GPT Image 2 prompts

Subject:
Summer soda crossing the screen

Main:
A photorealistic optical illusion piece where a woman's hand and a glass of {argument name="drink" default="blue soda"} three-dimensionally pop out from a {argument name="device" default="huge black-rimmed smartphone"} placed on a bright wooden table. From the top right, another hand tilts a transparent bottle, pouring a vigorous stream of water into a glass containing lemon. The smartphone screen radiates cracks centered at the base of the glass.

Character/Expression:
A portrait of a woman in her 20s inside the smartphone. Rounded oval face with a small chin, horizontal dark brown eyes, thin loose brows, small round nose tip, and glossy coral pink lips of natural thickness. Facing forward, looking at the camera, and smiling brightly to show white teeth. Thin brown eyeliner and peach cheeks. Dark brown hair tied high, with thin bangs, wisps of hair around the face, and blue-and-white frilled ornaments.

Clothing/Pose:
{argument name="character outfit" default="A puff-sleeved top layered with sheer white and pale blue fabric"}, fine ruffles around the shoulders, and a thin silver necklace. The subject's own right arm extends toward the foreground, firmly grasping a tall transparent glass. The glass contains bright blue carbonated water, ice, and thin lemon slices. Another adult's right hand at the top right holds a transparent bottle with water droplets horizontally, pouring water from the blue mouth.

Background/Light:
A pale wood-grain table, a transparent round wind chime containing blue-purple morning glories in the back left, a transparent saucer, and a strip of paper with a blue firework pattern in the front left. A white cloth with blue embroidery in the bottom right. The background inside the smartphone is a blue sky and white clouds. Strong summer diffused light enters from the window in the top left, creating white reflections on water droplets, ice, glass, and screen cracks.

Composition/Camera:
3:4 aspect ratio. A close-up overhead shot looking down diagonally at the table. The smartphone is placed diagonally from the bottom left to the top right, with the woman's face in the center right, the protruding glass in the center foreground, and the pouring bottle and hand at the top right edge. The lower part of the device is captured near the bottom left edge. Precisely focused on the glass, water stream, and face, with background items lightly blurred by a shallow depth of field.

Texture/Style:
A high-definition surreal advertising visual based on live-action photography. Realistically depicts wet transparent plastic, flowing water, carbonation bubbles, ice, glass, the cracked screen, skin, and wood grain, freshly emphasizing blue and white.

Negative:
Loss of the illusion where the glass and hand penetrate the screen.

A striking {argument name="character" default="female clown cyborg"} with an elegant yet unsettling presence, standing center frame in a {argument name="setting" default="dim futuristic carnival"}. Porcelain-white synthetic skin is cracked in places, revealing polished chrome machinery, exposed micro-cables, hydraulic joints, and faintly glowing circuitry beneath. Her clown makeup is graphic and asymmetrical: smeared crimson lips, sharp black eye shapes, tiny painted tears, and faded blush. One eye is human, the other a luminous mechanical lens. She wears a distressed retro clown costume fused with cybernetic armor, oversized ruffled collar, metallic corset, striped fabric, and mechanical prosthetic limbs. {argument name="lighting" default="Cold cyan neon mixes with dirty red carnival lights"}, reflecting across wet pavement. Cinematic portrait composition, shallow depth of field, eerie fog, gritty textures, beautiful but menacing, retro-futurist body-horror atmosphere, highly realistic materials and dramatic chiaroscuro lighting.

