GPT Image 2 prompts

Create a cozy cinematic vertical photograph inside a retro luxury cream-soda specialty café with a deep forest-green color theme. In the foreground, a pale hand wearing a dark green velvet sleeve with a cream lace cuff holds exactly three rectangular color swatch cards in front of a tall glass of bright emerald green cream soda: the swatches are labeled “Forest Green” on a dark green card, “Neon Green” on a vivid lime-green card, and “Cherry Red” on a deep red card, each with a white label strip and simple black sans-serif text. The drink is in a clear fluted parfait glass on a small ornate round table, filled with green soda, ice cubes, bubbles, topped with whipped cream, tiny green sprinkles, and one glossy red cherry with a long stem. Background: softly blurred vintage café lounge with dark green velvet sofa seating, a small white teddy bear sitting on the sofa, framed menu signs on cream walls, dark green curtains, warm wall sconces, and a green-shaded table lamp glowing warmly. Use shallow depth of field, nostalgic film photography, warm tungsten lighting, elegant pre-production color-reference mood, slight grain, realistic textures, and a 9:16 vertical composition. Customize the scene as {argument name="café concept" default="hypothetical cream-soda specialty café"}, the main drink color as {argument name="soda color" default="emerald green"}, the three swatch labels as {argument name="color swatch labels" default="Forest Green, Neon Green, Cherry Red"}, the garnish as {argument name="garnish" default="whipped cream, green sprinkles, and a red cherry"}, and the atmosphere as {argument name="atmosphere" default="cozy retro luxury café lounge"}.

Create a realistic studio portrait photograph of a young East Asian woman from the upper chest up, framed vertically in a 3:4 crop. She is turned slightly to the viewer’s left in a three-quarter pose, showing one visible ear and the side of her neck, with her face centered but intentionally covered by a large opaque square privacy block. She has {argument name="hair color" default="natural black"} hair styled in a loose low bun at the back, with soft bangs across the forehead area and a few wispy flyaway strands. She wears a simple {argument name="top color" default="white"} camisole with thin spaghetti straps, bare shoulders, natural skin texture, and soft collarbone highlights. Use a warm matte {argument name="background color" default="pale yellow"} seamless studio background. Lighting should be soft, flattering, and diffused, with gentle shadows, shallow depth of field, and a high-end AI photography look. Add exactly 1 censoring element: a large square block over the face, aligned vertically and horizontally over the central face area, filled with a smooth muted skin-tone vertical gradient from dark brown at the top to peach-beige at the bottom, with sharp edges and no text. Keep the composition clean and minimal, no jewelry, no props, no watermark, no extra text.

Create a cinematic dark gothic horror scene of an abandoned small-town railway station at night during a snowstorm. A lone {argument name="main character" default="young woman"} stands in the lower-left foreground on a wet, icy platform, seen from behind, wearing a long dark coat and holding a small suitcase at her side, staring down the tracks. The station building on the far left is decayed and partly in silhouette, with peeling paint, a hanging lantern, an old round station clock, and a readable sign that says {argument name="station sign" default="RAVENCROFT"}. Snow-covered train tracks begin near the bottom center and recede into the foggy distance, guiding the eye toward a haunted Victorian house in the mid-background with one glowing window. On the right side, include exactly three major background structures: a broken wooden house, a tall weathered utility pole, and a low chain-link fence with dead winter brush. The atmosphere should be bleak, suspenseful, and abandoned, with drifting snowflakes, heavy clouds, cold blue-gray moonlit lighting, wet reflective cobblestones, deep shadows, mist, and subtle volumetric light. Use a wide cinematic 4:3 composition, realistic film still style, high detail, moody contrast, no modern vehicles, no extra people, no bright colors, no watermark.

Create a dramatic cinematic sci-fi illustration of a {argument name="character description" default="female android astronaut"} kneeling on the exterior hull of a massive spacecraft in orbit above Earth, performing emergency maintenance with a handheld plasma welding torch. She has a sleek white-and-silver biomechanical body suit integrated with robotic joints, segmented armor plates, visible black cables, tiny bolts, micro-panels, and subtle weathering; her long {argument name="hair color" default="silver-white"} hair streams wildly backward in zero gravity. Her face is covered by a plain opaque square censor block in muted gray, centered over the head area. She is posed on one knee with the other leg bent, leaning toward a vertical wall of dark spacecraft hull panels while holding the torch in both hands. The torch emits an intense blue-white point of light and a spray of orange sparks against the wall. The background shows the curved blue edge of Earth with glowing city lights below, deep black space filled with stars, and the enormous cylindrical spaceship surface receding into perspective. Add readable industrial markings on the wall such as {argument name="hull label" default="HULL SECTOR 12-A MAINTENANCE ACCESS"}. Style: ultra-detailed realistic concept art, high contrast, cold blue rim lighting, orange spark highlights, glossy metal reflections, hard-surface sci-fi design, shallow depth of field, dynamic diagonal composition, 8k level detail, no cartoon styling, no extra characters, no watermark.

