GPT Image 2.5 API awesome prompts

Create a professional high-resolution paint-by-numbers template from the provided image as a single PNG only. Use an {argument name="canvas size" default="A4 portrait ratio canvas"} at 300 DPI with a pure white background. The main artwork must be a clean black-and-white line drawing containing only black outlines, black numbers, and white unfilled paintable areas. Do not create a PDF, document, mockup, multi-page layout, or colored version. Do not use color, gray, shading, texture, gradients, transparency, or pre-filled areas in the main artwork. Analyze the original image and choose the number of colors adaptively based on image complexity, from 1 to a maximum of {argument name="color count" default="14 colors"}. If the image has only 1–2 important colors, use only 1–2 colors; for simple images use 3–6 colors; for medium complexity use 7–10 colors; for complex, photographic, or highly detailed images use 11–14 colors. Do not force unnecessary colors. Assign each selected color one unique number and merge similar shades unless separation is essential for recognition. Convert the image into simplified, fully closed, easy-to-paint regions while preserving the recognizable composition, main objects, characters, text, logos, foreground, background, and key details. Every paintable white region must contain the correct black number, centered clearly inside the matching area, readable, upright, and not touching outlines. Large black or dark areas from the original must remain white paintable regions with the correct number, not filled black. Only tiny permanent black line details, such as mouth lines, nostrils, pupils, eyelashes, seams, or small decorative strokes, may remain unnumbered. At the bottom of the same PNG, add a neat legend section with the heading “Instructions”, the sentence “Paint the areas with the matching colors.”, and a list of every used number with a small color swatch and the English color name. Color swatches may appear only in the legend. Ensure the numbering exactly matches the legend, every paintable area is numbered, all regions are closed, and the final result is clean, professional, print-ready, and recognizable as the original image.

Using the provided reference image, redesign it as a polished retro Japanese music-release poster while preserving the same split layout, the same illustrated band artwork on the right, and the same overall composition. Upgrade the typography from plain placeholder text into authentic vintage poster styling: add distressed print texture, uneven ink, and a slightly worn screen-printed look. Keep the 3 vertical text columns on the left, but restyle them with stronger hierarchy and color separation: the Japanese service-announcement column in warm gold, the platform list in muted teal, and the date/release column in dusty pink. Change the bottom of the left Japanese column from the original release-start message to 「配信中♪」 and emphasize it in larger pink type with a musical note. Add thin gold decorative border lines and corner ornaments framing the left text block. On the right panel, keep the title text in the moon area but restyle it into bold distressed retro lettering that reads “MY MIDNIGHT KIDS’ COMPLEX.” Apply a cohesive nostalgic poster finish across the whole image: subtle paper grain, faded edges, light scratches, slightly darkened borders, and richer vintage teal/gold/pink color grading.

A professional hand drawn sketch storyboard sheet, pencil and ink illustration style, rough artistic linework, cross-hatching for shadows, loose expressive strokes, monochrome black and white on aged cream/off-white paper texture background, 2x2 grid layout with four equal storyboard panels bordered by thick hand-drawn black ink frames. Top-left panel sketch: {argument name="first panel" default="abandoned rusted cargo ship at dock, heavy hatching for rust texture, broken railings, algae, seagulls, no humans, dark moody pencil shading"}, handwritten label below: "SCENE 01 — ABANDONED STATE | Static wide dock shot | Overcast | No humans". Top-right panel sketch: {argument name="second panel" default="same ship with workers using pressure washers, dynamic water spray motion lines, figures in safety gear, debris piles"}, handwritten label: "SCENE 02 — CLEANING & STRIP-DOWN | Pressure wash + debris clear | Cloudy daylight". Bottom-left panel sketch: welding sparks as burst star lines, scaffolding structure, worker figures grinding and painting, primer sections with hatching, handwritten label: "SCENE 03 — REPAIR & REBUILD | Welding sparks + scaffolding | Primer applied". Bottom-right panel sketch: fully restored ship, clean hull, workers standing back, golden hour rays as radiating diagonal lines, handwritten label: "SCENE 04 — FULL RESTORATION | Final paint + golden hour | Completion". Bold hand-lettered title at top: "{argument name="title" default="SEA HARVEST VALLETTA — RESTORATION STORYBOARD"}". Handwritten production notes at bottom: "Cam: Static Wide-Angle Dock | Lens: Wide | Progression: Linear | Style: Photorealistic". Pencil graphite texture, ink pen outlines, rough paper feel, professional film storyboard aesthetic.