Use the attached image as the primary facial reference. Preserve her recognizable facial identity, natural facial proportions, subtle asymmetry, facial balance, and individual character. Keep the face completely authentic and unchanged—no beautification, reshaping, or feature alteration.\n\nCreate an authentic {argument name="style" default="early-2000s premium CCD compact-camera photograph"} with the unmistakable look of a straight-out-of-camera JPEG. Use a 24–28mm equivalent wide-angle lens and an unconventional low viewpoint, with the invisible camera placed naturally near ground level around her feet and aimed upward. The framing should feel spontaneous and slightly imperfect rather than professionally staged. Position her within the right third of the image, leaving roughly 70% of the frame to an expansive {argument name="background" default="cobalt-blue sky"}. The nearest arm can appear subtly larger due to the wide-angle perspective, while her face stays naturally recognizable and proportionate. The camera must never be visible.\n\nShe is dressed in a {argument name="outfit" default="tailored ivory satin sleeveless crop top"} with structured seams, paired with high-waisted black straight-cut jeans. Include a distressed heart-shaped detail on the rear pocket. She wears oversized glossy black rectangular sunglasses, chunky ivory resin bangles, a few delicate silver rings, and understated jewelry.\n\nHer pose should feel effortless: one hand casually tucked into the back pocket, the opposite arm hanging naturally by her side. Her torso is angled slightly away from the lens as she turns her head over one shoulder toward the camera, giving a calm, self-assured expression. Her long dark wavy hair is loose and moving naturally in the breeze, with fine flyaway strands catching the sunlight.\n\nKeep the setting spacious and uncluttered. A simple concrete rooftop or terrace should appear only along the bottom portion of the frame, while the deep cobalt sky dominates the composition. Add a handful of soft, naturally shaped warm-white clouds in the distance, maintaining plenty of clean negative space.\n\nThe sun is positioned just beyond the upper edge of the image, slightly behind and above her. This creates a bright natural halo along her hair, shoulders, sunglasses, and outer silhouette. Individual loose hairs should glow subtly against the blue sky. Combine direct daylight with gentle on-camera flash-style fill, giving the image the characteristic luminous CCD compact-camera appearance without looking artificially lit.\n\nPreserve realistic skin: naturally fair with a subtle peach undertone, lightly flushed cheeks and nose, visible pores, fine peach fuzz, authentic micro-texture, and a healthy hydrated appearance. Add restrained liquid highlighter reflecting naturally across the cheekbones, bridge of the nose, brow bone, inner eye corners, cupid’s bow, and chin. Makeup remains refined and realistic: champagne-toned shimmer on the eyelids, naturally brushed-up feathered brows, softly separated lashes, and glossy bubble-gum-pink lips.

9 hyper-realistic {argument name="material" default="enamel badges"}, 4x5 layout, minimalist Eastern aesthetics x {argument name="style" default="Song Dynasty style"} x modern premium cultural products; cream, dark teal, celadon, and cinnabar color scheme, large areas of negative space, thin gold borders, flat color blocks, exquisite {argument name="texture" default="enamel texture"}.\n\n1 medium-sized character badge: based on the woman in the reference photo, with double-bun hairstyle, white dress, delicate and natural expression, with a minimalist Eastern landscape background of lake water, arched bridge, distant mountains, and weeping willows.\n\n8 small badges: landscape folding fan, lotus flower, celadon vase, classical window lattice, small boat, cinnabar seal, "Elegant," and "DeepBlue".\n\nOverall restrained, elegant, primitive, and premium, resembling Song Dynasty art exhibition merchandise, avoiding cartoonish, candy-colored, or modern souvenir vibes.

Create a premium cinematic character bible sheet for {argument name="character name" default="DAX"}. Use uploaded character sheet as strict visual reference. Do not change his appearance.\n\nLEFT SIDE: Massive bold brushstroke text {argument name="character name" default="DAX"} top left, {argument name="accent color" default="deep blue"}. Small text: THE RIDER / NUMBER 7.

Create a cinematic vertical 3D cartoon scene on a realistic forest floor, framed like a mobile video still. In the center foreground, show one cute miniature bearded man, {argument name="character name" default="Montaser"}, with an oversized caricature head and worried expression, standing stiffly with arms slightly out while trapped in a large white spider web behind him. He has short dark brown hair, a trimmed beard, raised sad eyebrows, and big anxious eyes. Dress him in exactly four main outfit pieces: a dark denim jacket, a gray T-shirt, black pants, and white sneakers. To his right, show one giant glossy black cartoon ant with large expressive eyes, two antennae, six thin legs, and bent forearms, leaning toward him with a curious intimidating look. Behind them, the spider web radiates clearly across green leaves and stems. At the top center, add a game-style scoreboard overlay: a dark rounded rectangle with exactly two circular avatar icons, the left icon showing the man on a blue background, the right icon showing the ant on an orange background, and large white score text {argument name="score text" default="2 - 1"} between them. The environment is a macro-scale forest floor with dirt, small pebbles, tiny plants, blurred green foliage, warm daylight, shallow depth of field, and cinematic bokeh. Use polished Pixar-like 3D rendering, soft realistic lighting, detailed textures, playful comedy tension, and a vertical 9:16 composition. Add a small white stylized logo watermark at the bottom right reading {argument name="watermark text" default="MONTASER"}.

