图片提示词

Goal: Create a cinematic anime classroom still in the style of a polished Japanese romantic-comedy school series, with soft daylight, expressive characters, and promotional poster-like Japanese typography.

Canvas: Wide 16:9 landscape frame, 1200×675 composition, eye-level camera, shallow cinematic depth of field, cool blue-gray shadows and warm daylight highlights.

Scene and layout: Inside a modern high school classroom during daytime. Three students sit at desks near large windows. The outside background shows a bright sky, distant buildings, greenery, and concrete school architecture through the windows. The foreground desk runs horizontally across the bottom with school items on it. The mood is bittersweet and comedic, with handwritten commentary floating around the characters.

Characters: Exactly 3 anime students. 1) Left foreground: a cheerful blue-haired schoolgirl, the largest subject, leaning her cheek into one hand with a relaxed smug smile and slightly teary blue eyes; long messy navy-blue hair with bangs, white short-sleeve school shirt, blue bow, small yellow charm or ribbon accessory at her collar. She appears confident and teasing. 2) Center: a brown-haired schoolboy in a white short-sleeve shirt and dark green tie, seated slightly behind her, looking sideways at her with an anxious, awkward expression, small sweat drop on his face. 3) Right background: a short burgundy-haired schoolgirl with hair clips, sitting apart near the window, hunched over and hiding behind a small book or phone, giving a quiet jealous or embarrassed side-eye.

Objects: Include exactly 5 notable foreground objects: a pink lunchbox on the left desk, a folded pink-and-white cloth, a small strawberry yogurt drink carton with a straw in front of the boy, a dark navy school bag on the right desk, and the small book or phone held by the burgundy-haired girl.

Text and graphic elements: Add a small anime title logo in the upper-left corner with the main Japanese title {argument name="title logo text" default="負けヒロインが多すぎる!"}, a tiny romanized subtitle beneath it reading “MAKEINE Too Many Losing Heroines!” Add a classroom poster on the rear wall/window pillar with the Japanese slogan {argument name="poster slogan" default="負けても、きっと、青春は、終わらない。"} and small English subtitle “Too Many Losing Heroines!” Add white handwritten Japanese annotation text near each character with curved arrows and dotted lines, including one near the blue-haired girl saying {argument name="blue-haired girl annotation" default="やっぱ私は負けヒロインなんだよね〜"}, one near the boy saying {argument name="boy annotation" default="また始まった…"}, and one near the burgundy-haired girl saying {argument name="right girl annotation" default="別に気にしてない"}. Keep the handwriting light, casual, and integrated into the scene rather than looking like UI labels.

Visual style: High-quality 2D anime key visual, delicate line art, natural skin shading, detailed hair highlights, soft bloom from windows, subtle film grain, gentle atmospheric haze, realistic classroom perspective, muted cool palette with saturated blue hair as the focal point.

Constraints: Use exactly 3 characters and exactly 5 notable desk/handheld objects listed above. Do not add extra students. Keep the composition recognizable as a single anime promotional still, not a collage. Avoid watermarks, logos from real studios, or unreadable glitch noise.

Create a square, photorealistic fine-art portrait of {argument name="subject" default="a young woman in her early 20s with olive skin and hazel eyes"} facing the camera directly with a calm, neutral expression and relaxed posture. She has {argument name="hair style" default="long dark brown naturally curly hair with soft bangs across her forehead"}, minimal makeup, softly flushed cheeks, natural pink lips, and sharply focused eyes. She wears {argument name="clothing" default="an oversized dark forest-green knitted sweater with a simple crew neck"}. Frame her from mid-torso upward, centered, with her shoulders slightly angled but face symmetrical and looking straight into the lens. Use warm late-afternoon sunlight from the upper left, creating a glowing rim light on her hair and sweater, a soft circular patch of light behind her head, and a distinct but blurred shadow falling on the wall to the right. The background is a plain textured plaster wall in warm brown and charcoal tones, with heavy natural falloff and subtle vignette around the edges. Style should be realistic cinematic portrait photography with shallow depth of field, crisp facial detail, rich but natural skin texture, soft contrast, warm highlights, deep shadows, and no visible text, jewelry, props, or extra people.

Using REFERENCE_0 as the visual style reference, transform the scene from a trading desk setup into a cinematic anime view from inside {argument name="vehicle type" default="a car"} looking out through the windshield and side window at {argument name="city setting" default="a dense Tokyo commercial district"}. Preserve the same widescreen aspect ratio, polished cel-shaded/3D-anime look, dramatic red-and-green color accents, glossy reflections, and high-contrast lighting, but remove the monitors, keyboard, hands, speakers, and trading-chart interface entirely.

New composition: place the viewer in the front passenger area of the car, with the dark roof frame, windshield pillars, dashboard, steering wheel edge, side mirror, and a red seat partially visible in the foreground. Outside, show tall urban buildings packed tightly together under {argument name="time of day" default="bright late-afternoon sunlight"}, with strong shadows and reflective glass surfaces.

Added signage: include exactly 9 prominent billboard/sign elements across the buildings: 1 large central top sign with Japanese text, 1 red dragon/creature billboard, 1 vertical white Japanese sign, 1 wide red-and-white digital screen, 1 green character billboard, 1 blue-and-white Japanese storefront sign, 1 large red-and-white billboard on the upper right, 1 tall left-side cylindrical screen with a red figure, and 1 smaller distant red rectangular sign. Keep the signage mostly graphic and stylized, with plausible Japanese typography rather than perfectly readable text.

Style constraints: make it feel like a clean, upscale AI-video concept frame; sharp perspective, cinematic framing, no people in the foreground, no trading charts, no desktop equipment, no watermark.

Goal: Create a single widescreen educational presentation slide infographic about {argument name="headline text" default="TIMELESS ICONS:"} and the deconstruction of classic cocktails, in a luxurious vintage speakeasy style.

Canvas: 16:9 horizontal slide, dark emerald green background with a subtle vignette, thin double-line gold border, rounded dark frame, ornate Art Deco corner flourishes, and warm amber photographic lighting. Use elegant serif typography in metallic gold and cream, with the main title large and centered at the top and a small spaced subtitle beneath it reading {argument name="subtitle text" default="DECONSTRUCTING CLASSICS"}.

