and wears a loose black button-up shirt. His left elbow rests on the desk with his hand against his temple, while his right hand reaches forward toward the viewer, holding a small corked glass bottle wrapped with thin metal wire and a paper tag bearing a crescent moon symbol. Inside the bottle is glowing blue-violet magical mist shaped like sleepy stardust, with tiny stars and a crescent moon, spilling faint smoke upward. The desk is covered with manga manuscript pages, ink pens, nibs, brushes, black ink bottles, a dark coffee mug, crumpled paper, and scattered art tools. Behind him, shelves hold many glass bottles and jars filled with dim blue, amber, and purple light, suggesting a shop that sells bottled sleepiness. A large window on the left reveals a deep blue night cityscape with stars and distant lit buildings. Use dramatic cinematic lighting: cool moonlight from the window, warm amber lamp highlights from the shelves, and bright magical blue glow from the bottle illuminating his fingers and face. Place the large ornate Japanese title across the top in elegant glowing white calligraphy with thin decorative swashes and a tiny bottle icon near the final character; place the smaller subtitle centered beneath it with thin horizontal divider lines and a small ornamental flourish. Style: highly detailed modern seinen manga cover illustration, painterly anime realism, intricate linework, moody noir atmosphere, glossy magical highlights, sharp focus on the character and bottle, dark indigo and gold palette. Canvas: vertical 9:16 cover, no border, no watermark, keep the title readable and dominant.

Create a richly detailed vertical anime illustration of {argument name="scene theme" default="two young women celebrating Tanabata on a traditional Japanese veranda"} at twilight. The setting is an old wooden house with glowing shoji doors on the left, bamboo blinds under the eaves, polished wet-looking wooden floorboards, and a lush Japanese garden on the right filled with hydrangeas, stepping stones, a stone lantern, and a deep blue-to-pink sunset sky full of stars. Show exactly 2 seated women in yukata: the left woman has blonde hair in a low bun with purple hydrangea hair ornaments, a blue floral yukata, and a yellow obi, holding one pink wish strip; the right woman has dark hair in an elegant updo with blue flower ornaments, a pale lavender yukata with blue floral patterns, and a purple obi, reaching upward to tie a wish strip onto a bamboo branch. Include exactly 1 potted bamboo Tanabata branch in a small woven stand, decorated with exactly 8 visible ornaments: 3 hanging rectangular paper strips in blue, purple, and cream, 2 pink wish strips near the women’s hands, 2 gold star ornaments, and 1 dangling geometric paper garland. Include exactly 1 glass wind chime hanging from the eaves with a floral blue pattern and a white paper strip. On the veranda foreground include exactly 5 small objects: 2 glasses of iced tea on coasters, 1 round uchiwa fan painted with hydrangeas, 1 clear plate holding a pink wagashi sweet, and 1 white ceramic mosquito-repellent incense pig decorated with blue flowers. Near the garden edge include exactly 1 pair of blue geta sandals. Use luminous cinematic lighting: warm amber light from the interior and lanterns contrasting with cool violet-blue evening light, delicate reflections on the wooden floor, soft rim light on faces and hair, highly detailed fabric folds, floral patterns, and magical starry atmosphere. The mood should be tender, quiet, romantic, nostalgic, and summery. No captions, no watermark, no modern objects, no extra people.