Create an ultra-realistic wide-angle documentary landscape photograph of a quiet Mediterranean fishing village inlet at low tide under a clear deep blue sky. The scene is shot from ground level on a muddy, sandy tidal flat with shallow reflective water channels winding toward the sea in the center distance. On the right foreground, place one prominent small wooden fishing boat beached on the sand, painted {argument name="boat color" default="bright blue with a red-orange lower hull and yellow trim"}, casting a dark shadow. In the midground, include exactly three additional small boats floating or resting in the shallow harbor: one red-and-blue boat near the left quay, one small dark boat near the center-right water, and one blue-and-white boat farther right. Low-rise coastal buildings line both sides of the inlet: older tan and beige shopfronts and apartments on the left quay, white and cream apartment blocks on the right seawall, with sparse streetlights, railings, parked cars, and a stone embankment. The horizon opens to a calm blue sea beyond a low breakwater. Lighting should be harsh bright midday sun with crisp shadows, natural colors, candid travel-photography realism, slight lens softness and minor imperfections, 16:9 aspect ratio, no people visible, no text overlay, no watermark.

Create a cinematic modern Japanese theatrical anime-style wide landscape scene of two young girls standing barefoot in a shallow flooded concrete drainage canal at sunset. The girls are centered and full-body, wearing simple off-white long-sleeved institutional gowns with visible number tags: the left girl has {argument name="left girl label" default="5"} on her chest and the right girl has {argument name="right girl label" default="4.8"} at her waist. The left girl has {argument name="left girl hair" default="dark navy-black hair in a long side braid"}, a quiet serious expression, and looks slightly toward the other girl. The right girl has {argument name="right girl hair" default="short fluffy white hair"}, pale eyes, a delicate uncertain expression, and holds a small beige rabbit plush toy against her chest. They stand ankle-deep in mirrorlike water that reflects them, but the reflection transitions into a deep starry night sky filled with tiny blue-white stars, purple nebula-like light, and a bright crescent moon, creating a magical realism effect. Surround them with tall sloped concrete canal walls on both sides, weathered industrial details, ladders, railings, yellow-and-black hazard stripes, weeds and grasses growing from cracks, puddles and wet concrete, distant buildings, utility poles, wires, and an overpass bridge in the background. The sky is dramatic with towering sunlit clouds, blue upper sky, warm golden sunlight breaking through from the right horizon, soft rim light around the figures, and sparkling highlights on the water. Use a low eye-level camera, strong depth, ultra-detailed painted backgrounds, atmospheric perspective, cinematic lighting, melancholic but wondrous mood, high-quality anime film composition, 16:9 aspect ratio. Avoid text other than the two numeric tags, avoid extra characters, avoid watermark or logo.

Using the provided reference image as the source meal photo, regenerate it as a more natural, realistic restaurant tabletop shot taken from a slightly lower, landscape angle. Keep the same casual family-restaurant feel, wooden table, floral-rim plates, and the same meal components, but rearrange and simplify them into exactly 4 visible food/drink items: 1) a large plate of spaghetti on the right with a soft poached egg centered on top and herb flakes, 2) a white bowl of tomato-based soup on a saucer at the back left with visible beans/vegetables, 3) a white plate of sautéed spinach with bacon and garlic in the front left, and 4) a tall dark iced drink in a textured glass behind the pasta. Remove the black sauce pot, wooden serving boards, mug, and the brown rectangular spinach dish. Make the pasta strands look coherent and appetizing rather than fractal or tangled unnaturally. Use warm indoor lighting, realistic phone-camera perspective, natural shadows, and no added text or people.

Create a soft pastel anime close-up portrait of a cute chibi-style girl with {argument name="hair color" default="bubblegum pink"} hair in fluffy twin tails, large sleepy blue eyes, a tiny pouting mouth, and heavy rosy blush across both cheeks. She rests her chin on one hand with a shy, dreamy expression, wearing a loose white long-sleeve top with a frilled cuff and a visible shoulder cutout. Add exactly 1 pale yellow star hair clip on the right side of her hair, exactly 1 small white kitten with pink ears sleeping on top of her head, and exactly 2 glowing yellow sparkle stars floating to the left of her head. Use a warm indoor background with soft brown and cream tones, shallow depth of field, gentle bloom, glossy highlights on the hair and eyes, delicate line art, rounded proportions, kawaii Japanese MV-thumbnail aesthetic, square 1:1 composition, cropped close around the face and upper shoulders. Keep the mood sweet, sleepy, innocent, and pastel; no hard shadows, no realistic rendering, no extra characters, no readable text or watermark.