Layout: Divide the slide into exactly 3 vertical cocktail panels with thin gold ornamental borders and small gold title plaques at the top of each panel. Each panel contains one large realistic cocktail photo on the left or center and a vertical ingredient stack on the right, connected by fine dotted gold guide lines. Add a centered italic footer quote along the bottom reading {argument name="footer quote" default="Each cocktail is a masterclass in ingredient interplay."}, flanked by small gold flourishes.

Sections and visible content: Panel 1 is labeled OLD FASHIONED. It shows a large crystal rocks glass filled with amber whiskey, big clear ice cubes, and an orange peel garnish on a dark bar surface. Text paragraph: “Old Fashioned: Whiskey, sugar, bitters, water. A study in elegant simplicity.” Include exactly 4 ingredient mini-images with labels: Whiskey, Sugar, Bitters, Water.

Panel 2 is labeled NEGRONI. It shows a large lowball glass with a deep red cocktail, ice cubes, and a bright orange slice garnish. Text paragraph: “Negroni: Gin, Campari, sweet vermouth. The iconic bitter-sweet aperitif.” Include exactly 3 ingredient mini-images with labels: Gin, Campari, Sweet Vermouth.

Panel 3 is labeled DAIQUIRI. It shows a pale frothy daiquiri in a coupe glass with a lime wheel garnish. Text paragraph: “Daiquiri: Rum, lime juice, simple syrup. A pristine balance of sweet and sour.” Include exactly 3 ingredient mini-images with labels: Rum, Lime Juice, Simple Syrup.

Visual style: Premium cocktail-menu infographic, photorealistic drinks and ingredient glasses, gold engraved ornamentation, dark green velvet-like background, warm highlights, high contrast, crisp readable labels, classic luxury bar atmosphere.

Constraints: Use exactly 3 cocktail panels and exactly 10 ingredient mini-images total. Keep all text readable and correctly spelled. Do not add logos, watermarks, people, extra cocktails, or extra ingredient labels.

Create a cute Q-version sticker-style illustration in PNG format with a transparent background (show it on a gray-and-white checkerboard transparency preview). The subject is exactly 2 Shiba Inu dogs standing side by side, full body, facing forward, happy and friendly: the left dog is a {argument name="left dog color" default="black-and-tan"} Shiba Inu wearing a bright {argument name="left harness color" default="orange-red"} chest harness with small black and yellow straps and a white curved logo mark; the right dog is a {argument name="right dog color" default="cream-white"} Shiba Inu wearing a {argument name="right harness color" default="black"} chest harness with a white curved logo mark. Keep their individual pet features recognizable: upright triangular ears, curled fluffy tails, thick plush fur, round glossy eyes, open smiling mouths, visible pink tongues, small paws, and compact Shiba body proportions. Use a polished digital illustration style with soft realistic fur texture, clean cutout edges, subtle shadows under the feet, bright cheerful expression, and a high-resolution transparent PNG sticker look. Do not add text, background scenery, extra animals, collars beyond the harnesses, or props.

{
  "subject": {
    "person": "A beautiful young woman with long, wavy brunette hair and a warm smile.",
    "expression": "Polished, elegant, and festive.",
    "attire": {
      "dress": "An off-the-shoulder emerald green velvet mini dress featuring a giant, oversized white structural silk bow across the bust.",
      "hosiery": "Sheer black tights.",
      "accessories": "Small, elegant diamond stud earrings."
    },
    "pose": "Standing centered, facing the camera with arms relaxed at her sides."
  },
  "setting": {
    "location": "Grand luxury foyer with a classic winding staircase and ornate wrought-iron railings.",
    "background_elements": [
      "Large, towering Christmas tree heavily decorated with warm golden lights and gold ornaments.",
      "A life-sized traditional Nutcracker soldier standing to the left.",
      "Numerous gold-wrapped gift boxes piled at the base of the tree.",
      "Intricate patterned area rug on a polished marble floor."
    ]
  },
  "composition_and_lighting": {
    "shot_type": "Medium full shot, eye-level angle.",
    "lighting": "Warm, cinematic indoor lighting with a soft bokeh effect on the Christmas tree lights.",
    "color_palette": "Deep emerald green, crisp whites, rich golds, and festive reds.",
    "atmosphere": "Opulent, cozy, and high-end holiday celebration."
  },
  "technical_details": {
    "style": "High-resolution editorial photography.",
    "focus": "Sharp focus on the subject with a shallow depth of field for the background.",
    "image_quality": "8k, photorealistic, professional color grading."
  }
}

Goal: Create a minimalist editorial food poster that demonstrates reducing a realistic food photo into a clean low-detail illustration while preserving the composition.

Canvas: Vertical poster in a 3:4 aspect ratio, warm off-white paper background, generous margins, quiet Japanese magazine/poster design.

Layout: Divide the poster into exactly 2 main visual panels. The top half contains a realistic rectangular food photograph with a thin off-white border around it. The bottom half contains a simplified vector-style illustration of the same dish, centered and slightly smaller, surrounded by sparse typography and fine registration-mark lines.

Top panel: Show a close-up restaurant photograph of a {argument name="dish name" default="Japanese curry plate"} on a dark glossy tabletop. The dish sits on an oval dark brown ceramic plate viewed from a high three-quarter angle. Include exactly 8 visible food components: 1 large pool of brown curry sauce, 1 mound of white rice mostly hidden under toppings, 1 pile of thin fried shredded onions on the left, 1 heap of sliced pork or beef in the center, 1 scattering of pale garlic chips across the top, 1 scoop of white potato salad on the right, 1 small garnish of chopped greens near the back, and 1 lemon wedge at the back right. Add exactly 1 shiny metal spoon entering from the right edge into the curry. Use natural warm restaurant lighting, realistic texture, shallow shadows, and no people.

