Free GPT Image 2 Prompts & Examples

Create a cheerful, kawaii-style wide selfie photo of {argument name="group description" default="four young female friends"} celebrating around a birthday cake indoors. The camera is held close at arm’s length from the right side, creating a playful group selfie angle with everyone leaning tightly together toward the lens. Use a bright cozy bedroom or party-room background with soft daylight, white walls, pastel bedding, and small shelf decor. Place a frosted birthday cake in the lower foreground, slightly out of focus, topped with exactly 12 colorful lit birthday candles in pink, yellow, blue, green, and striped patterns. The people wear soft pastel casual sweaters and tops in cream, pink, white, and pale blue, with long dark or brown hair; make the faces anonymized by smooth soft rectangular blur blocks while keeping hair, clothing, hands, and poses visible. The left person reaches one hand toward the camera in a playful open-palm gesture, and the right person’s pale blue sleeve fills the near-right edge like they are taking the selfie. Add hand-drawn crayon doodles over the photo in a cute energetic style: a huge pink uppercase headline reading {argument name="headline text" default="HAPPY DAY"} across the top; exactly 10 heart doodles including pink and white hearts around the group; exactly 3 flower doodles, one large pink flower at top left, one small pink flower near bottom left, and one small yellow star-flower near the center; exactly 1 yellow crown doodle above the second person’s head; exactly 3 star doodles; exactly 5 green leafy sprig or curved accent doodles; and exactly 5 small pink motion marks or exclamation squiggles near the headline and right side. Use a clean bright aesthetic with high-key lighting, pastel pink and mint marker colors, whimsical birthday energy, shallow depth of field, and a polished social-media photo look. Avoid extra people, avoid realistic printed text besides the headline, and do not add watermarks.

Transform the image into a {argument name="art style" default="minimalist hand-drawn doodle illustration"}. Use simple {argument name="ink color" default="black"} brush-pen outlines with slightly wobbly, imperfect strokes. Reduce all details to cute, childlike shapes while preserving the main subject and composition. Naive sketchbook aesthetic, playful and whimsical character design, clean white background, minimal linework, expressive simplicity, handmade doodle style, black ink drawing, charming imperfections, simple cartoon illustration. Format 3:4.

To Xiaohongshu Food Cover Photo Edit. {argument name="preservation" default="Preserve the original food type and main content"}

Use 100% of the face and body from my reference image. Create a cinematic winter scene with a side face close-up view, inspired by a Minecraft death screen. The person is lying face up in the snow, wearing {argument name="clothing" default="casual winter clothing"}, in a realistic pose, with cool atmospheric lighting, slightly desaturated tones, and visible footprints in the snow. Surround the person with floating or placed pixelated objects in the Minecraft style, including a {argument name="weapon" default="diamond sword"}, bow, pickaxe, emerald, steak, Eye of Ender, music disc, and a mob of spiders. Add a realistic Minecraft "You Died!" overlay on top with pixel font, including "Score: 0," the reticle symbol, and two retro UI buttons labeled "Respawn" and "Title Screen." Naturally blend the realistic photography with pixel-art game elements. Soft shadows, melancholic winter aesthetic, ultra-detailed, vertical composition, DSLR realism, subtle depth of field, high contrast, 9:16 aspect ratio. Negative Prompt: "Low quality, blurry face, distorted body, extra limbs, incorrect facial features, cartoonish body, unrealistic"

Create a realistic warm bedtime photography scene of an adult woman with {argument name="hair color" default="long dark brown hair"} sleeping on her side on a gray bed, posed curled up with her knees bent and hands tucked under her cheek on a pillow. She wears a fitted {argument name="top color" default="cream ribbed camisole tank top"} and {argument name="shorts color" default="gray soft lounge shorts"}. Her face is intentionally hidden by a plain opaque brown square privacy mask, centered over the face, with no facial features visible. The room is dim and cozy, lit by a warm cylindrical bedside lamp on a wooden nightstand at the left, with a small potted plant beside it. Use gray pillows, gray sheets, a textured beige blanket in the foreground, and dark gray curtains across the background. Shoot from a slightly elevated bedside angle in a horizontal 4:3 composition, shallow depth of field, soft skin tones, warm amber lighting, photorealistic detail, quiet good-night mood, no text, no watermark, no extra people.