Goal: Create a clean square fashion product advertisement for an urban sling bag, using a minimalist Japanese e-commerce/catalog aesthetic.

Canvas: 1:1 square image, bright white to very light gray studio background, high-key lighting, sharp commercial photography, no borders or watermark.

Layout: Split composition with large typography on the left third and a male model on the right two-thirds. The model is cropped from upper thighs to just above the hair, standing relaxed with both hands in pants pockets. Leave generous negative space around the left text.

Subject details: A young adult East Asian male model with short slightly wavy black hair, face intentionally anonymized by a smooth rectangular skin-tone blur block. He wears an oversized plain {argument name="shirt color" default="white"} crew-neck T-shirt and loose dark trousers. Across his torso is a {argument name="bag color" default="black"} triangular urban sling bag worn diagonally from his right shoulder to his left waist. The sling bag has a wide black shoulder strap with buckle hardware, a clean matte fabric texture, one diagonal front zipper with a black pull cord, a small side zipper pull, and a tiny rectangular brand tag near the lower right edge. The bag should be the visual focus.

Text content: On the left, include exactly 4 text groups: 1) a large stacked uppercase headline reading {argument name="headline text" default="URBAN\nSLING\nSTYLE"}, set in bold modern sans-serif with wide letter spacing; 2) a smaller two-line Japanese feature statement reading {argument name="feature text" default="軽量 × 6L収納\n× タブレット対応"}; 3) a short thin horizontal divider line; 4) a small product code reading {argument name="product code" default="RGBR-11"}. Keep the text dark charcoal, aligned left, and spaced like a premium catalog poster.

Visual style: Photorealistic studio product photography, crisp fabric detail, soft shadows, monochrome black-and-white styling, modern Japanese streetwear catalog design, premium minimalism.

Constraints: Use exactly one model, one sling bag, one T-shirt, one pair of trousers, and four text groups. Do not add extra accessories, logos, scenery, props, decorative graphics, or additional text.

Goal: Create a vertical anime-style promotional profile/novel-cover image featuring a beautiful serious female teaching intern, with her face deliberately hidden by a solid rectangular privacy block.

Canvas: Portrait 2:3 aspect ratio, high-resolution, full-bleed illustration. The character is a close-up bust portrait centered in the frame, filling most of the canvas from the top hairline to the lower chest.

Subject details: A young adult Japanese woman, elegant and composed, with very long glossy black hair parted in the center. Her hair has detailed anime highlights, layered strands, slight waves near the sides, and falls over both shoulders. Visible ears peek through the hair. She wears a dark blazer over a white collared shirt, suggesting a school or teaching-intern setting. Cover the entire central face area with one large opaque rectangular block in muted brown-gray, placed horizontally across the eyes, nose, and mouth, leaving only the forehead, hair, ears, jawline edges, neck, and clothing partly visible.

Background: Soft dreamy pastel bokeh background in lavender, pink, and pale blue, with a bright vertical glow behind the head. Add sparkling starbursts and floating light particles throughout. Include exactly 7 visible pink flower petals: 1 near the upper left edge, 1 near the upper right edge, 2 on the left side, 1 on the right side near the hair, 1 over the notebook area, and 1 near the lower left hair area.

Foreground typography and objects: At the bottom, create a dramatic Japanese light-novel title layout with layered decorative lettering, sparkles, hearts, and school stationery. Use exactly 4 main text groups: large pink-and-white 「美人」 on the lower left with glowing outline and sparkle accents; small black 「の」 inside a white circular badge; blue 「教育実習生」 across a white lined notebook-paper shape; and large pink 「真面目」 with black handwritten-style surrounding text 「が」 and 「すぎる」. The full visible title should read 「美人の教育実習生が真面目すぎる」. Add a white notebook-paper panel behind the blue text, with faint horizontal ruled lines and blue binder holes on the right edge. Add one blue pen with gold trim at the lower right, angled diagonally. Add small pink heart doodles connected by a dotted flourish near the pen.

Visual style: Polished Japanese anime key visual, glossy hair rendering, romantic shoujo light-novel cover aesthetic, high contrast between dark hair and pastel sparkling background, crisp clean linework, luminous highlights, decorative glow effects around the title.

Parameters: The woman should be {argument name="character type" default="a beautiful serious female teaching intern"}; hair should be {argument name="hair color" default="glossy black"}; title text should be {argument name="Japanese title text" default="美人の教育実習生が真面目すぎる"}; privacy block color should be {argument name="privacy block color" default="muted brown-gray"}; background palette should be {argument name="background palette" default="lavender, pink, and pale blue"}.

Constraints: Keep the face fully obscured by the single rectangle, do not show eyes or mouth, do not add extra characters, preserve the vertical poster composition, and keep all visible text in Japanese exactly as specified.