Using REFERENCE_0 as the food subject, transform the photo into a refined vintage scientific specimen plate / illustrated food diagram on a clean off-white paper background. Preserve the same sizzling-plate meal components and overall arrangement, but redraw them as detailed hand-colored botanical/culinary field-guide illustrations with ink outlines, watercolor texture, subtle paper grain, and precise diagram callouts. Add a centered Traditional Chinese title {argument name="main title" default="鐵板標本圖鑑"}, a subtitle {argument name="subtitle" default="香煎雞腿排・荷包蛋・青花菜・蘑菇醬鐵板麵"}, and small plate-number typography in the top corners. Create exactly 4 numbered annotation sections around the main dish: I. 香煎雞腿排, II. 青花菜, III. 蘑菇醬鐵板麵, IV. 荷包蛋, each with short Traditional Chinese explanatory text and dotted leader lines pointing to the corresponding ingredient. Add exactly 4 square letter markers on the dish and legend: A 香煎雞腿排, B 青花菜, C 荷包蛋, D 蘑菇醬鐵板麵. Include exactly 6 supplemental illustrations/insets: a close-up of the seared chicken surface, a broccoli specimen with approximate size marks, a noodle close-up, a fried egg close-up, a small skillet of mushroom cream sauce, and a low side-view cross-section of the entire plated meal. Add two small boxed notes near the bottom, one legend box and one observation note box, plus footer text including {argument name="observer" default="美食愛好者"}, {argument name="observation date" default="2024 年"}, and {argument name="location" default="臺灣・日常餐桌"}. Keep the composition poster-like, balanced, educational, appetizing, and more polished than the original photo.

Create a warm, cozy anime-style illustration of {argument name="character description" default="a cheerful anthropomorphic calico cat person"} enjoying a lazy holiday afternoon in a sunlit bedroom. The character is reclining diagonally across a soft cream-colored bed, wearing {argument name="outfit" default="a plain white T-shirt and loose gray sweatpants with a drawstring"}, round wire-rim glasses, smiling with closed happy eyes, one arm tucked behind the head and the other holding a foamy glass of {argument name="drink" default="golden beer"}. Add a striped calico tail curled on the bed, visible paw pads on the feet, soft blush, and relaxed body language. The room should feel peaceful and lived-in, with golden sunlight streaming through a large open window on the left, white curtains, blue sky and fluffy clouds outside, and leafy green plants on the windowsill and in the foreground. Include exactly two pillows behind the character, one mustard yellow and one green gingham, plus cream bedding and a rounded wooden headboard. On the wall above the bed, include exactly three wall decorations: a cat poster reading {argument name="poster text" default="GOOD DAY"}, a small hanging sign reading "Relax" with a paw print, and a framed landscape picture. On the wall shelf to the upper right, include exactly five visible items: one potted trailing plant, one cat-face plush or ornament, two upright books, and one framed landscape photo. Add a bedside table on the right with a lamp, a small alarm clock, and a cup; place one open beverage can on the bed near the character. In the foreground, include a low table with a mug of tea and a magazine titled {argument name="magazine title" default="NEKO LIFE"} featuring a cat illustration, plus a small potted plant. Use soft painterly shading, gentle linework, warm beige and honey tones, detailed cozy clutter, dappled sunlight, and a relaxing slice-of-life mood.