Goal: Create a hand-drawn breaking-news style infographic in Japanese about {argument name="headline topic" default="Claude Code implementing infographic image generation for X breaking-news posts"}. Canvas: 16:9 horizontal image, dark navy chalkboard background, rough marker and chalk texture, high-contrast red, orange, yellow, white, and cream colors. Use imperfect hand-lettering, sketchy outlines, brush strokes, and doodle icons. No photorealism. Layout: Left side has 1 tall vertical red breaking-news banner with a white border, large white Japanese text 「速報」 stacked vertically, and 1 yellow lightning bolt near the bottom. Top center has 1 wide red-to-orange brush-stroke headline banner with 1 yellow lightning icon on the left and 1 megaphone icon on the right. Under it is 1 rounded source bar. Main middle area contains exactly 2 large rounded rectangular information panels stacked vertically. Right side contains exactly 2 smaller rounded cards stacked vertically: a source-information card above and a date/time card below. Bottom row contains exactly 3 hashtag pills. Text content: Top headline banner text: 「Claude Code」 on the first line, then 「インフォグラフィック画像生成機能を」, then 「X速報投稿に実装」. Source bar text: 「SOURCE: zenn-blog-automation PR #217」. Right source card text: 「情報源:」 then 「zenn-blog-automation」 then 「PR #217」. First main panel text: 「GPT-IMAGE-2で」, 「breaking_newsテーマの」, 「インフォグラフィックを生成し」, 「X速報投稿に添付できるようになった。」 Include a small orange label reading 「GPT-IMAGE-2」 on an image-thumbnail icon. Second main panel text: 「画像生成失敗時は」, 「テキストのみ投稿にフォールバック。」 Date/time card text: 「検出日時:」 then {argument name="timestamp" default="2026/06/24 19:15 JST"}. Bottom hashtag pills must read exactly 3 labels: {argument name="hashtag one" default="#claudecode"}, {argument name="hashtag two" default="#anthropic"}, and {argument name="hashtag three" default="#ai"}. Discrete visual elements to include: exactly 1 left vertical banner, 1 top headline brush banner, 1 source bar, 2 main info panels, 2 right-side cards, 3 hashtag pills, 2 yellow lightning bolts total, 1 megaphone, 1 document icon, 1 clock icon, 1 image-thumbnail icon with mountains, 2 orange arrow icons pointing right, 1 cloud-with-X failure icon, and 1 speech bubble with three dots. Visual style: Energetic Japanese breaking-news infographic, marker poster look, slightly irregular white handwritten type, orange rounded borders, yellow underline highlights under important phrases, small decorative radiating accent marks around the megaphone, banner, and hashtags. Keep the composition dense but readable, with strong hierarchy and a dramatic emergency-news feel. Constraints: Use Japanese text exactly as specified, do not add extra sections or extra hashtags, keep all icons hand-drawn, avoid logos, avoid watermarks, and maintain the rough chalk-and-paint texture throughout.