Create a low-resolution, surveillance-camera-style image on a mostly black canvas, with a small vertical rectangular photo crop centered in the frame. Inside the crop, show {argument name="subject" default="a woman"} standing on gray concrete pavement, facing slightly toward the camera with a neutral expression, shoulder-length dark hair, and wearing {argument name="outfit" default="a long dark brown or black dress/cardigan with light shoes"}. She is holding and pushing {argument name="object" default="a simple silver metal hand cart or dolly with a curved handle and two horizontal bars"} in front of her. The crop should look grainy, compressed, and slightly blurry, like a distant CCTV or old digital zoom capture, with muted colors, visible noise, and no clean studio lighting. Use a plain gray outdoor ground background with no signage or extra objects. The final image should have exactly 1 person and exactly 1 cart, with the photo crop surrounded by a large solid black border.

Create a realistic high-resolution food photography image of {argument name="dish name" default="unaju, Japanese grilled eel over rice"} served in a rectangular black lacquered bento box with a thin red inner rim, placed on a warm dark wooden restaurant table. The box is filled with white rice and topped edge-to-edge with exactly 4 long grilled eel fillets arranged horizontally in parallel, slightly overlapping, coated in a glossy caramel-brown tare sauce with visible char marks, tiny blistered skin texture, subtle grill lines, and faint skewer marks. The eel should look moist, oily, and sizzling, with bright specular highlights and appetizing lacquered reflections. Use a close-up three-quarter overhead angle, shallow depth of field, natural warm restaurant lighting, and rich contrast. In the softly blurred background include exactly 3 supporting table items: a small white side dish at the upper left with pale pickles or garnish, a dark cylindrical cup near the upper center, and a black soup bowl at the upper right containing clear broth and a small green herb leaf. Keep the focus sharply on the eel and lacquer box, with no people, no utensils, no text, and no watermark.

Create a vertical cinematic Korean webtoon-style fantasy illustration of two children standing ankle-deep in reflective water inside a dark abandoned underground station. The scene is dominated by one enormous translucent cosmic jellyfish floating overhead, its bell glowing with stars, nebulae, cyan, violet, and deep blue bioluminescence, with many long ribbon-like tendrils trailing down through the air. Below it, place exactly 2 children in the foreground: one dark-haired child on the left with a long braid, pale robe, a chest label reading “5”, and a small flashlight casting a warm beam into the water; one light-blond child on the right with short wavy hair, pale robe, a chest label reading “4.8”, holding a small plush bunny and raising one hand to present a glowing blue orb. The environment is a flooded, decaying subway concourse with escalators on the left, rows of pillars receding into a bright circular vanishing point, cracked walls, hanging droplets, glowing jellyfish-like bulbs scattered on the water surface, and illuminated poster panels showing abstract blue-purple light. Use dramatic perspective, glossy reflections, volumetric mist, tiny floating particles, high-detail anime linework, luminous painterly rendering, deep shadows, magical blue and purple rim lighting, and a melancholic dreamlike atmosphere. No extra characters, no modern signage beyond the two numeric chest labels, no watermark, no border.

Create a wide 16:9 polished anime illustration of three high-school girls walking together from left to right on a sunny residential Japanese street. Show exactly 3 characters in side profile or three-quarter profile: left girl with very long dark teal-green hair, yellow-green eyes, a green bead headband, green-accent sailor uniform, green pleated skirt, black thigh-high socks, and a navy shoulder school bag; center girl with very long pale sky-blue hair, red eyes, blue floral hair ornament with dangling beads, blue-accent sailor uniform, blue pleated skirt, black thigh-high socks, and a navy shoulder school bag with small blue keychain charms; right girl with very long pink hair, red eyes, pink floral hair ornament with dangling beads, smiling and looking back toward the others, pink-accent sailor uniform, pink pleated skirt, black thigh-high socks, and a navy shoulder school bag with small pink charms. Use the character inspiration {argument name="character trio" default="Kotonoha Akane, Kotonoha Aoi, and Tohoku Zunko"}, but render them as original clean anime schoolgirls. The uniforms should have white short-sleeve sailor tops with colored collar stripes matching each girl's skirt, red ribbon ties at the chest, fitted waists, and soft fabric folds. The scene is bright midday with warm sunlight, leafy green trees and bushes casting dappled shadows, a low stone wall, traditional wooden house facade, sliding-style wooden gate, and dark tiled roof in the background. Composition should be cinematic and detailed, with the girls filling most of the frame, hair flowing naturally behind them as they walk, crisp line art, vibrant colors, soft shading, glossy hair highlights, charming slice-of-life mood, no text, no watermark, no extra characters.