Create a square anime illustration in a high-contrast monochrome manga style with vivid hot-pink accents. Show {argument name="character name" default="a mysterious young woman"} seated sideways on a wooden chair at a rustic counter inside a dim retro cyberpunk tea room or café. She has very long {argument name="hair color" default="silver-white"} hair tied into a high ponytail that cascades behind her, pale skin, sharp bangs, and one glowing pink eye visible as she gazes to the left with a calm, aloof expression, resting her chin on one hand. Dress her in a bright {argument name="kimono color" default="hot-pink"} kimono-style top with wide sleeves, a light obi bow at the waist, and a long pale pleated skirt. The café background is dense with dark wooden panels, framed futuristic control screens, tiny interface diagrams, signs, cables, and a hanging lamp with a glowing pink bulb. On the counter place exactly five tea items: one large white teapot on the far left, two small teacups near it, one steaming mug in front of the character, and one small teapot or kettle on a rear shelf to the right. Surround the composition with exactly four clusters of white lilies: upper left corner, upper right corner, lower left corner, and lower right edge. Add sweeping ribbon-like pink energy trails and brush strokes crossing the image diagonally, with several white streaks and ink splatters, giving motion and magical atmosphere. Use thick black linework, sketchy crosshatching, limited palette of black, white, gray, and neon pink, detailed manga shading, dramatic lighting, slightly tilted perspective, ornate cluttered background, no readable modern watermark, no extra characters.

Create a highly detailed square anime illustration of {argument name="character name" default="a moonlit beauty"} seated in profile inside an ornate dark Chinese-style teahouse at night. The main subject is an elegant young woman with very long {argument name="hair color" default="silver-white"} hair tied in a high ponytail, loose flowing strands, floral hair ornaments, tassel earrings, pale skin, and glowing pink eyes. She sits gracefully on a carved black wooden chair, turned slightly toward the viewer, resting one hand near her lips with a calm, mysterious expression. Dress her in a vivid {argument name="robe color" default="hot pink"} embroidered kimono/hanfu-style robe with delicate pale floral patterns, layered white inner fabric, sash details, and soft folds across her lap. Use a mostly monochrome black, charcoal, white, and gray palette with saturated neon magenta accents.

The environment should be richly decorated: dark carved wooden wall panels, framed floral paintings, shelves with teaware, a round pink hanging lantern in the upper right, a low tea table with exactly 5 visible tea-service items: 1 white teapot, 2 teacups, 1 saucer/plate, and 1 small lidded bowl. Add steam curling upward from one cup. Include exactly 3 visible text signs in glowing pink Chinese characters: a large vertical sign on the right reading {argument name="main sign text" default="月下美人"}, a smaller vertical sign on the left reading {argument name="left sign text" default="花開富貴"}, and a tiny red seal-like mark beneath the main sign. Surround the composition with exactly 4 prominent white lilies in the foreground, plus clusters of small white blossoms near the right side.

Visual style: polished dark fantasy anime, ultra-detailed linework, dramatic rim lighting, glossy highlights, deep shadows, intricate ornamental patterns, cinematic composition, high contrast. Add multiple neon pink magical light trails: one crescent arc above her hair, one spiral around the foreground lilies, one horizontal swirl near the right wall, and one sweeping ring across the lower foreground. Scatter small pink glowing particles throughout the room. Keep the image square, centered on the woman, with no modern objects, no watermark, and no extra readable text beyond the specified signs.

Create a detailed anime-style fantasy illustration of two elegant characters having breakfast together in a sunlit gothic library. The scene shows {argument name="character pair" default="a petite young woman with very long black hair and a tall silver-haired man"} seated at a wooden breakfast table, both with their faces intentionally obscured by soft rectangular blur masks. The woman sits on the left wearing a deep navy silk robe or kimono-inspired morning gown with black lace trim, a blue gemstone necklace, earrings, and an ornate blue flower hair accessory; she holds a fork in one hand and rests her cheek on the other hand in a relaxed morning pose. The man sits on the right, much taller and broad-shouldered, with spiky pale silver hair, earrings, a pendant necklace, an open black shirt, and a dark embroidered robe with silver ornamental patterns; he holds a plate of scrambled eggs in one hand and a spoon in the other as if serving or eating. The table contains exactly 12 visible breakfast and table items: 1 plate of scrambled eggs with tomato and greens in front of the woman, 1 plate of scrambled eggs held by the man, 1 cup of coffee or tea on a saucer, 1 white lidded teapot or sugar pot, 1 small clear glass bottle, 1 small ornate metal condiment holder, 1 plate of sliced bread, 1 platter with greens, sliced meat, and a halved boiled egg, 1 glass bowl of green and dark purple grapes, 1 basket of croissants, 1 large round loaf or pastry at the front right, and 1 brown teapot at the far right. Add a folded newspaper on the table with the readable English headline {argument name="newspaper headline" default="CELESTIAL LIBRARY"}. The background is an ornate old-world library with tall bookshelves, carved wood furniture, gothic arched windows, long blue-and-gold hanging banners, a decorative sign that also reads Celestial Library, vases of purple flowers, small bottles and curios, and warm morning sunlight streaming through the windows. Use a rich painterly anime rendering style with intricate linework, cinematic lighting, warm gold highlights, cool blue shadows, soft dust-filled atmosphere, high detail in fabrics and food, and a cozy yet aristocratic fantasy mood. Use a horizontal 4:3 composition, waist-up framing of the two characters, the table in the foreground, and the library receding behind them. Avoid modern objects except the newspaper, avoid extra people, and keep the image text minimal and mostly decorative.