Goal: Create a glossy, high-contrast YouTube-style tech video thumbnail about choosing between {argument name="tool one name" default="Codex"} and {argument name="tool two name" default="Hermes"}. Canvas: 4:3 landscape thumbnail, 1024×768 composition, dark black-to-charcoal studio background with a dramatic white spotlight shining down from the top center, subtle smoky haze, cinematic rim lighting, and a premium “oily” glossy thumbnail look. Layout: Center a young male presenter from chest up, facing camera, wearing a plain black V-neck T-shirt. His face should be intentionally anonymized with a smooth blurred rectangular skin-tone mask. His arms are raised symmetrically, palms open upward as if weighing two choices. Place exactly 2 floating option graphics above his hands: on the left, a glowing blue cloud-shaped terminal icon with a white command prompt symbol; on the right, a black-and-white anime-style female avatar bust with glossy hair and a white outline glow, also anonymized with a small blurred square over the face. Text content: Add exactly 2 stacked text lines across the lower third. Top line: large bold white Chinese text with black shadow, {argument name="top headline text" default="到底选"}. Bottom line: huge mixed gold-and-white text reading {argument name="bottom comparison text" default="Codex 还是 Hermes?"}; make “Codex” and “Hermes?” metallic gold with bevel, gradient, black outline, and drop shadow, while the middle Chinese word stays white. Include one small golden star lens flare near the right side of the word “Hermes”. Subject details: The presenter has short messy brown hair lit by the overhead spotlight, natural hands with visible palms, and a neutral pose suggesting comparison or uncertainty. The left icon should glow electric blue with a neon rim; the right anime avatar should be monochrome with strong white edge highlights. Visual style: Hyper-polished Chinese tech creator thumbnail, sharp subject cutout, intense contrast, glossy 3D text, dramatic lighting, saturated highlights, clean composition, no extra logos, no watermark, no additional text.

Goal: Create a cinematic social media video thumbnail / award post image showing a fantasy shogi match transforming into a battlefield, with a dark PixVerse-style video-card presentation. Canvas: 600×483 px landscape card on a black background with rounded corners. The upper 82% is a cinematic image still; the lower strip is black with a large title. Use warm firelight, deep shadows, shallow depth of field, dramatic movie lighting, and high-detail photorealistic animal characters. Main scene: Inside a dim traditional Japanese room, two anthropomorphic dogs sit across a wooden shogi board that has become a miniature battlefield. On the left is {argument name="left character" default="PONTA, a pink Shih Tzu"}, small and fluffy, wearing a pale pink kimono, holding a thin wooden pointer toward the board with a confident strategic expression. On the right is {argument name="right character" default="a worried bulldog opponent"}, large and wrinkled, wearing a dark robe, gripping both sides of his head in despair. Behind them are blurred lanterns, flames, and traditional wall art. Board details: The shogi board fills the lower center of the image in perspective. Show a dense miniature battle scene on top of the grid: exactly two opposing armies, one accented in red/orange and one accented in blue. Include about 64 visible miniature soldier figures total, mixed with shogi pieces, banners, spears, and glowing embers. A bright fiery clash burns near the center of the board. Make the wooden grid and at least one shogi piece with Japanese characters visible near the front edge. Overlay text and UI: Add a rounded dark-brown pill label at the top left reading {argument name="award label" default="Top 10 Popular Picks"} in white sans-serif text. At the lower left of the image still, place a circular colorful avatar icon followed by white text reading {argument name="creator line" default="PONTA (Pink Shih Tzu)"}. At the lower right of the image still, place the white duration text {argument name="duration" default="00:08"}. In the black bottom area, add the large white title text {argument name="video title" default="Shogi on the Battlefield"} aligned left. Visible discrete text elements: exactly 4 text elements: the award label, the creator line, the duration, and the bottom title. Visible major characters: exactly 2 dogs. Visible UI elements: exactly 1 rounded award pill, exactly 1 circular avatar, exactly 1 video duration marker, exactly 1 bottom title area. Style constraints: photorealistic yet fantastical, cinematic, warm orange highlights, dark vignette, high contrast, crisp central details, no extra logos, no watermark, no additional captions beyond the specified text.