Create a photorealistic watercolor illustration with fine ink sketchwork of a young Taiwanese woman taking a wide-angle selfie inside a Taipei MRT train at night. She sits beside a large window on the left side of the frame, leaning toward the camera with one arm extended into the foreground, her hand resting along the window ledge and wearing a delicate ring. She has {argument name="hair color" default="dark brown"} hair loosely tied back with soft strands around her face, natural makeup, a gentle confident smile, and small earrings. Her outfit is a navy oversized blazer over a black top, paired with a light skirt or trousers, with a black structured handbag and sunglasses near her lap. The train interior should be bright and realistic: blue seats, white and silver walls, overhead rails, hanging triangular hand straps, fluorescent ceiling lights, a few seated passengers in the background, and a small digital route display showing Taipei City Hall and route code BL18. Through the window, show a vivid blue-hour Taipei skyline with Taipei 101 and surrounding illuminated skyscrapers, river or city lights below, and a clear reflection of the woman’s face and upper body layered over the night city view. Use a dynamic ultra-wide composition, slightly fisheye perspective, strong foreground hand perspective, crisp architectural lines, and transparent glass reflections. Visual style: fusion of realistic watercolor washes, visible paper texture, fine black ink outlines, crosshatching, and delicate sketch lines, while keeping the face and city lights highly detailed. Mood: calm urban nightlife, elegant, cinematic, intimate travel moment. No extra text, no watermark, no distorted hands, preserve exactly one main woman, one reflected woman in the window, and a small group of background passengers.

Create a wide horizontal anime-style Japanese mythological illustration featuring two elegant white-haired male shrine-deity figures facing inward toward the center. The left character has long pale blond-white hair tied low, sharp purple eyes, a calm sly smile, and wears a luxurious deep {argument name="left robe color" default="purple"} kimono with gold peony and cloud patterns; he holds an open folding fan decorated with cream peonies and dark ribs. The right character has very long silver-white hair with braids, ornaments, dangling red cords, a radiant crown-like hairpiece, pale intense eyes, and wears an ornate ivory-white ceremonial robe layered with gold chains, red tassels, floral motifs, and fine embroidery; he holds an open cream-and-gold folding fan across his chest. Place them symmetrically on the left and right sides with their bodies angled inward, occupying most of the frame. In the center, add large vertical Japanese calligraphy reading {argument name="central calligraphy" default="阿吽"} in rough metallic gold brush strokes. Behind the text, depict a huge golden sun disk with a smaller dark-gold circular disk and many radiating black and gold rays, like a sacred halo. Across the upper background, stretch one thick twisted shimenawa rope from left to right with tassels hanging at both far corners. The background is deep black-blue with antique gold texture, red and gold swirling clouds, scattered white petal-like shapes, and dense ukiyo-e decorative patterns. Use a lavish Edo-period fantasy aesthetic, highly detailed line art, metallic gold leaf texture, ornate kimono fabric patterns, dramatic symmetry, refined bishonen character design, cinematic composition, no modern objects, no watermark, and keep the image in a 16:9 landscape ratio.

Create an ultra-realistic vertical smartphone portrait photo of an adult East Asian woman in her early 20s, leaning forward across a café table and looking directly into the camera with a soft, friendly closed-mouth smile. She has {argument name="hair color and style" default="short dark brown bob hair with airy bangs and glossy highlights"}, clear warm-toned skin, subtle natural makeup, rosy cheeks, glossy lips, and large expressive brown eyes. She is wearing {argument name="outfit" default="a dark navy ribbed cardigan with a deep V neckline"} and a delicate silver necklace with a small pendant. The camera is very close, slightly above eye level, using a wide smartphone lens that creates an intimate selfie-like perspective; her face fills the upper half of the frame and her torso and arms extend toward the lower half. Place her in {argument name="setting" default="a warmly lit modern café with wooden tables, chairs, window frames, and softly blurred background patrons"}. Add a smartphone lying on the wooden table at the lower left edge, partially cropped. Use warm indoor lighting mixed with cool window light, realistic skin texture, detailed hair strands, shallow depth of field, natural bokeh, candid composition, high dynamic range, and photorealistic detail. Keep the image vertical 9:16, no text, no watermark, no illustration or anime styling.