Bottom panel illustration: Recreate the same plate as a flattened minimal vector illustration using reduced information and reduced color. Keep the exact same 8 food components and 1 spoon, but simplify them into layered organic shapes with thin light outlines and a limited palette of curry brown, tan, cream, pale yellow, muted green, and soft gray. The oval plate is suggested by a fine gray outline and a dark brown curry shape beneath the toppings. Avoid photorealistic detail; make it look like a tasteful poster illustration.

Text content: Add exactly 5 text blocks. Left of the illustration, small spaced uppercase text reads "JAPANESE" above a large elegant serif title reading "CURRY\nPLATE". Upper right, small serif copy reads "Comfort in\nEvery Layer." Bottom left, small text reads "2024" with a thin horizontal line beside it. Bottom right, small spaced uppercase text reads "MINIMAL\nPOSTER SERIES". Use refined black or dark brown typography.

Graphic details: Add exactly 3 fine-line decorative marks around the lower illustration: 1 thin vertical line near the left of the title area, 1 small crosshair mark on the lower right of the illustrated plate, and 1 subtle horizontal line near the year. Maintain a restrained, premium editorial look with lots of negative space.

Style constraints: Minimal colors, low visual noise, no extra captions, no logos, no watermark, no additional utensils, no extra plates, and no people. The final image should feel like a polished Japanese food poster comparing a realistic curry plate photograph with its simplified illustration.

Goal: Create a photorealistic live-action storyboard contact sheet for a corporate lobby fight scene, showing exactly 9 numbered cinematic frames in a 3x3 grid.

Canvas: Wide 16:9 contact-sheet image, thin white gutters separating the panels, each panel numbered in small white digits at the top-left from 1 to 9. Realistic film still look, sharp action photography, natural indoor daylight, 768p storyboard fidelity.

Location: A modern high-rise corporate lobby with polished beige marble floors, tall glass windows and sliding doors on the right and rear, neutral stone wall panels, and a dark marble reception desk. Behind the desk is a company logo and the text {argument name="company name" default="HINODE CORPORATION"}. The receptionist remains behind the desk in multiple shots.

Characters: Use exactly 3 recurring characters. Character 1 is {argument name="female fighter" default="a young East Asian woman in a navy blue business suit, white blouse, red high heels, long brown hair in a ponytail with loose bangs"}; she is athletic, focused, and calm under pressure. Character 2 is {argument name="male attacker" default="a middle-aged East Asian man in a black suit, black shirt, black shoes, messy dark hair, wearing black sunglasses"}; he is aggressive and increasingly defeated. Character 3 is {argument name="receptionist" default="a young East Asian woman in a white blouse behind the reception counter"}; she looks shocked and frightened.

Panel-by-panel layout, exactly 9 panels:
1. Wide establishing shot: the male attacker stands with his back to camera in the open lobby, facing the female fighter several meters away near the windows; the receptionist is visible behind the reception desk. Tense standoff composition.
2. Medium-wide action shot: the male attacker lunges forward and throws a straight punch toward the female fighter; she squares up defensively in her navy suit, fists raised, feet apart.
3. Over-the-shoulder/side action shot: the female fighter pivots and dodges the punch, her ponytail swinging, her face alert; the attacker’s black-suited arm cuts across the foreground.
4. Close-up reaction shot: the receptionist behind the counter clasps her hands nervously near her chest, eyes wide, watching the fight.
5. Medium action shot: the female fighter traps and twists the attacker’s arm from behind or beside him, forcing him to bend forward in pain; her expression is controlled and determined.
6. Tight medium shot: the female fighter strikes or shoves the attacker’s face with an open palm, compressing his sunglasses and cheek; he grimaces while she stares him down.
7. Wide dynamic throw shot: the female fighter executes a judo-style takedown or sweeping throw; the attacker is inverted or falling headfirst toward the marble floor, one leg flung high in the air, while she stays balanced in red heels.
8. Ground-control shot: the attacker lies face-down on the floor, grimacing; the female fighter kneels or crouches over him, pinning his arm behind his back with both hands. Reception desk and receptionist visible in the background.
9. Final victory shot: the attacker lies prone and defeated on the marble floor; the female fighter stands over him with one red high heel planted firmly on his back, looking composed and victorious; receptionist and corporate logo remain visible behind them.

Visual style: Photorealistic Japanese corporate action-drama stills, cinematic framing, realistic motion blur only where appropriate, consistent character identities and wardrobe across all 9 panels, cool neutral office lighting, polished reflections on the floor, dramatic but believable stunt choreography.

Constraints: Do not add extra panels, captions, speech bubbles, weapons, blood, gore, or fantasy effects. Keep the grid clean and readable. Maintain the exact 9 numbered frames and the same lobby location throughout.

Goal: Create a polished editorial SaaS product exploration image for an AI agent collaboration platform called {argument name="product name" default="AgentBoard"}, showing two large web UI screens side by side on a warm off-white background: a marketing landing page on the left and the product dashboard on the right. Use refined startup design, soft shadows, rounded cards, thin gray dividers, lots of whitespace, muted forest-green accents, black serif display typography for the landing headline, and small clean sans-serif UI text.

Canvas: Wide 4:3 composition, approximately 1200×900. Place two tall browser-like panels with rounded corners and subtle drop shadows, separated by a narrow gutter. Background is light cream with faint grain and very soft vignette.

Left screen, marketing landing page: Top navigation bar with the logo text “AgentBoard” at top left, 4 centered nav links labeled Product, Solutions, Resources, Pricing, then Log in and a dark green “Start for free” button on the right. Hero area has a small pill label “AI-NATIVE WORKSPACE”, a large three-line headline reading “One board. Many agents. Real outcomes.” with “Real outcomes.” italic and green, supporting line “Plan, build, and ship work—together.”, and 2 CTA buttons: dark green “Start for free” and light outlined “Book a demo”. To the right of the hero text, show an isometric workflow board with exactly 5 connected agent cards: Research Agent, Strategy Agent, Content Agent, Design Agent, and Data Agent, linked by curved connector lines with small status chips such as Done and In progress. Below, include exactly 4 metric tiles: “2.4M+ Tasks automated”, “180K+ Agents this month”, “98.6% On-time”, and “4.9★ User rating”. Under the metrics, include a row labeled “Connects with the tools you use” with exactly 7 integration icons/buttons: Slack, Notion, GitHub, Figma, Google Drive, HubSpot, and “+ More”. Add a testimonial block with a circular female headshot and the quote “AgentBoard gives us complete visibility and keeps our team aligned.” credited to Sarah Chen, Head of Product, Northline. At the bottom, include exactly 4 feature blurbs with tiny line icons: Visual workspace, Multi-agent collaboration, Real-time progress, and Secure & governed.