Goal: Create a dreamy warm pastel mirror-selfie fantasy scene featuring {argument name="main subject" default="a young woman"} in a cozy bedroom, surrounded by miniature chibi versions of herself. The image should feel like a soft magical Instagram aesthetic: glowing candles, fairy lights, flowers, motivational wall art, and a large elegant oval mirror. Canvas: Vertical portrait image, 2:3 aspect ratio. Warm peach, blush pink, rose-gold, and amber color palette. Soft cinematic lighting with shallow depth of field, glowing bokeh particles, and a dreamy haze. Main composition: Place a large wooden oval mirror in the center, occupying most of the frame. In the mirror reflection, show the main subject standing upright and taking a mirror selfie with a black smartphone held in her right hand. Her face is intentionally covered by a smooth rectangular skin-tone blur/blank mask, with no facial features visible. She has long dark brown hair styled half-up with volume at the crown and loose straight lengths over her shoulder. She wears {argument name="outfit" default="a black embroidered traditional kurta with white floral embroidery, magenta scalloped trim, and a matching black dupatta"}. Add small hoop earrings and a beige quilted shoulder bag with a gold chain strap on her left shoulder. Miniature character details: Include exactly 6 small chibi versions of the same woman, all with the same long dark hair, black embroidered outfit, sandals, and blank blurred face style. Count and place them as follows: 1 chibi sitting on the upper edge of the mirror, leaning forward with one hand hanging over the frame; 1 chibi sitting on the left shelf reading an open book; 1 chibi sitting at the bottom left near flowers, holding a bouquet of pink tulips; 1 chibi standing on the right shelf holding a small camera; 1 chibi standing on a small mug or pedestal on the lower right, one hand raised as if waving; 1 chibi partially visible in the bottom-right corner, sitting and looking inward. Make all chibis cute, doll-like, and proportionally tiny compared with the main woman. Room and props: Design a cozy pastel bedroom or vanity corner with wooden shelves on both sides of the mirror. Add pink tulips in a vase on the upper left, small pink baby’s-breath flowers, a large white ceramic vase, candles in glass holders, jars of glowing fairy lights, a tiny teddy bear near the bottom center, a pink notebook with a pen on the vanity surface, and warm string lights with star-shaped bulbs draped around the mirror and shelves. Add floating golden sparkles throughout the air. Visible text: Include exactly 3 readable motivational signs: a framed pink poster on the upper right reading {argument name="upper right poster text" default="YOU ARE enough"} with a small heart underneath; a pink sign on the left reading {argument name="left sign text" default="You glow girl"} with a heart; a smaller orange-pink sign on the lower right reading {argument name="lower right sign text" default="Focus on the good"}. Use handwritten or soft modern script fonts where appropriate. Visual style: Photorealistic base with whimsical miniature figures, soft editorial beauty lighting, warm candle glow, rose-gold highlights, realistic wood mirror frame, delicate embroidery detail, cozy feminine decor, magical sparkle overlay, high-resolution polished composition. Constraints: Keep the central mirror reflection realistic and dominant. Preserve the blank blurred-face look for the main subject and all 6 chibi figures. Do not add extra chibis beyond the exact count of 6. Avoid harsh shadows, neon colors, cluttered text, watermarks, logos, or distorted hands.

Create a transparent-background decorative Japanese lower-third banner for a short video overlay. Canvas: wide horizontal 3:1 ratio, about 1200×400 px, with a light gray checkerboard transparency preview behind the artwork. Main object: a cream-white rounded rectangular speech-bubble plaque spanning most of the width, with layered offset shadows and thin gold and pale blue outline strokes, slightly vintage stationery style. Layout: place a small tab on the upper left of the plaque containing a circular icon of a laptop with code brackets and the label {argument name="category label" default="開発者コメント"}; place the main headline centered in large elegant Japanese Mincho-style calligraphy reading {argument name="headline text" default="ハイネを消す方法"}, with the characters for “消す” colored muted blue while the rest is black. On the left side inside the plaque, include exactly 1 watercolor-style dolphin illustration facing right, pale blue-gray. On the upper right, overlapping the plaque, add exactly 1 comic burst sticker with scalloped edges, beige outline, and handwritten text {argument name="sound effect text" default="ピコン"}, plus exactly 2 exclamation mark rays. Decorative details: include exactly 6 small sparkle stars distributed around the dolphin, headline, and right edge; exactly 1 thin blue wavy underline beneath the headline; exactly 4 tiny dots near the lower right center; subtle beige halftone dot patches on the left and right; small gold ornamental corner flourishes. Visual style: soft pastel cream, dusty blue, muted gold, clean Japanese YouTube Shorts caption graphic, gentle drop shadows, crisp vector edges mixed with light watercolor texture. Constraints: no characters or people, no extra text, keep the banner readable with generous empty space, preserve a transparent PNG overlay feel.