Create a vertical retro manga-style science-fiction illustration in a delicate hand-inked line art look with muted screen-print colors. Show {argument name="character count" default="two"} young anime girls riding in a small boxy open-top orange utility car through a dusty orange desert. The girl driving on the left has long dark navy-black hair in a side braid, serious eyes, and a pale work-robe outfit with a small chest patch reading “5”; the passenger on the right has short messy white-blond hair tied low, looks off to the right, wears a matching pale robe with a patch reading “4.8,” and holds one plush beige rabbit with floppy ears. The car has two round headlights, a red front grille, red wheel rims, simple angular panels, and a curling dust trail behind it on a winding pale road. The landscape is an alien radio-telescope desert: include exactly 14 visible satellite dish elements total, counted as 1 huge vine-covered dish dominating the left foreground, 1 partial dish cut off at the far left edge, 1 medium upright dish in the middle distance, 1 small dish just behind it, 3 tiny horizon dishes to the right of center, 1 large upright dish at the right background, 1 tiny far-right horizon dish, and 5 damaged or tilted dish fragments scattered in the foreground and midground. Above the turquoise sky, place an inverted white futuristic city skyline hanging upside down from the top edge, with dense rectilinear towers and antenna-like spires. Add exactly 2 pale cream suns/moons in the sky. Use a warm peach-orange ground, teal-blue sky, cream highlights, fine blue-gray ink outlines, subtle paper grain, sparse rocks and scrub plants, a dreamy quiet atmosphere, and a composition reminiscent of a 1980s/1990s Japanese sci-fi manga cover. No extra text besides the two chest labels “5” and “4.8,” no watermark, no photorealism.

Create a side-by-side comparison graphic on a white background showing two AI image results from the same prompt. Canvas: landscape aspect ratio, approximately 580×420 px, clean white margins. Layout: two equal vertical image panels aligned horizontally with a narrow white gutter between them; each panel has a black sans-serif heading centered above it. Use exactly 2 headings: left heading “ChatGPT Image 2”, right heading “Grok Imagine 2.0”. Left panel: a highly realistic warm-toned portrait photo of an elderly East Asian woodcarver in a cluttered traditional workshop, gray hair tied back, long white beard and mustache, weathered face with fine wrinkles, wearing dark work clothes, holding a detailed carved wooden statue close to his chest with both hands, soft window light from the left, shelves and tools blurred in the background, dusty particles in the air, documentary realism. Right panel: a more cinematic but less natural portrait of an elderly East Asian craftsman seated at a workbench in a woodworking studio, balding gray hair, deeply wrinkled face, dark shirt, carving a small wooden figure with hand tools, strong dramatic sunlight from a window, many chisels and woodworking tools standing upright behind him, wood shavings scattered across the table, high contrast shadows, slightly artificial glossy look. Include exactly 1 small blue circular overlay icon near the center-left of the right panel, with a simple white symbol inside, resembling an interface marker. Overall style: realistic photographic comparison, left image more natural and detailed, right image more stylized and dramatic. No extra captions, no watermarks beyond the single blue overlay icon, no borders except the white page background.