Using REFERENCE_0 as the character, cat, and soft anime illustration style base, create a different composition: change the canvas to a horizontal living-room scene where the girl sits curled up on the sofa with her knees hugged close, turned slightly toward a large bright window. Keep the relaxed white loungewear mood and the orange tabby cat, but make the cat smaller and sleeping beside her on the sofa. Emphasize {argument name="morning atmosphere" default="dazzling sunrise backlight and a fresh,爽やかな morning feeling"}: strong warm sunlight flooding in from the right, soft overexposure near the window, gentle rim light, and airy dust motes. Add sheer white curtains billowing softly in the breeze. Add a calm interior background not present in the reference: exactly 2 visible plant areas, one leafy plant and small potted plant on a rear shelf at left, and one potted plant on a low table at right; also include a framed wall picture, a mug, and stacked books. Keep the image quiet, cozy, pastel, and cinematic, with delicate linework and soft shadows. Avoid adding text or extra characters.

Create a cinematic anime VTuber-style half-body portrait of {argument name="character name" default="a gothic cyber-diva"} standing centered in a dark cathedral-like server hall. The character has very long glossy {argument name="hair color" default="black hair with subtle purple strands"}, straight bangs, pale skin, and her face is intentionally covered by a soft blurred gray square placeholder, as if the facial area is masked for later Live2D design. She wears an ornate black gothic off-shoulder dress with sheer translucent sleeves, lace trim, a fitted bodice decorated with gold constellation, star, moon, and chain motifs, a black belt with a gold buckle, layered necklaces with green jewels and star charms, and a gold star hair ornament with dangling emerald accents on one side. Show exactly 8 notable outfit/accessory elements: 1 black off-shoulder lace dress, 2 sheer sleeves, 3 gold constellation embroidery, 4 black belt with gold buckle, 5 layered jewel necklaces, 6 dangling cross/star chest charm, 7 gold star hairpin, 8 green gem earrings or hair-chain accents. The background is a vast ruined cyber-gothic cathedral interior with high arches, tall columns, dark stone, server racks glowing with green lights on the left, hanging streams of luminous green cables from the ceiling and side walls, smoky mist, a warm circular skylight beam in the far upper center, and a thin neon-green horizontal ring of light passing behind the character at chest height. Use a 16:9 widescreen composition, centered symmetrical framing, dramatic low-key lighting, teal-green cyberpunk glow mixed with warm amber backlight, high detail, polished anime illustration blended with realistic architectural atmosphere, mysterious music-game / AITuber key visual mood. No text, no watermark, no extra characters.

Create a vertical, richly detailed anime storybook illustration of cute pet birds mining emerald crystals on a sunny mountain cliff. The scene is a cheerful fantasy mining site with rocky terrain, sparkling green gems, wooden equipment, and a bright blue sky. Use a warm, polished Japanese anime style with soft outlines, expressive faces, luminous highlights, and high-detail natural textures.

Canvas: vertical 2:3 composition, tall poster framing, bright daylight. Put the sun in the upper left with lens flare, blue sky and white clouds behind jagged green mountains. The rocky cliff rises up the right side and background, with a dark mine entrance in the mid-right rock face and a small wooden lookout platform near the upper right.

Main subject: In the center foreground, place a joyful cream cockatiel with yellow crest and orange cheeks, wearing a miner belt with small tools and pouches, holding up one oversized faceted emerald crystal that sparkles with a starburst. This central bird is the largest figure and the focal point.

Bird count and roles: Show exactly 9 cute birds on the ground/cliff plus 4 small seabirds flying in the sky. The 9 main birds are: 1 central cream cockatiel proudly lifting the giant emerald; 1 blue-and-white budgie on the right reading an old treasure map with a red X; 1 dusty pink Java sparrow on the left drinking from a blue water bottle; 1 small pale yellow chick at lower left-center wearing a yellow scarf and holding a pickaxe; 1 teal-and-blue budgie at the bottom left wearing a tiny backpack; 1 green peach-faced lovebird at lower left holding a hammer; 1 green peach-faced lovebird at lower center-right using a pickaxe on the stones; 1 small white cockatiel on the right with a pink flower near its head, standing by a mine cart full of gems; 1 tiny white bird wearing a yellow miner helmet on the high wooden platform.

Environment and objects: Include a rustic wooden crane on the left with ropes and a suspended metal cage bucket. Include exactly 2 wooden mine carts or crates filled with emeralds: one near the right bird and one at the lower right. Include 1 hanging lantern glowing at the bottom right, 1 wooden sign at bottom left with a simple crystal drawing and small bird doodle, 1 broken wooden rail or beam across the bottom foreground, and 1 coiled rope tied to a post. Place emerald crystals throughout the scene: 1 giant emerald in the central bird’s wings, 1 large embedded emerald on the right cliff, 1 medium emerald cluster beside the mine entrance, 1 crystal cluster in the right mine cart, 1 crystal cluster in the lower-right crate, 1 cluster near the green bird with the hammer, 1 cluster near the green bird with the pickaxe, 2 clusters along the lower foreground, and small green sparkle flecks on the cliff.