Goal: Create a bright Korean YouTube-style thumbnail about improving sleep, combining a lifestyle portrait with bold editorial typography. Canvas: 1200×675 horizontal 16:9 thumbnail, soft warm indoor bedroom/living-room background, thin white rectangular border inset from the edges. Main subject: On the right half, show a young East Asian woman with long dark wavy hair wearing an off-shoulder cream knit sweater, sitting in a cozy room and touching her cheek with one hand. Her face is intentionally covered by a large centered square blur/mosaic block in muted skin-tone colors, obscuring all facial features. Keep the portrait softly lit, feminine, and clean. Layout: Left half is dominated by oversized stacked Korean headline typography with thick white outline and subtle black drop shadow. Use three main headline lines: line 1 black text, line 2 pink text, line 3 pink text. The woman overlaps slightly behind the typography, creating a magazine thumbnail composition. Text content: Include exactly 4 text blocks: 1) a pink brush-stroke banner at upper left reading {argument name="top banner text" default="요즘 제가 빠진!"}; 2) the main headline reading {argument name="headline text" default="삶의 질을 높여준 8가지"}; 3) a pink oval brush label near the lower center reading {argument name="callout text" default="꼭 해보세요!"}; 4) a small handwritten note on the right reading {argument name="side note text" default="일상이 달라져요!"}, with the second phrase in pink and underlined. Decorative elements: Include exactly 2 pink outline hearts, one near the top banner and one near the right handwritten note. Include exactly 4 star/sparkle doodles: two white sparkle marks above the headline near the center, one pink star outline on the left edge, and one small white star near the lower callout label. Add hand-drawn underline strokes under the callout and side note. Visual style: Korean lifestyle content thumbnail, cute and polished, high contrast, bold rounded sans-serif Korean lettering, black and hot-pink palette with white outlines, soft beige background, playful brush strokes, clean social-media design. Constraints: Do not add extra text blocks, extra people, logos, or watermarks. Keep the face fully obscured by the square blur block while preserving the hair, sweater, hand, and shoulders.

Create a cinematic photorealistic blog cover image showing a chaotic crowd of zombie-like shoppers pressed against the glass entrance of a Japanese retail store on a bright sunny day. The camera is inside the store looking outward through automatic sliding glass doors, wide 16:9 composition, low slightly wide-angle perspective, intense backlight, blue sky and small clouds outside, reflections on the glass, metal door frames, posters on the side windows, and a stack of red shopping baskets at the lower right. About 14 pale, dirty, frantic people in torn light-colored clothes crowd the doorway with outstretched hands and smeared grime, like a consumer stampede. Exactly 10 visible smartphone screens are held up toward the camera, each displaying large red Japanese sale text {argument name="phone screen text" default="本日ポイントアップ↑"} on a white background with small red confetti dots; include 3 large foreground phones, 5 raised background phones, and 2 smaller partially visible side phones. Include 1 large raised rectangular placard near the upper left-center with the same red Japanese text, and 1 bright red-and-yellow store poster on the right window advertising {argument name="poster sale text" default="ポイント 5倍"}. Add exactly 9 opaque dark gray rectangular anonymization blocks covering faces in different sizes, including one large block in the center foreground, one large block on the right foreground, one large block on the left foreground, and six smaller blocks across the background crowd. Keep the scene realistic rather than cartoonish: gritty clothing texture, bluish corpse-like hands, motion blur on reaching arms, shallow depth of field, high-contrast daylight, documentary disaster-movie atmosphere, no extra readable English text, no watermark.