Create a cinematic vertical full-body anime-realistic portrait of {argument name="character name" default="Emilia"}, a beautiful young woman at a Japanese Tanabata festival at night, standing on a narrow path lined with bamboo branches, warm fairy lights, and many hanging tanzaku wish papers. She has {argument name="hair color" default="long blonde"} hair in a high side ponytail tied with a black ribbon, soft bangs, large pinkish eyes, pale skin, delicate makeup, and a slightly curious upward gaze. She is reaching up with her right arm to touch or hang one visible pink tanzaku strip with handwritten Japanese-style text, while her left arm hangs relaxed by her side. Dress her in a somewhat awkwardly worn {argument name="yukata color and pattern" default="deep red yukata with pale pink floral prints"}, loose short hem, wide sleeves, and a {argument name="obi color" default="mustard yellow obi"} tied around the waist; show one decorative white-and-red thigh garter on her right thigh. The setting is an intimate bamboo-wish tunnel with exactly three prominent categories of repeated decorations: many green bamboo leaves overhead and along both sides, numerous small warm glowing bulbs forming golden bokeh, and multiple pastel tanzaku strips in pink, cream, yellow, and blue hanging vertically. Background includes a softly blurred urban night skyline with bridge or riverside lights, strong depth of field, warm amber lighting on the subject, dark blue night sky, realistic skin and fabric texture, high-detail gpt-image-2 style, shallow focus, 9:16 composition, slightly tilted dynamic camera angle, subject centered slightly lower than the raised hand, romantic festival atmosphere, no watermark, no extra people, no readable modern signage.

Create a highly polished vertical anime illustration of a beautiful young woman at a summer Tanabata festival, shown in a close three-quarter portrait from the waist up. She has {argument name="hair color" default="light ash-brown blonde"} hair styled in a soft loose updo with airy bangs and wispy strands, decorated with small white and indigo flowers and delicate bead chains. She has large sparkling green eyes, a gentle shy smile, soft blush, and elegant dangling earrings with blue teardrop gems. She wears a deep navy yukata with pale floral patterns and subtle gold line accents, tied with a muted gold obi visible at the back. She holds one vertical paper wish strip in both hands near the foreground; the strip contains handwritten Japanese text reading {argument name="wish text" default="みんなが幸せでありますように♡"}. The setting is a romantic evening festival under a star-filled blue sky with a bright Milky Way, warm lantern bokeh, bamboo leaves and hanging tanzaku wish papers on the left side, and a distant bridge or riverside railing with a small silhouetted couple in the background. Use cinematic backlighting and rim light from the upper left, glowing warm highlights on hair and face, luminous eyes, detailed fabric folds, delicate linework, shallow depth of field, dreamy atmosphere, rich contrast between cool indigo night tones and warm lantern light, ultra-detailed modern anime style, portrait orientation 2:3. Avoid extra characters in the foreground, avoid distorted hands, and keep the Japanese text legible on exactly one main wish strip.

Goal: Create a wide two-panel comparison image of an arctic warship harbor at night, separated by a thick vertical black divider, with subtle black margins around the whole composition.

Canvas: Ultra-wide horizontal canvas, about 3:1 aspect ratio. Split the image into exactly 2 equal rectangular panels: the left panel is dark, cinematic, realistic concept art; the right panel is brighter, stylized, clean anime/cartoon concept art.

Left panel: Show exactly 1 massive futuristic battleship moored in an icy industrial harbor, viewed from a low front-left angle. The ship is dark gray, imposing, with a tall central superstructure, radar masts, antennas, and large forward naval guns. The water is almost frozen, with broken ice floes in the foreground and reflections on the black-blue surface. Include exactly 3 visible harbor cranes as dark silhouettes: 1 large crane on the far left, 1 crane near the right edge, and 1 smaller industrial crane behind the ship. Add distant dock structures, piers, and faint city/port lights. The sky is deep midnight blue with stars and a dramatic blue aurora borealis sweeping vertically and horizontally behind the ship. Mood: cold, ominous, militaristic, high-detail sci-fi realism.

Right panel: Show exactly 1 white icebreaking patrol warship in a frozen harbor, viewed from a high three-quarter front-left angle. Paint it in a crisp illustrated style with clean outlines, cel-shaded colors, and detailed ship equipment. The ship name on the hull must read {argument name="ship name" default="NORTHERN GHOST"}. The ship has a tall bridge with glowing cyan windows, radar domes, antennas, mast equipment, side lifeboats, deck guns, missile pods, railings, hatches, and warm yellow-lit interior openings. It is surrounded by cracked ice sheets and snow-covered docks. Include exactly 3 visible harbor cranes in the background on the left side, plus small warehouses and pier lights. Add a full moon in the upper-right sky with moonlight reflected on the icy sea. Mood: adventurous, cute but technical, blue winter palette, clean worldbuilding illustration.