{"type":"Japanese train-hanging advertisement style anime movie poster and original novel concept art","format":"landscape 4:3 theatrical key visual, glossy commercial print design, dense typography, dramatic school music room background","setting":{"location":"sunlit Japanese high school music classroom","details":"large windows on the left with blue sky and city greenery outside, cream curtains, wooden floor, chalkboard with musical staff notation, wall clock centered above the board, bookshelf and black speaker at back left, piano and sheet music stands in foreground, warm afternoon light"},"main cast":{"count":3,"characters":[{"position":"left","description":"teenage schoolgirl in navy blazer uniform with red tie, white shirt, pleated skirt, knee socks, twin-tail black hair tied with red ribbons, playing violin with bow, determined action pose"},{"position":"center","description":"adult woman teacher or guardian in black blazer and white blouse, shoulder-length brown hair, standing confidently while playing acoustic guitar"},{"position":"right","description":"teenage schoolboy in navy blazer uniform with red tie and crest badge, messy black hair, playing golden trumpet, upright heroic pose"}],"face treatment":"faces are intentionally blurred/obscured as in a concept mockup while bodies and costumes remain sharp","instruments":{"count":3,"items":["violin","acoustic guitar","trumpet"]}},"typography":{"main title":{"position":"large center foreground","text":"私立あかつき学園","style":"huge distressed block Japanese lettering, white and blood red with black outline, cracked grunge texture, target crosshair graphic behind the red center text"},"english title":{"position":"below main Japanese title","text":"THE SPY","style":"oversized distressed white stencil letters with black outline"},"english subtitle":{"position":"right of THE SPY","text":"WHO FORGOT THE BONDS","style":"stacked compact white bold lettering"},"japanese subtitle":{"position":"under title","text":"ー命と絆とスパイー","style":"small white cinematic subtitle"}},"layout":{"sections":[{"title":"top-left tagline","position":"upper left on torn red paint banner","count":1,"labels":["奪われた人生を取り戻せ。\nその絆が、世界を動かす。"]},{"title":"right vertical tagline","position":"upper right beside trumpet player","count":1,"labels":["音楽がつないだ、\nたった一つの家族。"]},{"title":"release date badge","position":"bottom left red circular badge","count":1,"labels":["2025年","10.10","FRI","ROADSHOW"]},{"title":"staff credit box","position":"lower right black rectangle with gold border","count":5,"labels":["原作：サブサン","監督：サブサン","脚本：Chat GPT+Claude","音楽：SUNO","公式サイト：akatsuki-spy.jp"]},{"title":"bottom announcement band","position":"full width dark red strip along bottom","count":3,"labels":["少年たちの青春スパイアクション、開幕！","主題歌：私立あかつき学園のテーマ","公式サイトはこちら"]},{"title":"QR area","position":"bottom right","count":1,"labels":["black and white QR code"]}]},"color and style":"high-detail modern anime illustration blended with Japanese cinema poster graphic design, saturated navy uniforms, red accents, gold trumpet, warm classroom browns, dramatic contrast, grunge action typography, professional railway hanging ad composition","customization":{"school name":"{argument name=\"school name\" default=\"私立あかつき学園\"}","movie title":"{argument name=\"movie title\" default=\"THE SPY\"}","release date":"{argument name=\"release date\" default=\"2025年 10.10 FRI\"}","main tagline":"{argument name=\"main tagline\" default=\"奪われた人生を取り戻せ。\\nその絆が、世界を動かす。\"}","website":"{argument name=\"website\" default=\"akatsuki-spy.jp\"}"}}