Mood and lighting: Make everything whimsical, adventurous, and optimistic. Crystals should glow translucent emerald green with bright white specular glints. Use sunlit rim lighting, soft shadows, warm highlights on feathers, and detailed rocky textures. Keep the birds round, adorable, and expressive, with oversized glossy eyes and tiny feet.

Customizable details: The main crystal color should be {argument name="crystal color" default="emerald green"}. The central bird species should be {argument name="main bird species" default="cream cockatiel"}. The setting should be {argument name="setting" default="sunny mountain emerald mine"}. The art style should be {argument name="art style" default="polished Japanese anime storybook illustration"}. The mood should be {argument name="mood" default="cheerful treasure-hunting adventure"}.

Constraints: no humans, no text labels except the simple crystal drawing on the sign, no watermark, no photorealism, no dark horror tone, keep the composition dense but readable.

Create a polished anime-style vertical portrait of a high school girl in a close-up three-quarter side view, cropped from upper chest to above the head. She has long slightly messy dark navy-black hair with thin loose strands blowing across her face and shoulders, with soft pink rim light on the left and cool blue shadows on the right. Her face is deliberately covered by a large opaque muted mauve square censor block centered over the facial area, hiding all facial features. She wears a Japanese school uniform: dark blazer, white shirt collar, red ribbon bow tie, light gray hoodie layered under the blazer with the hood visible behind her neck, and a deep blue backpack with visible shoulder straps and rounded bag body at the lower right. One hand grips the front strap or lapel near her chest. Use delicate line art, painterly cel shading, high-detail hair strands, soft backlighting, pale white background, gentle spring atmosphere, and a clean modern anime illustration style. Keep the composition intimate and vertical, with no text, no watermark, and no extra characters.

Create a polished anime-style illustration of {argument name="character name" default="a young maid girl"} sitting on a soft patterned rug in a bright, cozy living room, viewed from a high slightly overhead angle in a vertical portrait composition. She has {argument name="hair color" default="dark brown almost black"} hair in exactly 2 high twin ponytails with loose strands, wearing a classic black-and-white frilly maid outfit with exactly 7 main visible outfit elements: a white ruffled maid headband, a black puff-sleeve dress, a white pleated bib front, a large black bow at the collar, a white apron with ruffled edges, white thigh-high stockings with lace tops, and glossy black Mary Jane shoes with buckles. Her face is intentionally covered by a centered opaque square censor block in a muted brown-to-beige vertical gradient, hiding all facial features. Pose her seated demurely with knees bent, one hand resting on her skirt and one hand touching the floor, with the skirt and apron spread around her. The room should feel warm and sunlit, with golden afternoon light pouring through a window at the upper right, soft bloom, dust sparkles, and a faint rainbow lens flare. Include exactly 6 notable background props: a round white coffee table, a glass of orange drink on the table, a small book or notebook on the table, a low shelf with books and a basket, a small green potted plant, and a pink cushion or sofa at the right. Use delicate pastel colors, detailed fabric folds, soft shading, clean linework, and high-quality Japanese anime illustration aesthetics. Make the mood gentle, intimate, and cozy, with no text, no watermark, and no extra characters.

Create a luxurious pastel product photograph of an open high-end chocolate gift box themed around an ethereal ocean princess. The scene is a dreamy macro-style tabletop arrangement in pale aqua blue, pearl white, blush pink, and champagne gold, with soft diffused lighting, shallow depth of field, glossy reflections, and sparkling bokeh. The open rectangular box is light blue with gold trim and a hinged illustrated lid. On the inside lid, show a delicate anime-style mermaid-like girl with long flowing iridescent white-blonde hair, bare shoulders, pearl necklace, translucent blue shell accents, coral and bubbles around her, and a plain pale square censor/placeholder block covering the center of her face. The lid design has ornate gold corner filigree, small shell motifs, underwater bubbles, pink coral, and elegant gold branding text: Japanese hiragana “のぞむ”, large serif “NOZOMU”, and smaller uppercase “LIMITED CHOCOLAT COLLECTION”. In the lower half, show exactly 15 individual luxury chocolates arranged in a 3-row by 5-column gold compartment tray, each in a brown paper cup: row 1 has a faceted blue-purple jewel dome, a pale aqua shell chocolate, a white pearl-topped gold-speckled dome, a blue crystal pyramid bonbon, and a dark blue round bonbon with pink coral sprinkles; row 2 has a white gold-speckled dome, a glossy dark blue-purple orb, a blue dome with pink coral sprinkles, a white ridged shell, and a translucent pale blue faceted gem; row 3 has a pale blue heart, a ribbed aqua shell with small pearls, a turquoise faceted gold-flake jewel, a pale blue dome topped with a pearl, and a teal-gold faceted gem. Surround the box with satin blue ribbons printed with subtle gold lettering and shell icons, pearl strands, glass bubbles, seashells, small pink coral pieces, crystal ornaments, and glittering highlights. Add a small matching square card in front of the box with ornate border and the same branding text “のぞむ”, “NOZOMU”, “LIMITED CHOCOLAT COLLECTION”. Use premium Japanese confectionery packaging aesthetics, ultra-detailed textures, photorealistic product photography mixed with soft anime illustration on the lid, elegant composition, no extra text, no watermark. Optional custom details: {argument name="brand name" default="NOZOMU"}, {argument name="Japanese brand text" default="のぞむ"}, {argument name="collection subtitle" default="LIMITED CHOCOLAT COLLECTION"}, {argument name="main color palette" default="pale aqua blue, pearl white, blush pink, champagne gold"}, {argument name="lid character" default="ethereal ocean princess with iridescent white-blonde hair"}.