Style contrast: The left panel should feel like {argument name="left panel style" default="dark realistic cinematic naval sci-fi concept art"}; the right panel should feel like {argument name="right panel style" default="bright anime-style arctic ship illustration"}. Keep both panels depicting the same basic subject, {argument name="subject" default="an advanced arctic warship in a frozen naval port"}, but with clearly different rendering styles.

Color and lighting: Use a cold palette of navy, steel gray, icy blue, and white. Left panel lighting is low-key and dramatic, dominated by blue aurora glow. Right panel lighting is clearer and more readable, with moonlight, cyan bridge lights, and warm yellow accents.

Constraints: Include no people, no captions other than the hull text, no logos, no watermark. Maintain exactly 2 panels and exactly 1 primary ship per panel. Emphasize frozen water, broken ice floes, industrial harbor machinery, and night atmosphere.

Create a vertical 9:16 ultra-detailed photorealistic anime-style portrait of a beautiful young Japanese woman at a Tanabata night festival, wearing a deep navy-blue yukata with purple bellflower motifs and tiny star-speckled patterns. She has long glossy black hair tied in a high side ponytail with a small purple floral hair ornament, soft bangs, large expressive brown eyes, smooth fair skin, and a gentle confident smile. Pose her standing three-quarter view, one arm raised with her hand touching the top of her ponytail and the other hand lightly holding the edge of her collar, creating an elegant emotional ponytail-themed composition. The yukata should have a low neckline, wide sleeves, a high side slit revealing one leg, and a metallic silver obi tied at the waist with trailing pale lavender ribbons. Place her in a dreamy bamboo grove at night under a starry sky and visible Milky Way, with Tanabata wish papers hanging from bamboo branches in multiple colors, warm lanterns glowing in the background, and a foreground bed of luminous purple bellflowers. Use cinematic bokeh, soft rim lighting on her hair, magical blue-purple firefly-like sparkles, shallow depth of field, highly polished glossy illustration rendering, realistic fabric texture, and a romantic festival atmosphere. Add two visible text overlays: a semi-transparent “CapCut AI” watermark in the upper left and a pink-white signature “@Me_Rock369” in the lower right. Main customizable details: {argument name="character name" default="unnamed young Japanese woman"}, {argument name="yukata color" default="deep navy blue"}, {argument name="flower motif" default="purple bellflowers"}, {argument name="festival setting" default="Tanabata night bamboo grove"}, {argument name="signature text" default="@Me_Rock369"}.

Goal: Create a polished anime-style Tanabata night illustration featuring {argument name="character name" default="WAGAHAI"}, a thoughtful young girl in a yukata gazing up at the Milky Way and wondering about Hikoboshi.

Canvas: Wide 16:9 horizontal illustration, cinematic composition, deep blue starry night palette, high detail, soft glowing lights, magical summer-festival atmosphere.

Layout: Place the girl on the right third of the image from waist up, facing left and looking upward with a contemplative expression, one finger resting near her cheek. Fill the left and far-right edges with tall bamboo branches framing the scene. Put the main glowing Japanese headline in the center-left sky. In the distant background, show a river, mountain silhouettes, and a small arched bridge lined with warm festival lights. Add faint celestial silhouettes of Orihime and Hikoboshi floating near the upper center of the sky, plus two shooting stars.

Main subject details: The girl has long straight {argument name="hair color" default="black hair"}, large round blue eyes behind round glasses, fair skin, and a white headband. Add a decorative hair ornament on the right side of her head with beige leaves and a red ribbon bow. She wears a dark navy-blue starry yukata patterned like a galaxy, with a small illustration of Orihime on the chest, a wide yellow obi, and red obi cords. Her mood is quiet, wistful, and romantic.