Create a hyper-realistic cinematic square portrait of {argument name="subject" default="a modern mafia boss"} seated inside a luxury black car at night, framed from the chest up in a 1:1 movie-poster composition. He wears a tailored {argument name="suit color" default="black"} three-piece suit with a white dress shirt, black textured tie, black waistcoat, and a crisp white pocket square. His dark wavy hair is sharply detailed, while the entire central face area is covered by a flat opaque brown rectangular censor block, obscuring all facial features. Wisps of cigar smoke curl beside the left side of his head, even though the cigar itself is mostly out of frame, adding a noir crime-drama atmosphere. The car interior is glossy black leather and chrome, with the open door frame arcing behind him and a dark tinted window on the right. Outside the car, show a blurred city street with warm bokeh lights and a distant tower silhouette under a stormy, moody sky. Use cool desaturated tones, high contrast lighting, sharp fabric texture, crisp smoke detail, shallow depth of field, dramatic shadows, premium 8K hyper-realistic photography, cinematic noir style.

Create a cozy hand-drawn colored-pencil and crayon style illustration of a cute pastel study room corner, with warm beige paper texture and sketchy brown outlines. Show one young woman with {argument name="hair color" default="long wavy light brown hair"} sitting at a wooden desk and writing in an open journal with a black pen; her face is intentionally covered by a plain beige square censor block. She wears a frilly pink gingham dress with lace trim and a bow, seated on a wooden chair over a pink scalloped rug. The room contains exactly 1 large wooden desk with 2 visible drawers, 1 open notebook, 1 closed pink book stack, 1 teacup on a saucer, 1 pen, 1 wooden chair, 1 curtained window with flowers outside, 1 hanging vine, 4 potted plants, 1 bookshelf with colorful books, 1 small framed photo, 1 floral table lamp, and 1 framed flower picture on the wall. Include exactly 2 visible text labels: a pink hanging wall sign reading {argument name="wall sign text" default="luffynodoo"} and a cloud speech-bubble sign reading {argument name="cloud text" default="ももいろ\nメモリーズ♡"}. Surround the scene with kawaii doodles: 4 large pink flowers, 8 yellow stars, 1 blue smiling cloud, 1 pink lollipop, 1 wrapped candy, small hearts and sparkles. Use soft pastel pink, cream, honey brown, pale yellow, mint green, and baby blue, with a whimsical Japanese stationery aesthetic, delicate crosshatching, cozy sunlight, and no photorealism.

Create a highly detailed cozy miniature diorama room in a soft pastel, hand-drawn storybook-anime style, shown in a three-quarter isometric view as a tiny open box with stitched beige fabric walls and floor edges. The scene centers on a young woman seated at a wooden writing desk, wearing a pink pastel plaid dress with ruffled off-shoulder sleeves, lace trim, a bow at the chest, and a delicate necklace; her long {argument name="hair color" default="wavy light brown"} hair falls over her shoulders, and her face is covered by a flat plain tan square censor patch. She is writing with a pen in an open notebook on the desk. The desk contains exactly 10 visible tabletop items: one open notebook, one pen in her hand, one teacup on a saucer filled with tea, one pink closed notebook, two round buttons, one blue wrapped candy, one pink striped candy, one tiny yellow candy, and one small flower. Add a warm wooden window on the left with lace curtains and garden light shining through; outside the window are pink roses and greenery. Include exactly three fluffy white clouds floating near the window, plus small star and flower doodles around the room. On the right wall place a pink stitched name plaque reading {argument name="wall sign text" default="luffynodoo"} with tiny heart doodles around it, plus two blue cloud doodles and two yellow star doodles. Furnish the right side with a small wooden shelf and cabinet containing exactly four upright books, one potted succulent, one lampshade, and one jar labeled {argument name="jar label" default="CANDY"}. Add a woven basket plant on the floor, a wooden chair, a small red rug, scattered embroidery-like flowers, a looping yellow thread, and delicate candy, flower, star, and cloud motifs throughout. Use warm sunlight, soft shadows, realistic miniature textures, embroidered fabric seams, intricate lace, wood grain, pastel pink and beige tones, cute whimsical atmosphere, high detail, square composition, no extra text except the wall sign and candy jar label.