Create a vertical anime-style cinematic illustration of {argument name="character name" default="a young female military cadet"} kneeling on a dark wooden floor inside a cramped old room at night, viewed from a high overhead angle looking down at her face and body. She has short layered {argument name="hair color" default="deep teal hair with lime-green streaks"}, bright green eyes, pale skin, and a quiet serious expression as she looks slightly upward and to the side. She wears a fitted dark green military-inspired uniform dress with gold piping, brass buttons, red-trimmed shoulder epaulettes, a high collar with a small insignia, black gloves, and structured sleeves with red cuffs. Her right hand holds one bright flashlight aimed downward, casting a warm circular glow across the floor and papers; include a second flashlight partially visible at the lower left edge as if held by someone just outside the frame. Scatter exactly 4 visible sheets of music notation on the wooden floor: 2 sheets on the left side near the wall, 1 sheet illuminated near the bottom center, and 1 partially cropped sheet at the lower left. The room should have vertical wooden wall panels, floorboards with strong perspective, a heavy red curtain on the left, and a small window in the upper right showing a dark blue starry night. Use dramatic chiaroscuro lighting, warm flashlight glow against cool moonlight rim lighting, glossy highlights on the hair and uniform, detailed linework, polished modern anime rendering, intimate mystery atmosphere, no text, no watermark, no extra characters visible except the cropped second flashlight hand/object at the edge.

Goal: Create a clean fashion-photography character sheet for a peach jelly commercial heroine, centered on a young Japanese woman named {argument name="character name" default="Momoka"} holding a peach, showing how a close-up concept model expands into full-body and expression references.

Canvas: Square 1:1 white editorial layout, high resolution, soft peach accents, thin pale grid lines, airy commercial mood.

Top section: Three large panels across the top, connected by exactly 2 thick hand-drawn orange arrows pointing left to right. Panel 1 is a large warm golden-hour outdoor half-body portrait of the woman in a white off-shoulder gauzy dress, windblown long dark wavy hair, green mountains blurred behind her, holding a peach near her chest. Panel 2 is a studio half-body front view on a plain cream-white background, the same woman holding one peach with green leaves at stomach height. Panel 3 is a studio full-body front view, the same pose and outfit, showing white sandals and the full translucent dress.

Main lower layout: A character reference sheet divided into a large left full-body hero portrait, a central grid of reference images, and a narrow right information column. Keep all panels aligned in a precise catalog-style layout with delicate peach-pink labels and Japanese/English bilingual headings.

Subject details: Young Japanese woman, early 20s, slim and graceful, natural fresh makeup, soft rosy cheeks, calm gentle expression, long wavy dark brown hair, fair skin. Outfit is a flowing semi-sheer white long dress with loose sleeves, deep soft neckline, layered skirt, and flat white sandals. She holds a realistic peach with leaves as the signature prop. Overall brand feeling is fresh, innocent, natural, summery, and slightly nostalgic.

Central reference grid, count every item exactly:
- Full body row labeled 「全身アングル / FULL BODY」 with exactly 5 full-body views: front holding peach, left side profile, three-quarter/front angled view, right side profile, and back view.
- Face angles row labeled 「顔アングル / FACE ANGLES」 with exactly 4 bust portraits: front face, three-quarter left, three-quarter right, and right side profile.
- Expressions row labeled 「表情 / EXPRESSIONS」 with exactly 6 bust portraits: neutral, soft smile, bright smile, thoughtful finger near lips, slightly serious/concerned, and eyes closed calm expression.

Right column content: A profile card labeled 「プロフィール / PROFILE」 listing compact Japanese-style stats for the character: name {argument name="displayed name" default="桃香（ももか）"}, age {argument name="age" default="28歳"}, gender female, nationality Japan, height 160cm, weight 48kg, body type slim, occupation model, personality gentle, natural, and soothing. Below it, a palette card labeled 「カラー / PALETTE」 with exactly 6 circular color swatches and labels: dark brown hair, pale peach skin, pure white dress, light pink, peach orange, and leaf green. At the bottom, a usage card labeled 「使用イメージ / USAGE」 with the text “MOMOCHI FRUITS” and “桃ゼリーCM 主人公”.

Visual style: Photorealistic DSLR fashion editorial, soft diffused studio lighting for the sheet, warm backlit sunlight in the first top image, natural skin texture, consistent face identity across all views, creamy whites and peach tones, elegant minimal Japanese design. Avoid cartoon styling, avoid heavy makeup, avoid extra characters, avoid clutter, avoid incorrect item counts, and keep the woman recognizably the same in every panel.