Tanabata decorations and counted elements: Include exactly 14 hanging paper wish strips in total: 8 on the left bamboo cluster, 1 small vertical strip near the central hanging lantern, and 5 on the right-side bamboo/yukata area. Include exactly 4 tall cylindrical Tanabata lanterns with streamers: 2 on the left side, 1 near the upper center-left, and 1 on the far right. Include exactly 6 glowing star-shaped ornaments hanging among the left bamboo. Include exactly 4 rectangular floor or railing lanterns glowing warm yellow along the bottom edge. Include exactly 1 large round creator-logo badge in the top-left corner reading “AiRT AI ART CREATOR for children/images.”

Text content: Add the central glowing handwritten Japanese phrase {argument name="headline text" default="私の彦星さん、どこにいるのかな？"} in pale pink-white neon lettering, with a decorative heart-shaped light trail and curved underline below it. The tanzaku strips may contain small vertical Japanese wish text, delicate and partially readable, but the headline must be clear.

Visual style: High-quality Japanese anime illustration, luminous fantasy realism, detailed star field with Milky Way, soft bloom, clean character line art, rich blues and purples, warm lantern contrast, reflective river water, romantic Tanabata festival mood. Avoid photorealism, avoid extra characters in the foreground, avoid cluttering the central headline.

Create a cinematic vertical photograph of {argument name="subject" default="a teenage schoolgirl"} walking toward the camera inside a dim concrete parking garage at dusk. She is centered slightly left, captured mid-stride with natural motion blur: long {argument name="hair color" default="black"} hair whipping sideways, her right hand raised in front of her face so it partially hides her features, creating strong foreground occlusion. She wears a crisp white button-up shirt with sleeves slightly loose, a blue plaid necktie hanging casually, a dark navy pleated skirt, black knee-high socks, and black loafers; she carries a dark blazer or jacket bunched in her left arm. The environment is an empty multi-level garage with rough concrete floor, square pillars, low ceiling panels, exposed pipes, fluorescent strip lights overhead, and open sides showing a blue-purple evening sky and distant blurry parking-lot lights. Use a handheld candid fashion-photo feel, full-body framing from slightly below eye level, shallow depth of field, realistic physics in the clothing and hair, soft low-contrast lighting mixed with harsh fluorescent highlights, subtle film grain, muted colors, and a moody urban coming-of-age atmosphere. No text, no watermark, no posed studio look.

Create a wide cinematic editorial collage about {argument name="composer name" default="Gabriel Fauré"} and the concept of a gentle requiem rather than divine punishment. Use a 16:9 horizontal canvas with a dramatic split composition: the left third is dark, black and crimson, with a realistic sepia portrait of an older 19th-century French composer in a black suit, white shirt, curled gray hair and mustache, looking upward in profile toward the center. Behind him, show a ruined Gothic church arch with a glowing red circular stained-glass window and a shadowy statue of justice holding scales, creating an atmosphere of judgment and wrath. The center of the image is a luminous monumental funerary cathedral or mausoleum made of pale marble and translucent sheet music, with a grand arched entrance, stairs leading upward, a glowing golden light inside, a large domed roof, and many vertical organ pipes rising around it like architecture. The right third becomes soft ivory and gold: include a large classical marble angel or mother figure above, tenderly holding a child, and below it a peaceful sleeping young person wrapped in white fabric and clouds, suggesting death as sleep. Add flowing golden musical staff lines and notes sweeping diagonally from the central monument across the sleeping figure, tying the composition together. Use exactly four main symbolic zones: 1) composer portrait, 2) dark Catholic judgment scene, 3) central requiem monument with organ pipes and sheet music, 4) serene angelic sleep scene. Visual style should be high-end historical photomontage, baroque sacred art, marble sculpture, antique paper texture, dramatic chiaroscuro on the left, soft heavenly diffusion on the right, warm gold accents, subtle dust, cracks, and layered geometric collage shapes. No text, no captions, no watermark, no modern objects.