Goal: Create a loud, hyper-saturated Korean seafood restaurant advertising poster for {argument name="restaurant theme" default="fresh natural flatfish sashimi and giant lobster"}, with an energetic travel-at-the-sea mood, aimed at attracting customers to a sashimi restaurant.

Canvas: Vertical 4:5 poster, full-bleed, high contrast, glossy commercial food-ad style. Use a bright tropical blue sky and deep blue ocean background with white sea spray, seagulls, a harbor breakwater on the left horizon, and explosive brush-stroke graphics.

Layout: Top third contains a huge hand-painted Korean calligraphy headline on a white and blue paint-splash background. Right side features a giant red spiny lobster emerging dramatically from ocean waves, angled upward, with long antennae crossing the sky. Bottom half is dominated by a lavish wooden platter of fresh sliced white fish sashimi over crushed ice, green garnish, seaweed, lemon slices, wasabi, and splashing water. Add a grilled or raw fish head at the lower right edge of the platter. Use diagonal composition, comic-book impact, and many overlapping stickers and banners.

Text content: Include exactly 10 distinct Korean text blocks, styled as bold brush lettering, stickers, or banners: 1) top pink brush banner: {argument name="top slogan" default="바다의 맛! 여행의 즐거움!"}; 2) huge main black-and-red headline: {argument name="main headline" default="바다여행 최고의 맛!"}; 3) small blue slanted banner under headline: "싱싱함이 입안 가득!"; 4) right flag beside lobster: "대박 맛집 발견!"; 5) yellow starburst badge on right: "초대형! 랍스터 해산물의 왕!"; 6) pink circular badge on left: "지금이 제철! 자연산 광어회"; 7) black brush box on lower left: "동해바다에서 만나는 진짜 신선함!"; 8) yellow splash headline near bottom left: "맛있다! 푸짐하다! 여행가자!"; 9) pink bottom ribbon: "잊지 못할 추억! ♥"; 10) bottom blue footer strip with three icon-separated phrases: "인생샷 명소! | 푸른 바다! | 맛있는 여행!". Also add two narrow vertical white paper strips at lower right reading "맛도 보고!" and "바다도 보고! 힐링도 하고!".

Subject details: Make the lobster extremely large, glossy, bright red-orange, covered in water droplets, with visible eyes, segmented shell, spiny legs, and dramatic claws/antennae. The sashimi platter should contain exactly 4 neat diagonal rows of translucent white flatfish slices with red accents, arranged in overlapping layers. Add exactly 3 visible lemon slices, 1 small patterned bowl of green wasabi, 1 fish head, 1 small camera icon, and at least 6 seagulls in the sky.

Visual style: Korean street-market seafood flyer, maximalist, punchy, celebratory, with thick black ink strokes, red handwritten calligraphy, yellow highlights, magenta badges, blue wave graphics, white foam, sharp cutout edges, glossy food photography mixed with illustrated poster effects. Use saturated cyan, royal blue, hot pink, lemon yellow, white, black, and vivid lobster red.

Constraints: Keep all Korean text legible and integrated into the design. Do not add English text. Avoid minimalist spacing; the poster should feel crowded, festive, and attention-grabbing. No watermark, no QR code, no phone number.

Goal: Create a humorous four-panel cosmic comic strip about a godlike universe creator oversleeping and accidentally causing a messy Big Bang.

Canvas: Wide horizontal comic page, 16:9 aspect ratio, with four rectangular panels arranged in a 2-by-2 grid. Use thick black comic borders, slight perspective wobble, and a smoky grayscale outer background with glowing specks and sketchy crosshatch marks.

Visual style: Highly detailed painterly comic art with dramatic ink outlines, cosmic fantasy lighting, saturated nebula colors, stars, galaxies, explosions, and watercolor-like textures. The tone is funny, epic, and slightly chaotic. Add a deliberate blurred square over the godlike character’s face in each panel, as if censored or anonymized.

Main character: A long-haired, bearded, elderly cosmic wizard or deity wearing ornate purple robes patterned with stars and gold embroidered cuffs. He is sleepy and irritated, shown in bed or beside the alarm clock. Use {argument name="character name" default="anonymous cosmic creator"} only as an internal character concept, not visible text.