Create a soft, detailed anime-style illustration of a young girl sitting on a wooden park bench beside a calm pond in a lush Japanese garden. The girl is centered slightly left, with {argument name="hair color" default="warm brown"} hair in twin low pigtails tied with white ribbon bows, large gentle brown eyes, a small smile, and a peaceful, shy posture with both hands folded on her lap. She wears a light blue sleeveless dress with tiny white floral patterns, a white cardigan, tan strappy sandals, and a small pale yellow crossbody satchel. The setting is a bright spring or early summer park with dappled sunlight, dense green trees, flowering plants, grasses, lily pads, and a reflective pond behind her. Include exactly one wooden arched bridge in the mid-background crossing the water, exactly one small pavilion or gazebo on the far right bank, exactly one large tree trunk on the far left foreground, and exactly one small wooden sign at the lower left with simple Japanese characters and a pink flower symbol. Use a square canvas, warm daylight, soft watercolor-like rendering, delicate linework, rich foliage detail, sparkling reflections on the water, and a wholesome nostalgic mood. Avoid modern city elements, extra characters, logos, watermarks, or random added text beyond the small sign.

Create a dynamic comic-book/anime action illustration of {argument name="character name" default="a bearded photographer scientist"} lunging toward the viewer in a dramatic three-quarter pose on a warm off-white background. He wears a backwards black baseball cap with large white text reading {argument name="cap text" default="SEATTLE"}, brown rectangular glasses, black ear gauges, a trimmed brown beard and mustache, a white lab coat over a dark charcoal pinstripe vest, white shirt, and black tie, with a small rainbow pride pin on his chest. His left hand holds a black Canon EOS DSLR camera with a large lens pointed toward the viewer, while vivid electric-blue lightning arcs from the lens and across the left side of the composition. His right hand holds a black rectangular camera flash angled upward, and his open center hand presents a swirling dark purple-black portal or vortex surrounded by cracked energy and pale blue-white lightning. Show exactly three main handheld objects: the Canon EOS camera, the separate camera flash, and the glowing vortex hovering above his open palm. Add floating broken stone fragments around him, with large shards near the top left and right edges and smaller debris scattered throughout, implying explosive supernatural force. Use bold manga ink outlines, sharp hatching, cel-shaded colors, exaggerated perspective, intense facial expression, detailed hands, energetic speed lines, and high-contrast highlights; keep the image as a single action panel with no frame border, no extra characters, and no additional text besides the cap and camera branding.

Create a bright, dreamy anime-style full-body illustration of {argument name="character name" default="a young summer girl"} standing in a sunlit white garden pavilion. She has {argument name="hair color" default="short black hair with white highlights"}, large clear blue eyes, fair skin, and a gentle shy morning expression. She wears a pale sky-blue off-shoulder sundress with thin straps, translucent puff sleeves, layered sheer ruffled hems, and light blue trim, plus white platform sandals. Add a cream woven straw sunhat decorated with a small pale-blue flower on the side; her right hand lightly holds the brim of the hat while her left hand delicately lifts the edge of the skirt. The pose is graceful and slightly tilted, one leg relaxed forward, creating a soft breezy summer feeling. Set the scene in an elegant open-air gazebo with tall white columns, ornate arched glass-and-metal roof details, white marble floor, distant greenery, blue sky, and fluffy clouds. Use strong morning backlight from behind her head, glowing sun rays, soft lens flares, overexposed highlights, pastel colors, airy transparency in the fabric, delicate line art, refined anime rendering, and a peaceful vacation-morning atmosphere. No text, no watermark, no extra characters.

Create a high-quality professional overhead product photograph of a whole pepperoni pizza centered on a clean bright white seamless studio background. The pizza should be perfectly round and fill most of the square frame, photographed from directly above with soft studio lighting and a subtle natural shadow around the crust. Show a thin-to-medium golden-brown crust with darker toasted bubbles around the rim, rich red tomato sauce visible near the edges, glossy melted mozzarella cheese with lightly browned spots, fine dried herb seasoning scattered across the cheese, and exactly 21 round pepperoni slices arranged evenly across the surface in a balanced pattern from edge to center. Make the pepperoni deep red-orange, shiny with rendered oil, slightly curled at the edges, and speckled with seasoning. Keep the composition appetizing, realistic, crisp, and commercial-food-photography quality, with no plate, no text, no hands, no props, and no background clutter.