Right screen, product dashboard: Create a dense but elegant web app interface. Top app bar shows “AgentBoard” on the left, main tabs Workspace, Agents, Integrations, Data, Reports, plus an Invite button, a row of 5 small user avatars, a “+3” badge, and a dropdown chevron. Left sidebar for workspace “Acme Inc.” with exactly 11 navigation items: Home, Inbox with an 8 badge, Boards, Agents, Workflows, Data, Reports, Integrations, Settings, Help, and Collapse. Main page title reads {argument name="dashboard title" default="Q2 Go-to-Market Launch"} with a green “Live” pill and subtitle “Coordinating agents across research, strategy, content, design, and ops.” Include Share and dark green Add agent buttons.

Dashboard content: Below the title, show exactly 4 tabs: Board, Progress, Outputs, Activity, with Board active. The main board canvas contains exactly 6 connected workflow cards arranged left-to-right with curved connector lines: Research Agent “Market & audience research” Done; Data Agent “Performance analysis” Done; Strategy Agent “Go-to-market plan” In progress; Content Agent “Messaging & content framework” In progress; Design Agent “Creative concepts & brand” In progress; Ops Agent “Launch readiness & checklist” In progress; Analytics Agent “Performance forecast” Queued. Include small board controls at top right: search icon, layout icon, fit icon, zoom “100%”, and dropdown.

Under the board, add a KPI strip with exactly 5 items: Live progress 72% with green progress bar, 9 Active agents, 24 Tasks running, 156 Tasks completed, and 92% On-time, plus a “View all runs” button. Add a Recent outputs section with exactly 4 document cards: GTM Strategy Brief, Messaging Framework, Creative Concepts, and Performance Forecast, each with a small preview thumbnail and update time. Add a Recent runs table with exactly 5 rows: Q2 Go-to-Market Launch, Customer Onboarding Revamp, Pricing Strategy Analysis, Brand Refresh Campaign, and Enterprise Sales Playbook; include tiny agent avatar clusters, colored status pills, started times, and durations. At the bottom, include a Composer input bar with placeholder “What would you like the agents to work on?”, buttons for Attach and Add context, an “All agents” dropdown, and a green send button.

Visual style: Premium minimal product mockup, realistic UI proportions, crisp vector-like rendering, tiny legible interface text where possible, subtle dotted background texture, pale beige cards, green status indicators, fine connector lines, modern editorial SaaS aesthetic suitable for a product-design exploration tweet.

Constraints: Use exactly 2 main screens, exactly 5 agent cards on the landing hero, exactly 4 metric tiles, exactly 7 integration icons, exactly 4 bottom feature blurbs, exactly 11 sidebar items, exactly 4 dashboard tabs, exactly 6 workflow agent cards on the dashboard board plus the queued Analytics card as the seventh endpoint, exactly 5 KPI items, exactly 4 output cards, and exactly 5 table rows. Do not add extra panels, watermarks, browser chrome, or decorative text beyond the described UI.

Create a vertical anime-style portrait illustration of {argument name="character name" default="a quiet androgynous young person"} in a deep indigo night scene, shown from the chest up in a close, intimate crop. The character has {argument name="hair color" default="cool blue-black"} layered bangs covering much of the forehead and one eye, a long thick side braid over the right shoulder tied near the end, pale skin, delicate eyelashes, downcast blue eyes, and a melancholy, serene expression as they lean toward a flower near their lips. Dress them in a dark blue kimono or robe with simple folded collar lines and subtle shadowed fabric folds. Surround the face with exactly 5 luminous blue lotus-like flowers: 1 large flower directly in front of the mouth and nose, 1 large flower at the lower left foreground, 1 partial flower on the far left edge, 1 flower in the upper right corner, and 1 smaller flower on the mid-right background; also include exactly 2 unopened buds near the lower-left stems. Use intense blue monochrome lighting with glowing cyan highlights, dark navy shadows, strong rim light from the upper right, translucent layered petals, golden-yellow stamens, fine hair strands, and crisp anime cel shading blended with painterly gradients. Composition should feel poetic and nocturnal, with the character partially hidden by flowers, high detail in hair and petals, shallow depth of field, no text, no watermark, no border.

Goal: Create a cinematic anime key visual for a romantic school comedy titled {argument name="series title" default="負けヒロインが多すぎる!"}, with a melancholy but funny classroom atmosphere and polished feature-film lighting.

Canvas: Wide 16:9 horizontal frame, 1200×675 style composition, soft depth of field, cool daylight entering from large classroom windows. Use a realistic modern Japanese high school classroom with concrete pillars, window frames, city buildings and pale blue sky outside.

Layout: Place exactly 3 students around a classroom table. Foreground left is the main heroine, largest in frame, leaning on one hand and smiling with a teasing, slightly lovesick expression. Center is a schoolboy sitting behind her, looking uneasy and awkward with a small sweat drop. Right is a second girl, smaller and partly isolated, sitting behind a dark school bag while hiding her mouth behind a pale smartphone, looking jealous or indifferent. Keep the desk edge across the bottom foreground.

Characters: The main heroine has long messy deep blue hair, blue eyes, rosy cheeks, a white short-sleeve school shirt, and a blue ribbon with a small yellow ornament; she rests her cheek on her hand and looks toward the viewer with a smug smile. The boy has short dark brown hair, a white school shirt, green necktie, and a worried sideways glance. The right-side girl has short burgundy bobbed hair with two small yellow hairpins, a white school shirt, and narrowed eyes above her phone.