Panel layout and exact content:
1. Top-left panel: The cosmic creator sleeps under blankets in a bed floating in deep space. Beside the bed on a small wooden table sits a large vintage twin-bell alarm clock. The clock face clearly reads “BIG BANG” in bold black letters, with Roman numerals around the edge. Background filled with colorful nebula clouds and stars. No speech bubble.
2. Top-right panel: The alarm clock rings violently with motion lines. Add exactly two comic sound effects above it: “BEEEEEP!” on the left and “BEEEEP!” on the right, in yellow bold comic lettering with black outline. The creator reaches from the right side toward a small rectangular orange snooze button on a base in front of the clock. The button label reads “SNOOZE.” The clock face still reads “BIG BANG.”
3. Bottom-left panel: The creator is back asleep on the left while the clock on the bedside table is wrapped tightly in messy bandages or cloth, muffled and restrained. The universe behind it erupts with galaxies, stars, and colorful energy swirls. Add a small beige caption box in the upper right that reads exactly: “13.8 billion years pass...”
4. Bottom-right panel: The creator is awake and exasperated, pointing or gesturing toward the cosmic chaos. Behind him is a wild explosion of galaxies, starbursts, and nebula fragments. Add a glowing neon sign on the left that reads exactly: “Big Bang is a Mess.” with “Mess.” emphasized in red. Add one speech bubble above the creator that reads exactly: “DAMN..., I hit Snooze just once...”

Text content: Include exactly these visible text elements: “BIG BANG” on the alarm clock in panels 1, 2, and partially visible in panel 3; “BEEEEEP!”; “BEEEEP!”; “SNOOZE”; “13.8 billion years pass...”; “Big Bang is a Mess.”; and “DAMN..., I hit Snooze just once...”

Constraints: Use exactly four panels, exactly one recurring main character, exactly one alarm clock, exactly one snooze button, exactly one neon sign, and exactly one caption box. Keep all text legible and spelled as specified. Do not add extra panels, extra speech bubbles, watermarks, logos, or signatures.

Use my uploaded portrait photo as the only identity reference. Preserve only my real facial features, face shape, skin tone, hairstyle, hairline, and overall personal vibe. Do not use the face of the person in the reference image. All other elements — including visual style, clothing, scene, composition, lighting, and mood — should follow the reference photo.

Generate an ultra-realistic cinematic vertical portrait. The subject is standing in a {argument name="setting" default="traditional Chinese palace courtyard"}, photographed from a low angle, in a half-body close-up. The person is slightly turned to the side, with the head naturally facing off to one side, not looking at the camera, showing a calm, restrained, story-driven expression. The background features ancient palace eaves, red walls, stone railings, and traditional roof structures, creating a sense of historical Chinese architecture.

The subject wears a {argument name="clothing" default="deep black traditional robe or black Chinese-style outer garment"}, with subtle fabric texture and an overall dark color palette. {argument name="lighting" default="Sunlight comes from the front-side direction"}, illuminating part of the face and creating strong light-and-shadow contrast, with one side of the face lit and the other side naturally falling into shadow. The hair is gently blown by the wind, with loose strands moving naturally, but it must still preserve the hairstyle characteristics of my uploaded portrait and should not be changed into someone else’s hairstyle.

The overall style should feel like a realistic cinematic still, with authentic photographic texture, shallow depth of field, slight film grain, muted colors, high-contrast lighting, and a subtle vintage tone. The skin must look real and natural, with normal skin texture. Do not over-smooth the skin. No plastic look, no anime look. The image should be clean and premium, with no text, no watermark, and no logo. Vertical 9:16, high detail.

Negative Prompt:

watermark, logo, text, Chinese characters, Doubao watermark, AI watermark, anime, cartoon, illustration, plastic skin, overly smooth skin, doll-like face, fake face, face swap look, distorted facial features, wrong identity, different person, bad eyes, unnatural hair, messy hair clumps, overexposed face, blurry face, low quality, extra fingers, deformed hands, modern building, modern clothes, sci-fi style, fantasy armor, excessive beauty filter, unnatural lighting, pasted face, collage look

Create a cinematic Hollywood-style character design board for a big-budget action film featuring a {argument name="protagonist" default="male protagonist"}.

Avoid clean grid layouts; composition should feel like a senior concept artist’s film pitch board used in a studio meeting.

Design a grounded, realistic {argument name="character type" default="male lead actor-type character"} with strong screen presence, subtle imperfections, and emotional depth.

Include:

Full-body turnaround (hero suit / tactical modern outfit)
Close-up facial expression studies (intensity, calm before action, anger)
Wardrobe breakdown (jacket layers, tactical gear, fabric realism)
Signature weapon or gadget design sheet
Motion/action pose studies (running, fighting, aiming)
Height scale with production notes and casting-style annotations

Style: Hollywood blockbuster realism, cinematic lighting, shallow depth of field, premium film still aesthetic, dramatic contrast.