Create an ultra-photorealistic cinematic miniature advertising scene for {argument name="brand name" default="Spotify"}: a glossy black vintage convertible toy car driving on a rain-soaked miniature city street at night, with neon green branding accents. The car has exactly 3 visible brand marks: one large green circular Spotify logo on the front grille, one green Spotify icon plus the word “Spotify” on the left door, and one small green Spotify icon on the visible wheel hub. Show exactly 2 glowing round headlights, 2 visible wheels with green tire rims, chrome mirrors and bumper details, and a black leather interior. A small realistic male driver sits in the car, with short dark hair, glasses, and a trimmed beard, wearing a dark jacket and holding the steering wheel while looking forward. In the background, place one enormous out-of-focus circular green Spotify logo towering like a billboard, plus soft bokeh city lights and miniature greenery. The road should be wet and highly reflective, with green and warm gold reflections on the asphalt, shallow depth of field, macro lens perspective, dramatic low-angle composition, cinematic lighting, realistic reflections, tilt-shift miniature scale, premium commercial photography look, no extra text besides the brand word on the car.

Goal: Create a dense editorial cultural-exhibition poster about {argument name="exhibition title" default="Dunhuang Rhythms"}, combining Dunhuang mural fragments, music-and-dance archive typography, and layered museum ephemera.

Canvas: Tall vertical poster, approximately 2:3 ratio, aged warm beige rice-paper background with visible grain, stains, torn edges, faded ink, and pasted-collage seams. Overall mood is scholarly, archaeological, and highly designed, with intentional clutter but a strong grid.

Layout: Use an asymmetric high-density collage. On the far left, place four enormous stacked black Chinese calligraphy characters reading 「壁上乐舞」, occupying nearly the full height, distressed and ink-textured. Beside the large characters, add a narrow vertical Chinese subtitle reading 「敦煌壁画中的乐舞与文明」. At the top center-left, add a small bilingual series label: 「丝路映像 系列展」 with “Silk Road Imaging Series” underneath. At the top right, place the main English title in elegant serif capitals: “DUNHUANG RHYTHMS”, with the smaller subtitle “Fragments of Murals and Music” below.

Main imagery: Use exactly 6 visible mural/image fragments, all in faded Dunhuang cave-painting colors of turquoise, cinnabar red, ochre, black, and parchment beige: 1) a large upper-right flying apsara musician fragment with flowing ribbons crossing diagonally; 2) a lower-left torn fragment showing three mural musicians or dancers; 3) a mid-right small rectangular fragment of a standing dancer; 4) a mid-lower ornate floral scroll pattern fragment; 5) a large bottom-right close-up of a female bodhisattva or apsara face with jewelry and flowing ribbons; 6) a right-edge partial vertical mural strip cropped off the page. Make the fragments look like scanned ancient wall paintings with cracks, missing plaster, chipped edges, and uneven opacity.

Information blocks: Add many thin vertical Chinese text columns like museum notes and archival captions around the collage, especially along the far left margin, middle columns, and lower center. Include exactly 4 primary info panels: 1) a right-side date and venue panel with “2026 09.05 SAT / 2027 01.16 SUN”, Chinese venue text, and “Hexi Culture Center Shanghai”; 2) a lower-left educational panel listing “Instruments” and “Dance Poses” with a small line drawing of an instrument; 3) a bottom-left organizer/credits panel with Host, Co-operation, and Visual Design lines; 4) a lower-right cave caption panel reading “Cave 285, South Wall Music and Dance Scene (detail)” with Chinese above it.

Typography and style: Mix oversized rough black Chinese brush type, refined Western serif uppercase, small Chinese Song/Ming-style vertical text, and tiny English museum captions. Use black, dark brown, muted red stamp ink, and occasional faded grid marks. Add small red seal stamps, thin archival diagrams, faint measurement lines, and marginal annotations. The design should feel like a contemporary museum poster, editorial magazine spread, and historical archive board fused together.

Constraints: Preserve a clear exhibition-poster hierarchy despite the dense collage. Do not make it modern glossy; keep everything matte, aged, weathered, and printed on textured paper. Use no photorealistic people outside the mural fragments, no 3D effects, no bright neon colors, and no empty clean white background. The most prominent readable texts should be 「壁上乐舞」, {argument name="Chinese subtitle" default="敦煌壁画中的乐舞与文明"}, “DUNHUANG RHYTHMS”, and the date range {argument name="exhibition dates" default="2026 09.05 / 2027 01.16"}.