Visible objects, exactly 5 on the desk: 1 pink lunchbox in front of the blue-haired girl, 1 crumpled pink-and-white cloth or napkin, 1 small pink yogurt drink carton labeled in Japanese with a strawberry graphic, 1 large dark navy school bag in front of the burgundy-haired girl, and 1 pale decorated smartphone held by the burgundy-haired girl.

Text overlays, exactly 6 distinct text elements: 1 top-left title logo reading {argument name="title logo text" default="負けヒロインが多すぎる!"} with small romanized/English subtitle beneath it, 2 handwritten white speech near the blue-haired girl reading {argument name="heroine speech" default="やっぱ 私は負けヒロインなんだよね〜"} with a curved arrow, 3 small white thought text near the boy reading {argument name="boy thought" default="また始まった…"}, 4 vertical wall poster behind the boy with Japanese slogan and the English subtitle “Too Many Losing Heroines!”, 5 handwritten white slogan in the upper-right corner reading {argument name="upper right slogan" default="それでも、好きな気持ちは、きっと、負けじゃない。"}, and 6 small handwritten white text near the burgundy-haired girl meaning she is pretending not to notice, with a dotted line and curved arrow.

Visual style: High-quality Japanese anime film still, delicate line art, natural skin shading, expressive eyes, subtle blush, atmospheric classroom lighting, slight bloom on windows, muted blues and grays contrasted with the heroine’s vivid blue hair and the burgundy-haired girl’s hair. Add poster-design typography and handwritten annotation graphics integrated into the scene.

Constraints: Do not add extra characters, extra desk objects, or extra text elements beyond the six listed. Keep all text legible but naturally hand-drawn where appropriate. No watermark, no photorealism, no 3D render.

Create a vertical 2:3 anime-style portrait illustration of a charismatic young male scholar in a warm, old-world study. The subject is a slim, handsome young man with {argument name="hair color" default="deep auburn red"} wavy medium-length hair, long bangs falling over one eye, sharp {argument name="eye color" default="emerald green"} eyes, fair skin, and a subtle confident smile. He stands in a three-quarter pose, looking directly at the viewer, with one hand raised thoughtfully to his lips and the other arm folded across his torso while holding a closed dark book. Dress him in a crisp cream dress shirt with slightly rolled, wrinkled sleeves, a small tied cravat, and a fitted black waistcoat with buttons, giving him a refined professor or alchemist look. Place him beside a dark chalkboard covered with glowing handwritten cursive formulas, equations, symbols, and sketch marks, but keep the writing decorative and mostly unreadable. Add wooden bookshelves and antique objects on the left side, including exactly 3 visible brass or gold-toned instruments/ornaments and a row of old books near the bottom. Use dramatic golden sunset light streaming from the upper left, strong rim lighting on his hair and sleeves, warm amber reflections, deep shadows, and atmospheric dust motes. Render with polished high-detail anime painting, cinematic composition, rich linework, glossy eyes, soft bloom, warm environmental color grading, and elegant romantic academia mood. Keep the background detailed but secondary, avoid modern items, avoid extra characters, and do not include readable text or watermark.

Create a cinematic science-fiction landscape titled {argument name="title concept" default="The Ring on Earth"}: a colossal vertical ring megastructure standing on the horizon in a vast grassy plain, so enormous it dwarfs mountains and dominates the sky. The ring is a single continuous circular band, slightly tilted in perspective, made of weathered gray metal and concrete-like panels with intricate industrial seams, layered armor plates, faint windows, exposed structural ribs, and one visible circular mechanical docking module on the right side. The bottom edges of the ring disappear into atmospheric haze near the ground, suggesting impossible scale. Foreground: a quiet wet meadow with wild grasses, small flowers, scattered dark rocks, and a shallow reflective pond occupying the lower center, mirroring the sky softly. Midground: rolling green fields, low hills, tiny waterways, and distant mountain silhouettes, with a few snowy peaks far right. Background: pale blue sky with soft clouds, warm low-angle sunlight from the left, mist and volumetric haze around the ring. Style: ultra-detailed photorealistic concept art, epic scale, naturalistic lighting, muted greens and blue-grays, soft golden highlights, serene yet awe-inspiring mood, 4:5 vertical composition, wide-angle lens feel, high dynamic range, subtle film grain. Constraints: no people, no vehicles, no text, no logo, no watermark; keep exactly one giant ring structure and one foreground pond.

[MODEL]
version = photorealistic_v6
quality = ultra
resolution = 8k
aspect_ratio = 3:4
style = cinematic_realism

[REFERENCE_IMAGE]
use_reference = true
face_fidelity = 1.0
body_fidelity = 1.0
instruction = Maintain exact same face, facial features, skin texture, body proportions, and figure as the uploaded image with zero deviation. No beautification, no morphing, no AI face changes.

[SUBJECT]
gender = female
age = same_as_reference
identity = exact_same_person_from_reference
expression = calm, confident, elegant
gaze = looking slightly toward camera
pose = elegant standing pose, relaxed posture

[FACE]
face_shape = unchanged
eyes = unchanged
nose = unchanged
lips = unchanged
skin_texture = realistic, natural
skin_tone = exact_same_as_reference

[HAIR]
style = exact_same_hairstyle_as_reference
length = unchanged
parting = unchanged
texture = natural, realistic

[OUTFIT]
top = light green halter-neck bikini top
top_details = crisscross straps, fitted, elegant
bottom = sheer light green sarong
sarong_details = polka-dot texture, semi-transparent, tied at the hip
fit = flattering, natural drape
no_clothing_change_on_body_shape = true

[BODY]
figure = exact_same_as_reference
proportions = unchanged
no_slimming = true
no_enhancement = true

[ENVIRONMENT]
location = luxurious bright interior
windows = large floor-to-ceiling windows
outside_view = lush tropical foliage, greenery, palm leaves
interior_style = modern luxury, clean, airy
floor = polished or marble
ambience = tropical, serene, high-end

[LIGHTING]
type = natural daylight
source = window light
quality = soft, diffused
shadows = gentle, realistic
skin_highlights = natural

[CAMERA]
shot_type = captivating mid-to-full body shot
angle = eye-level
lens = 50mm DSLR
depth_of_field = shallow
focus = sharp on face and body

[RENDERING_RULES]
photorealistic = true
no_cartoon = true
no_illustration = true
no_anime = true
no_artifacts = true
no_extra_limbs = true
no_distortion = true
no_text = true
no_watermark = true

Goal: Create a nostalgic Kyoto travel memory corkboard collage titled {argument name="headline text" default="Kyoto Memories"}, as if made from real paper souvenirs pinned inside a wooden frame.