Task Objective: Generate a set of 16 chat-style sticker emojis based on the reference images. The result should feel expressive, highly shareable, and stylistically unified — but intentionally chaotic, low-quality, and humorous. The final output should look like: Someone who cannot draw, using a mouse in MS Paint, randomly doodling and writing text at the same time — messy, awkward, low-effort, but unexpectedly funny. ———————— Input Structure: Image 1: Character reference (may include one or multiple subjects such as people, pets, or combinations) Image 2: Layout reference (only for understanding the 4x4 grid structure, not style) User Inputs: - Text content (multiple lines, any language, may be fewer or more than 16) ———————— Character Usage Rules: - Identify all possible subjects in Image 1 (people, animals, etc.) - Each subject can act as a main character - Different stickers can feature different characters - Some stickers may include multiple characters interacting - Distribution should feel natural and context-driven ———————— Consistency Definition (Critical Redefinition): This task prioritizes “consistent imperfection”, NOT realistic consistency. The same character across stickers: Does NOT need to look identical, but must feel like it was drawn by the same unskilled person. Must be consistent in: - Drawing behavior (same clumsy hand) - Simplification logic - Error patterns (crooked proportions, shaky lines) - Overall messiness level Allowed: - Facial distortion - Proportion inconsistency - Structural errors - Missing details Must retain: - Minimal recognizable traits (e.g. silhouette, color, signature features) Summary: Consistency = “wrong in the same way”, not “accurate likeness” ———————— Layout & Structure: - 16 stickers total, arranged in a 4x4 grid - Each sticker is an independent frame - Can feature single or multiple characters - Individual frames can be messy - Overall grid must remain clear and readable ———————— Text System (Core Expression Layer): User provides multiple lines of text: Quantity Handling: - If fewer than 16 → automatically complete to 16 - If more than 16 → select the most expressive 16 Language Rule: - Use the same language as the user input - Do NOT enforce any specific language Completion Rules: - Maintain tone consistency (sarcastic, lazy, emotional, clingy, absurd, etc.) - Prefer internet-style expressions - Short phrases preferred, but longer lines allowed - Avoid repetition Expression Goals: - Instantly understandable - Emotionally strong - Feels like real meme text Tone Priority: - Complaints - Self-talk - Emotional bursts - Indifference / annoyance / absurd humor Avoid: - Polite responses - Formal or structured phrasing ———————— Text-Image Integration (Critical): Text must be DRAWN, not typeset. Must: - Look like mouse handwriting (crooked, shaky, uneven size) - Be messy (tilted, overlapping, misaligned) - Have inconsistent spacing - Be placed freely (on face, beside, edges, etc.) Allowed: - Repeated letters (aaaaa) - Stretched words (soooo tired) - Random punctuation (????!!!) - Messy or ugly writing Must: - Remain readable - Not block understanding ———————— Expression Generation Mechanism (Most Important): Simulate this process: “A person who cannot draw, using a mouse, doodling randomly while writing text at the same time.” Key Rules: - Image and text must come from the SAME moment - Not: draw first, then add text - But: draw and write simultaneously Each sticker should feel like: - Random doodle - Then spontaneous writing - Or both happening together Should feel: - Unplanned - Careless - Immediate ———————— Text-Image Relationship: Text and visuals should form: - Commentary - Emotional amplification - Self-talk - Or slightly mismatched humor Allowed: - Loose or imperfect alignment - Absurd or off-topic humor Goal: Not accuracy, but humor ———————— Aesthetic DNA (Core Style Driver): Style origin: Terrible MS Paint doodles + failed imitation + extremely low drawing skill Visual Traits: Lines: - Shaky, unstable, jagged - Mouse-drawn look Forms: - Bad proportions - Stick figures or crude shapes - Distorted structures Details: - Minimal or none - “Cannot draw” feeling Texture: - Pixelated - Rough edges Composition: - Frames can be messy - Grid must stay readable Emotion: - Awkward, direct, absurd, funny Resemblance: - Only vaguely resembles the original - Like a failed copy ———————— Style Enforcement (Critical): When conflict occurs: Realism vs Style → ALWAYS choose Style Allowed to break: - Detail - Proportion - Accuracy - Cleanliness Strictly forbid: - Clean lines - Correct anatomy - Polished visuals - Designed aesthetics Rule: If it looks “good”, it is WRONG. Force it back to messy, ugly, low-effort. ———————— Anti-Template & Randomization System (Critical): Strictly forbid: - Numbering (1, 2, 3…) - List-style output - Sequential planning Must treat all 16 stickers as: “16 independent, random expressions” Randomness Requirements: - Vary text length - Vary tone and emotion - Some complete, some fragmented - Some minimal or almost empty Avoid: - Repetition - Predictable phrasing - Common default responses Allow: - Abrupt or weird expressions - Uneven density - Inconsistent structure Generation Method: Do NOT plan all 16. Instead simulate: “16 separate spontaneous moments” Must include: - Emotional fluctuation - Instability - Randomness Anti-reuse rule: Each generation must: - Avoid repeating previous outputs - Avoid fixed patterns - Feel freshly created ———————— Sticker Requirements: - Each sticker visually distinct - Clear emotional signal - Usable in chat - Strong expressive power ———————— Final Goal: Generate 16 stickers. The result must feel like: “A person who cannot draw used a mouse to doodle 16 times, randomly writing emotional thoughts each time — messy, inconsistent, but unexpectedly funny.” NOT: “A clean, well-designed AI sticker set”