Use only my uploaded portrait as the reference for my facial features, face shape, hairstyle, hair color, natural skin texture, and overall vibe. Do not copy the identity of the person in the reference image. All other visual elements should follow the reference photo: a {argument name="setting" default="gothic castle"} in the background on a rainy day, cool cinematic color grading, a wet and moody atmosphere, the subject sitting beside a wooden terrace table, leaning slightly forward, with one hand naturally touching or lifting her hair, long hair softly blown by the wind, wearing a {argument name="headwear" default="black beret"} and a {argument name="dress" default="light pink sequin spaghetti-strap dress"}, with delicate yet natural makeup. Visible rain, a blurred castle in the background, sharp focus on the subject, shallow depth of field, low-saturation blue-gray tones, and a cold, mysterious, romantic cinematic portrait mood.

Keep my character highly realistic, with accurate facial features and no distortion. Skin should have natural texture, not overly airbrushed, with no plastic look, no anime look, and no influencer-style filter effect. The final image should look like a real high-quality cinematic portrait captured by a camera, with soft natural lighting and consistent environmental light on the face and body. The subject should feel naturally integrated into the scene, not pasted in. Clean and premium composition, with no text, no watermark, and no logo

Negative Prompt:

watermark, logo, text, AI generated mark, anime, cartoon, plastic skin, over-smoothed skin, doll face, fake face, face swap look, distorted face, bad eyes, bad hands, extra fingers, unnatural pose, low quality, blurry face, overexposed skin, unrealistic lighting, messy background, duplicated body, deformed body, cheap filter, heavy retouching

Use my uploaded portrait only for facial identity, facial structure, skin tone, and hairstyle. Create a realistic mirror selfie in a {argument name="location" default="dim indoor hallway or elevator-like space"}, inspired by the reference image. The person is standing in front of a mirror, holding a phone that partially covers the face, with a casual slightly messy pose. Keep the face recognizable and natural, not beautified or over-smoothed.

Style: raw smartphone flash photography, cool-toned low-light atmosphere, slightly grainy mirror reflection, realistic shadows, natural skin texture, candid social-media snapshot feeling. Outfit: {argument name="top" default="oversized black sweatshirt"} slightly lifted at the waist, {argument name="pants" default="loose black-and-white cow-print lounge pants"}, relaxed streetwear styling. The sweatshirt should have no brand logo, only a simple small white abstract graphic or be plain.

Composition: square photo, waist-up to upper-thigh framing, phone in one hand, other hand casually lifting the sweatshirt, messy but realistic hair movement, muted gray wall panels in the background, strong phone-flash look, high realism, natural body proportions, realistic fabric folds, authentic mirror selfie perspective

Negative Prompt：

watermark, logo, text, anime, cartoon, plastic skin, over-smoothed face, fake beauty filter, distorted body, exaggerated waist, unrealistic proportions, bad hands, extra fingers, duplicated phone, blurry face, low quality, glossy doll skin, AI-generated look, oversexualized pose, messy background text

Create a visual trend analysis moodboard for {argument name="trend" default="[trend]"}, including colors, textures, objects, typography inspiration and 5 readable trend keywords

Ultra realistic cinematic portrait of a {argument name="subject" default="young handsome man"} sitting closely beside a {argument name="animal" default="black bull"} at golden hour, emotional bond between human and animal, man gently resting his hand on the bull’s head while feeding green grass, calm affectionate expression, strong love and trust vibe, black bull with textured fur and slightly reddish highlights on forehead, decorated with aesthetic red traditional rope harness and small bells, warm sunset lighting falling dramatically on faces, soft orange glow, shallow depth of field, rustic farm background blurred with bokeh, cozy countryside atmosphere, detailed skin texture, messy dark hair, casual brown plaid shirt over white t-shirt, luxury wrist watch visible, highly detailed eyes, emotional storytelling composition, rich earthy tones, cinematic color grading, ultra sharp focus, photorealistic, DSLR quality, 85mm lens, f/1.8, natural shadows, intimate framing, premium editorial photography style, aesthetic mood, handwritten text in corner saying “Love. Respect. Bond. ❤️”

Use my uploaded portrait only for my facial identity and hairstyle. Create a realistic fantasy portrait in an elegant {argument name="style" default="dark oriental style"}: side-profile composition, long flowing black hair, an {argument name="accessory" default="intricate jeweled gold headdress with blue gemstones"}, delicate floral face art, {argument name="element" default="luminous butterfly"} near the face, translucent shimmering robe, floral decorations on the shoulder and chest, and soft glowing blossoms around me. Black background, dreamy blue-pink lighting, ethereal atmosphere, ornate vintage frame border, cinematic fine-art photography. Keep my face recognizable and realistic, with natural skin texture and a real human look, not anime or plastic. No watermark, no logo, no text

Negative Prompt:
watermark, logo, text, anime, cartoon, plastic skin, doll-like face, over-smoothed skin, fake eyes, distorted face, bad anatomy, low quality, blurry details, oversaturated colors, extra accessories, messy composition