Canvas: Vertical 4:5 image, warm dim museum-like lighting, realistic corkboard background with a dark wooden frame, soft shadows, antique beige paper textures, torn edges, masking tape, brass pushpins, and binder clips. Overall mood is cozy, handmade, slightly sepia, and autumnal.

Layout: Arrange exactly 12 main pinned or taped items on the corkboard, overlapping naturally:
1. Top-left small painterly postcard of a glowing Kyoto shrine building at dusk, pinned with one brass tack.
2. Top-center torn handmade title strip reading “Kyoto Memories” in elegant handwritten script, underlined in red, with a small red maple leaf stamp.
3. Top-right bus pass ticket clipped with a small brass binder clip, reading “KYOTO CITY BUS / ONE-DAY PASS,” Japanese text, date “24.11.08,” and “¥700,” with a simple bus icon.
4. Center large landscape painting postcard of {argument name="main landmark" default="Fushimi Inari Taisha endless red torii gates"}, showing a stone path receding through many vermilion gates, forest foliage, lantern glow, and textured pointillist brushwork; secure it with two brass tacks.
5. Left-middle torn note with handwritten English text: “Fushimi Inari Taisha / endless gates, / quiet woods, / fox messengers. / 2024.11.08”.
6. Right-middle vertical torn washi note with Japanese calligraphy poem and a small red seal stamp near the bottom.
7. Lower-left folded Kyoto map with red circled destinations labeled Arashiyama, Kinkakuji, Gion, and Fushimi Inari; partially overlapped by a red maple leaf and a yellow ginkgo leaf.
8. Bottom-left tea house receipt labeled “OMATCHA TEA HOUSE,” with small Japanese text, “MATCHA SET ¥880,” “WAGASHI ¥380,” and total “¥1,260”.
9. Lower-center small painterly postcard of a quiet shrine walkway lined with red lanterns, pinned at the top.
10. Lower-right-center small painterly postcard of a fox statue wearing a red bib at a shrine, pinned at the top.
11. Right-lower shrine entry ticket reading “SHRINE ENTRY,” “ADULT,” “¥500,” date “24.11.08,” and ticket number, pinned near the top with a brass tack.
12. Bottom-right painterly postcard of {argument name="secondary landmark" default="Kinkaku-ji Golden Pavilion"} reflected in water, with dark trees and muted golden highlights.

Additional loose objects: Add exactly 3 autumn leaves: one red maple leaf over the map, one yellow ginkgo leaf near the tea receipt, and one small red maple motif stamped on the lower handwritten note. Add exactly 1 small fabric travel charm tag at the bottom-right corner, tied with a red-and-white cord, with Japanese-style characters suggesting a travel amulet.

Visual style: The postcards should look like thick impasto oil or gouache paintings with visible stippled brush marks, rich but muted autumn colors, rough handmade paper borders, and no glossy modern finish. Use warm amber highlights, deep forest greens, vermilion reds, tan paper, brass pins, and dark walnut wood.

Text content: Keep the headline “Kyoto Memories” prominent and legible. Use the date {argument name="travel date" default="2024.11.08"} where handwritten notes or tickets show a date. Include tasteful small Japanese text on tickets and notes, but do not overcrowd the image.

Constraints: Photorealistic physical collage, not a flat graphic design. Preserve the exact count of 12 pinned/taped paper items and 3 autumn leaves. No people, no modern devices, no watermark, no extra headline, no clean digital UI elements.

Create a cute kawaii sticker illustration of a happy potted cactus on a transparent background, shown as a centered front-facing character with a thick dark green outline and a soft white sticker border. The cactus is bright glossy {argument name="cactus color" default="lime green"}, with one large rounded central barrel body and exactly 2 raised side arms, one on the left and one on the right, both curved upward. Add vertical ribbed shading, shiny highlight spots, and many small star-shaped cream spines with orange bases distributed across the body and arms. Give the cactus a cheerful face with 2 large glossy black oval eyes, 2 pink circular blush cheeks, 2 small curved eyebrows, and a wide open smiling mouth with a red interior and pink tongue. Place exactly 1 large red-pink flower on the upper left of the cactus, with 6 rounded petals and a cluster of small yellow pollen dots in the center. The cactus sits in a warm {argument name="pot color" default="orange terracotta"} flower pot with a thick rounded rim, darker orange shading, speckled texture, and glossy white highlights; show brown soil and exactly 5 visible rounded pebbles along the top. Style should be polished 2D digital art, chibi mascot, vibrant saturated colors, smooth gradients, soft highlights, clean vector-like edges, no text, no watermark, no background scene, transparent PNG look with only a faint checkerboard preview visible outside the sticker silhouette.

Create a polished semi-realistic anime portrait of a {argument name="character description" default="young adult woman with fair skin, light freckles, large blue eyes, and a gentle slightly coy smile"} leaning forward toward the camera in a close three-quarter view. She has {argument name="hair style" default="long chestnut-brown hair with wispy bangs and a loose side braid over one shoulder"}, delicate round wire-frame glasses, soft natural makeup, and luminous skin with detailed highlights. Dress her in a {argument name="outfit" default="traditional alpine dirndl: white puff-sleeve blouse with lace trim, navy bodice with crisscross lacing, and a blue skirt tied with a bow"}; the visible outfit has 4 main pieces: white lace blouse, navy laced bodice, blue skirt, and waist bow. Set the scene beside a clear mountain stream with smooth rocks in the foreground, green forest slopes and misty mountains behind her, and exactly 2 blurred wooden chalet houses on the right side of the background. Use warm morning sunlight, soft rim lighting in the hair, sparkling water reflections, shallow depth of field, high-detail digital painting, romantic alpine atmosphere, 3:4 vertical composition, sharp face and eyes, creamy bokeh background, no text, no watermark.