Create a photorealistic vertical portrait of a miniature tiny doll-like girl sitting cross-legged in the open palm of a giant realistic human hand, with a second giant hand entering from the right and gently pinching or adjusting near the tiny girl's ear to emphasize the scale illusion. The girl should have {argument name="hair color" default="long chestnut-brown hair"}, smooth realistic skin, a calm neutral expression, and a face intentionally obscured by a plain centered rectangular censor block in warm medium brown. She is wearing {argument name="outfit color and style" default="a pastel dusty-pink embroidered traditional kurta pajama set"} with delicate lace and threadwork around the neckline, cuffs, hem, and trousers, plus matching soft slippers. Her arms are folded across her chest and her legs are crossed neatly, seated securely on the palm. Show the palm lines, skin texture, fingers, nails, and natural hand creases in sharp macro detail, making the human hands look very large compared to the tiny seated figure. Background: {argument name="background setting" default="a nighttime city skyline"} with tall buildings and colorful out-of-focus bokeh lights in blue, purple, red, gold, and white. Use shallow depth of field, cinematic lighting, warm skin tones, realistic shadows where the tiny body touches the hand, high detail, vertical 2:3 composition, centered subject, magical miniature realism, no text, no watermark, no extra people, and no distorted anatomy.

{argument name="platform" default="Netflix"} original series poster, close-up portrait, intense expression, cinematic color grading, dramatic atmosphere, subtle film grain, premium streaming-series aesthetic, shallow depth of field, ultra-detailed face, award-winning photography, 8K.

Create a warm, detailed anime-style morning livestream illustration inside a cozy wooden cabin cafe. The scene features exactly 2 main characters seated at a wooden table: on the left, a small cat-eared adventurer girl with short light-brown hair, amber eyes, fluffy ears and tail, wearing a dark brown explorer outfit, red scarf, fingerless gloves, backpack straps, and a rifle or long adventuring gear strapped behind her; she raises one hand in a shy greeting. On the right, a taller streamer girl with long twin-braided {argument name="hair color" default="turquoise hair with pink tips"}, oversized black headphones with cyan glowing rings, a white hoodie with a pixel heart design in teal, pink, orange, and green, and a tiny cow hair clip; her face is deliberately covered by a plain square censor block in a muted gray-beige color, while she waves with one hand. Place exactly 7 visible cow-themed items: 1 cow hair clip on the streamer’s head, 3 round hanging cow lanterns near the ceiling, 1 cow-face mug on the breakfast tray, 1 cow plush at the front-left corner of the table, and 1 cow doodle on the rear chalkboard. Add a breakfast tray in front of the streamer with exactly 4 food items: toast, a sunny-side-up egg, cherry tomatoes, and leafy greens. Add a large microphone on the right edge of the table for livestreaming, a small potted plant near the center front, and a wooden sign on the front-right reading 「いってらっしゃい & いってきます!」 with paw prints. Add a small spiral notebook on the table labeled 「今日の予定」 with exactly 3 checkbox lines: 「配信」, 「牧場シミュレーター」, and 「冒険の準備♪」. In the upper-left, add large bubbly Japanese headline text 「{argument name="headline text" default="おはよう"}」 in pink with a thick white outline and soft shadow. Beneath it, add smaller handwritten text 「{argument name="morning message" default="今日もいい一日にしようねっ♪"}」 next to a simple yellow sun doodle and small hearts. In the background, include a sunlit window with greenery outside, warm wooden beams, string lights, and a chalkboard on the back wall with English text “Good Morning!” plus hearts and a cow drawing. Use golden morning sunlight, soft rim lighting, cute VTuber fan-art energy, high detail, clean linework, glossy anime rendering, pastel pink accents, cozy rustic textures, and a cheerful welcoming mood. Keep the composition vertical, close-up, crowded but readable, with no extra characters and no additional text beyond the specified signs and labels.