Futuristic portrait of a young woman facing camera, wearing a {argument name="clothing" default="transparent neon jacket"} with glowing {argument name="accent colors" default="green and orange"} edges, large illuminated logo on chest, black inner outfit, sleek sunglasses, soft smoke light trails behind, {argument name="background" default="dark teal background"}, cyberpunk fashion campaign, ultra-realistic textures, cinematic lighting, sharp focus, luxury sportswear branding style, 8k Style keywords: neon edges, glowing logo, fashion campaign, high-end branding, moody lighting

Using REFERENCE_0 as the base landscape, remove all YAMAP branding, date, route, and metric overlay text, then transform the scene into a polished anime-style travel illustration while preserving the mountaintop viewpoint over the city, lake, distant mountains, hazy sky, and green forested foreground. Add one foreground character: a young anime hiker viewed from behind/three-quarter back, standing on an old stone castle-wall lookout and gazing over the scenery. Give the character {argument name="hair color" default="long aqua-blue hair"} in a high ponytail with a blue ribbon and star hair ornament, a light blue hiking shirt, beige cargo hiking pants, brown hiking boots, fingerless gloves, and a large dark hiking backpack. Add exactly 4 visible gear/accessory elements on or near the character: 1 rolled sleeping mat strapped to the backpack, 1 dangling blue star charm, 1 small blue bird-like companion perched near the shoulder, and 1 small hiking pouch/strap detail. Add weathered stone blocks under the character and soft blurred green leaves in the lower foreground for depth. Keep the composition scenic and peaceful, with bright morning light, gentle atmospheric haze, crisp anime linework on the character, and a softly detailed semi-realistic background. Do not include any text, logos, UI, or watermark.