Goal: Create a vertical Japanese manga page in polished anime style about a schoolgirl oversleeping, rushing to school, then realizing it is Sunday.

Canvas: Portrait comic page, roughly 3:4 aspect ratio, white margins, thick black outer border, divided into exactly 4 horizontal panels stacked from top to bottom with clean black gutters.

Main character: A cute teenage Japanese schoolgirl with short dark navy bob hair, straight bangs, large amber-gold eyes, fair skin, blush marks, and expressive sweat drops. She wears a classic white sailor school uniform with a dark navy collar and a red neckerchief. Keep her design consistent in all 4 panels.

Panel count and layout: Use exactly 4 panels.
1. Top panel: bedroom wake-up scene, angled view from above. The girl lies in bed on a white pillow and blanket, eyes wide and mouth open in panic. A wooden bedside table at left holds a yellow twin-bell alarm clock and books. Add a jagged red panic mark near her head. One large speech bubble on the right with vertical Japanese text: {argument name="wake up text" default="もう朝!?"}.
2. Second panel: outdoor rushing-to-school scene. The girl runs in her sailor uniform with toast in her mouth, cheeks flushed, sweat drops, one fist raised. Background shows blue sky, speed lines, school-like fence/building and green trees. Include exactly 2 large speech bubbles: right bubble text {argument name="right rushing text" default="遅刻しちゃう!"}; left bubble text {argument name="left rushing text" default="遅刻しちゃう!"}.
3. Third panel: school gate scene. The girl stands front-facing in front of a brick school gate and metal fence, holding a pink smartphone, looking surprised and confused with a small open mouth. Background includes leafy green trees and bright blue sky. One speech bubble on the right with vertical Japanese text: {argument name="empty school text" default="あれ…誰もいない?"}.
4. Bottom panel: same school gate composition as panel 3. The girl smiles awkwardly with closed happy eyes, blushing, still holding the pink smartphone, realizing her mistake. One speech bubble on the right with vertical Japanese text: {argument name="sunday realization text" default="今日、日曜日だった…"}.

Visual style: Bright, clean modern anime manga coloring, crisp line art, soft gradients, expressive face close-ups, high saturation but gentle lighting. Use large rounded white speech bubbles with black outlines and bold brush-style vertical Japanese lettering. Maintain a humorous slice-of-life tone.

Constraints: Use exactly 4 panels and exactly 5 speech bubbles total. Do not add extra characters, extra narration boxes, logos, watermarks, or page numbers. Make the Japanese text legible and placed inside the speech bubbles only.

Create a highly detailed vertical anime-style illustration of {argument name="character name" default="an elegant steampunk violinist girl"} performing in a grand dark opera house. She is shown from the thighs up in a dramatic three-quarter perspective, leaning slightly forward while playing a polished wooden violin under her chin; the violin and bow must be anatomically coherent, accurately held, and angled diagonally across the composition. The character has long flowing twin-tail hair in {argument name="hair color" default="pale mint green"}, soft violet eyes, delicate facial features, and a calm, melancholic expression. She wears bronze steampunk goggles with teal lenses on her head, black ribbon hair ornaments, dangling purple gemstone earrings, a black gothic lace dress with sheer bell sleeves, an ornate corset bodice, gold embroidery, chains, tassels, floral metal ornaments, and layered ruffled skirts. Include exactly 3 prominent steampunk accessory types: goggles, jewelry chains, and gemstone ornaments. The background is a luxurious baroque concert hall with curved balconies, carved columns, gilded railings, chandeliers, candlelight, and warm amber bokeh, softly blurred for depth. Use cinematic chiaroscuro lighting, warm gold highlights, rich shadows, intricate textures, glossy violin varnish, fine lace detail, elegant hand poses, dynamic strands of hair, and a refined dark fantasy atmosphere. No text, no watermark, no extra instruments, no distorted hands, no broken violin geometry.

Create a full-body chibi anime character sticker on a transparent-looking black background: a serious young female tactical police officer with an oversized head and small body, standing in a confident contrapposto pose. She has {argument name="hair color" default="short glossy black"} bob-cut hair with heavy bangs covering her left eye, one visible muted gray-violet eye, pale skin, and a neutral slightly stern expression. Add a luminous futuristic halo above her head: a thin glowing blue-white elliptical ring with broken arc segments and one bright star-like cross flare at the front-right of the ring. Outfit: navy police jacket worn open, light blue button-up uniform shirt, black necktie, dark navy tactical pants, black belt with silver buckle, black gloves, and black lace-up combat boots. Include exactly 8 visible police/tactical accessories and insignia: 1 gold chest badge on the shirt, 1 shoulder patch on the jacket sleeve, 1 pale reflective armband on the upper arm, 1 handheld radio clipped near the shoulder with an antenna and coiled cable, 1 belt pouch on the right hip, 1 large thigh holster on the left leg, 1 strapped thigh pouch on the left thigh, and 1 cargo pocket on the right thigh. Use polished high-quality Japanese gacha-game chibi art, clean linework, soft cel shading, glossy hair highlights, subtle fabric folds, compact proportions, and crisp edges suitable for a transparent PNG sticker. Center the character vertically, leave a small margin around the halo and boots, no text, no watermark, no extra characters, no scenery.

Create a square album-cover illustration for a mellow song titled {argument name="song title" default="Matcha-Orange"}. Show a romantic lakeside or seaside terrace at sunset, viewed from a slightly low frontal angle. In the foreground, two people appear as dark silhouettes seated across from each other at a small round café table, suggesting an intimate conversation or quiet date. Count exactly 2 human figures: one on the left leaning forward with a hand near the chin, and one on the right with long hair, seated in profile. Include exactly 1 small round table between them, exactly 2 chairs, and exactly 3 visible table objects: a teacup, a small teapot or kettle, and a shallow plate or dessert dish. The terrace has white wooden floorboards with strong inked perspective lines and long deep shadows stretching toward the viewer, plus railings on the left and right edges. Beyond the railing is a calm glowing body of water with distant dark shoreline silhouettes on both sides. The sky is a dramatic gradient from warm orange near the horizon to deep teal-green at the top, filled with many tiny white stars and several long meteor-like streaks sweeping diagonally from upper left toward the center. Add ornate cream-white line-art clouds on the left and right, drawn with dense parallel contour hatching, and leafy flowering branches entering from the upper right corner. Use a vintage Japanese woodblock print and modern linocut poster style: high-contrast cream linework, dark teal silhouettes, hand-drawn etched texture, subtle paper grain, elegant nostalgic atmosphere, no visible text, no logo, no watermark.