Design a '4:5' product poster for an '{argument name="fruit" default="orange"}' juice 3d bottle using playful inflatable-plastic packaging surrealism where the bottle behaves like a squeezed toy object mid-pressure. The poster should communicate juiciness, tension, and tactile freshness through a bottle that visibly bulges, stretches, and compresses like a soft object being squeezed from the inside out. SUBJECT: A single bottle dominates the center-left, illustrated in a semi-3D stylized form (not photoreal). The bottle is visibly distorted—its midsection bulges outward while the neck is slightly compressed, as if internal juice pressure is pushing against the container walls. The liquid inside exaggerates this effect, forming rounded convex surfaces pressing against the plastic. The cap is slightly tilted from pressure. The bottle feels elastic, alive, and reactive rather than rigid. COMPOSITION: The composition mechanic is “internal pressure distortion”. The eye enters through the most inflated part of the bottle (center bulge), then follows curved tension lines outward toward stretched typography on the right side. The camera uses a slightly low, close-up perspective with mild fisheye distortion, amplifying the sense of pressure and expansion. The bottle leans diagonally into the frame as if pushing against invisible resistance. Negative space on the right is intentionally stretched and warped, echoing the bottle’s deformation. TYPOGRAPHY: Headline: “{argument name="headline" default="SQUEEZE BACK"}” in thick, rounded, inflated lettering that mimics air-filled plastic. The type is physically distorted—letters stretch horizontally near the bottle and compress toward the edges, as if affected by the same pressure force. The headline sits to the right of the bottle but partially overlaps its bulging edge. Subheadline: “Juice that pushes back.” in a narrow condensed sans, straight and rigid, contrasting the soft headline. Small circular microcopy labels (e.g., “100% organic”, “no added sugar”) appear like printed stickers slightly warped by the surface tension. LIGHTING / GRAPHIC TREATMENT: Soft studio lighting adapted for stylized rendering—broad diffused highlights across curved surfaces, subtle specular streaks that follow the deformation. No realistic reflections—only controlled, simplified highlight bands that emphasize volume and elasticity. COLORS: 'orange' juice bottle color and 3d header text. Small accents of 'bright lime green' for sticker elements. High contrast between glowing orange and dark surroundings. BACKGROUND: A smooth dark gradient field with subtle radial tension lines expanding outward from the bottle, like invisible pressure waves. No texture clutter—clean but dynamically warped space. STYLE: Neo-pop surreal product illustration with inflatable material logic, playful distortion, and high-impact commercial clarity. FINISH: Soft plastic-like rendering, clean edges, controlled gradient transitions, no noise or grain, polis

Japanese negative film aesthetic, {argument name="setting" default="rooftop summer scene"}, soft natural sunlight, slight overexposure highlights, low contrast, muted faded colors, subtle grain a stunning beautiful young woman with subtle sensual presence, natural body line, effortless charm wearing a slightly {argument name="clothing" default="oversized white shirt loosely unbuttoned at the collar"}, paired with high-waisted shorts; shirt softly moving in the wind, occasionally slipping off one shoulder holding a {argument name="prop" default="cold glass bottle drink with condensation"}, one hand lifting it near her neck or cheek, fingers lightly touching the surface subject sitting or leaning on rooftop edge, body relaxed but with slight weight shift, one hand supporting behind, torso subtly opening, one knee bent and the other leg softly extended hair gently blown by summer wind, loose strands across face expression calm and distant, lips slightly parted, looking toward camera or slightly away open sky, minimal environment, a light plastic bag resting beside her moving slightly with the wind imperfect composition, quiet isolated mood, nostalgic and reflective, “memory-like realism”, subtle sensuality through natural gesture --2:3

{"type":"wide hand-drawn Japanese educational infographic","canvas":{"aspectRatio":"16:9 landscape","background":"warm off-white paper texture","style":"friendly marker-and-watercolor explainer, sketchnote, clean readable Japanese typography, blue/orange/green/red accents, rounded boxes, arrows, small doodle icons"},"headline":{"text":"{argument name=\"headline text\" default=\"まだ『画像生成＝絵を描くAI』だと思っていませんか？\"}","position":"top center","emphasis":"orange underline and large orange phrase for 『画像生成＝絵を描くAI』"},"subheadline":{"text":"{argument name=\"subheadline text\" default=\"その認識、もう少しアップデートが必要かもしれません。\"}","position":"below headline","color":"blue","decorations":"small radiating blue strokes"},"layout":{"mainSections":{"count":3,"arrangement":"three large rounded rectangular panels across the middle, connected by thick blue right-pointing arrows","sections":[{"title":"第１段階","subtitle":"アート","position":"left","borderColor":"blue","content":"framed colorful mountain-and-lake landscape illustration; text block titled 代表的なモデル例 with 3 bullet items: Stable Diffusion, DALL-E, Midjourney; bottom line きれいな絵・世界観・雰囲気をつくる","icons":{"count":3,"descriptions":["small picture icon","paintbrush","artist palette"]}},{"title":"第２段階","subtitle":"デザイン","position":"center","borderColor":"green","content":"mock website or poster layout with headline 未来を、デザインするか。, small paragraph text, button 詳しくはこちら →, document typography, image placeholder, and chart elements; text block titled 代表的なモデル例 with 2 bullet items: GPT-Image-1, Nano Banana Pro/2; bottom line 文字・レイアウト・訴求を組み立てる","icons":{"count":3,"descriptions":["T typography icon","layout grid icon","crossed pen or design tool icon"]}},{"title":"第３段階","subtitle":"図解・理解","position":"right","borderColor":"red","content":"simple flowchart diagram with 課題 at top, branching to 原因, 本質（構造）, 効果, and 解決策, with arrows and a small bar chart; text block titled 代表的なモデル例 with 1 bullet item: GPT-Image-2; bottom line 構造を見せる・意味を伝える・思考を前に進める"}]},"lowerExplanatoryElements":{"count":4,"items":[{"type":"yellow rounded callout","position":"lower left","text":"文字が “画像の中で機能する” ようになった瞬間、AI画像生成は “画像” を超え始めた。","icon":"light bulb"},{"type":"process strip","position":"lower middle-right","count":3,"labels":["見る画像","読む画像","考える画像"],"icons":["eye","document","light bulb"],"connectors":"blue arrows"},{"type":"bold statement","position":"near bottom center","text":"いま起きているのは、絵づくりだけではない。『思考の可視化』だ。","emphasis":"large blue handwritten 『思考の可視化』 with yellow underline"},{"type":"bottom ribbon callout","position":"bottom left to center","text":"この変化に気づくと、AI活用の見え方が変わる。","icon":"small sprout in a pot"}]},"characters":{"count":2,"description":"same friendly teacher character appears twice on the right edge, anime-style young man in a light blue shirt holding a pointer; face area intentionally covered by a beige rectangle","instances":[{"position":"top right","pose":"holding pointer upward beside headline"},{"position":"bottom right","pose":"holding pointer and green notebook labeled メモ, next to speech bubble 一緒に学んでいきましょう!"}],"customCharacter":"{argument name=\"teacher character\" default=\"young male teacher with brown hair, light blue shirt, pointer, face obscured by beige square\"}"},"extraDecorations":{"count":4,"items":["yellow light bulb doodle at top left","small sunburst marks around important text","thin hand-drawn underlines in orange, blue, and yellow","rounded speech-bubble outlines and sketched arrows"]},"textRequirements":"All Japanese text must be crisp, legible, and accurately placed; preserve the hand-lettered sketchnote feel while keeping diagrams clean and instructional.","theme":"{argument name=\"main message\" default=\"AI image generation has evolved from art to design to diagrams and understanding, making text functional inside images and visualizing thought.\"}"}

Create a bold street-fashion poster featuring a {argument name="subject" default="handsome man with short black hair"}, and a fearless expression. Dress him in a {argument name="jacket style" default="glossy emerald satin bomber jacket"} and black pants. Place him in front of a {argument name="background" default="distressed concrete wall with emerald, black, and white layered textures"}. Use dramatic editorial lighting, crisp contrast, and a premium urban mood. Add bold uppercase typography with a clean modern poster style.