Create a cute kawaii sticker illustration on a transparent background featuring an excited chibi {argument name="animal" default="orange tabby kitten"} reacting dramatically. The kitten is centered, facing forward, with orange striped fur, cream muzzle and belly, pink inner ears, huge glossy amber eyes filled with star highlights, rosy blush cheeks, an open happy shocked mouth, and both front paws raised against its cheeks. Add a thick white sticker outline with a thin dark shadow/edge around the entire character and graphics. Include exactly 1 large comic word at the upper left reading {argument name="comic text" default="OMG!"} in bold rounded yellow-orange bubble letters with black outline and white outer stroke. Add exactly 2 separate yellow exclamation marks to the upper right of the kitten, exactly 2 small yellow sparkle icons, one on the left and one on the right, and exactly 3 short white motion lines near the right cheek. Use a vibrant anime sticker style, clean vector-like line art, warm orange gradients, high contrast, playful expression, no extra objects, no background pattern, no watermark.

{
  "task": "hyper_realistic_vintage_film_portrait_grid",
  "identity": {
    "reference_face": "use_uploaded_reference_images_only",
    "face_accuracy": "100_percent_identical_no_modifications",
    "constraints": [
      "no_face_shape_changes",
      "no_eye_shape_changes",
      "no_nose_or_lip_changes",
      "no_age_modification"
    ]
  },
  "output": {
    "type": "single_image_4_grid",
    "layout": "2x2_grid",
    "quality": "ultra_high_resolution",
    "realism": "hyper_realistic_photography",
    "skin_detail": "fully_preserved_natural_texture",
    "post_processing": "vintage_film_soft_focus"
  },
  "subject": {
    "gender": "female",
    "age_range": "late_20s",
    "skin": {
      "tone": "very_pale",
      "finish": "extremely_dewy_glass_skin",
      "details": [
        "natural_freckles",
        "intense_highlighter_cheekbones",
        "intense_highlighter_nose_bridge"
      ]
    },
    "eyes": {
      "color": "yellowish_green",
      "lashes": "long_curled_individually_separated_doll_like",
      "expression": "dreamy_sultry"
    },
    "eyebrows": "thick_brows_brushed_upwards_with_gel",
    "lips": {
      "color": "peach_light_orange",
      "finish": "highly_glossy_wet"
    },
    "hair": {
      "color": "brown",
      "length": "long",
      "texture": "naturally_wavy",
      "style": "messy_butterfly_cut",
      "state": "wet_strands_falling_on_face",
      "accessory": "single_off_white_orange_tulip_tucked_in_hair"
    }
  },
  "apparel_and_accessories": {
    "top": "light_gray_thin_spaghetti_strap_top",
    "earrings": "large_heart_shaped_imitation_pearl_earrings",
    "nails": {
      "length": "long",
      "design": "intricate_3D_nail_art",
      "details": "reddish_orange_floral_elements"
    }
  },
  "makeup_style": {
    "overall": "very_wet_glossy_fresh_from_water_look",
    "cheeks": "glassy_reflective_blush",
    "eyes": "soft_shimmer_with_definition",
    "lips": "mirror_like_gloss"
  },
  "poses_by_grid": {
    "top_left": {
      "shot_type": "full_face_high_angle",
      "pose": "head_resting_on_surface",
      "expression": "dreamy_sultry",
      "mouth": "slightly_open"
    },
    "top_right": {
      "shot_type": "extreme_close_up",
      "focus": "right_eye",
      "details": [
        "iris_color",
        "eyelash_definition",
        "eye_makeup_texture"
      ]
    },
    "bottom_left": {
      "shot_type": "close_up",
      "focus": "lips_and_fingers",
      "action": "gently_biting_white_straw",
      "details": [
        "gloss_texture",
        "3D_nail_art"
      ]
    },
    "bottom_right": {
      "shot_type": "full_face_high_angle",
      "pose": "head_resting_on_surface",
      "expression": "dreamy_sultry",
      "mouth": "slightly_open"
    }
  },
  "lighting_and_color": {
    "lighting": "warm_golden_dramatic",
    "mood": "intimate_sensual_contemplative",
    "time_feel": "golden_hour_or_warm_indoor_light",
  }

Create a cinematic surreal 16:9 wide image about {argument name="music theme" default="Brahms Piano Concerto No. 1"}: a young {argument name="pianist description" default="romantic-era male pianist with wavy brown hair in a black velvet tuxedo"} performs at the center of a dark glossy circular stage on an impossible black grand piano shaped like a Möbius loop, with the keyboard wrapping around the front curve and the piano body arching into a sculptural oval. Surround the pianist with exactly 24 life-size audience statues arranged in a circular amphitheater ring: 12 matte white neoclassical conservative figures on the left side, dressed in 19th-century formal coats, cravats, bonnets, and top hats, some cracked with glowing yellow fissures; and 12 futuristic progressive figures on the right side, polished chrome and iridescent rainbow like liquid metal, with abstract modern silhouettes and reflective surfaces. A thick circular waveform of explosive paint and light encircles the piano and cuts between both camps, transitioning from hot red and orange on the left through yellow and green to electric cyan and blue on the right, with splashes, sparks, and jagged audio-wave peaks. The composition should feel like conservatives and progressives frozen in judgment around a wild piano performance, viewed from a high three-quarter angle, ultra-detailed, dramatic studio lighting, deep navy-black background, glossy reflections on the floor, high contrast, no text, no watermark.