Create a cinematic anime key visual for a sci-fi fantasy project titled {argument name="title text" default="SHADOW SYNC"}. Wide 16:9 poster composition, ultra-detailed, split-world contrast: the left half is a brilliant utopian future city in daylight with white crystalline skyscrapers, blue sky, sun flare, greenery, waterways, sleek architecture, and golden energy arcs; the right half is a dark cyberpunk megacity at night with storm clouds, neon blue and purple lights, dense towers, holographic signs, flying structures, and rainlike atmosphere. Center foreground: two back-to-back protagonists standing in profile, separated by the divide between worlds. On the left is a blonde young woman with a low ponytail, slim athletic build, wearing a black sleeveless top and dark short outfit, lit by warm daylight; her face is intentionally covered by a soft square beige blur. On the right is a black-haired young man in futuristic black tactical armor, lit by cool neon blue, with a glowing cyan infinity symbol on his shoulder; his face is intentionally covered by a soft square dark blur. Add exactly 2 large flags or banners: one on the bright left side, purple fabric with gold streaks and a glowing blue circular emblem; one on the dark right side, black fabric with a glowing cyan infinity emblem. Add multiple visible infinity symbols across the cyberpunk side, including one large cyan infinity hologram near the center-right, one on the man's shoulder, one on the right flag, one near the top tower display, and one small sign near the lower-right buildings. Include tower text on the dark side reading "REASON", "AGI", "VALUE", "REASON" stacked vertically. Across the lower center, place the main title in large sharp italic cracked lettering, white on the left blending into electric purple on the right, with energy streaks and shards cutting through it. Beneath the title, add small Japanese subtitle text: {argument name="subtitle text" default="影は、未来を選ぶ"}. Visual style: high-end anime illustration, dramatic lighting, extreme detail, glossy sci-fi rendering, dynamic particles, petals and glass shards, purple lightning, blue neon bloom, high contrast, poster-quality composition. No extra logos, no watermark, no additional characters.

Ultra-realistic IMAX-level Netflix-style cinematic six-panel beauty collage portrait, 9:16 vertical composition, use the uploaded image as the primary facial and visual reference with high iden…

Create a cinematic horizontal title card for a classical music video about {argument name="composer and work" default="Reinecke: Undine"}. Use a 16:9 widescreen canvas. The scene is a dark Romantic forest at night, filled with blue-gray mist and water vapor, with a shallow reflective pool along the bottom edge. A powerful diagonal shaft of cold moonlight enters from the upper right and illuminates a ghostly water spirit or undine standing near the lower center, small in scale, with very long pale blonde hair and a translucent silver-blue gown that blends into the fog and water. Frame the image with exactly 9 prominent black tree trunks in the foreground: 5 clustered on the left side and 4 on the right side, with thin hanging branches and moss near the top. The atmosphere should feel mysterious, mythic, and operatic, with deep greens, blacks, cyan-blue haze, and soft volumetric lighting. Add bold title text in the upper left: exactly 2 lines, line 1 reads {argument name="top title text" default="Reinecke:"} in white, smaller bold sans-serif, and line 2 reads {argument name="main title text" default="Undine"} in large bright yellow bold sans-serif with a subtle dark shadow. Keep the typography clean and highly legible, leave the rest of the image free of text, no logos, no watermark.

Cinematic {argument name="lighting" default="sunset"} football match in a massive stadium, score {argument name="match score" default="1-1"} in the final moments, intense crowd cheering, dramatic orange and golden sunset sky, player about to score a {argument name="goal moment" default="last-minute winning goal"}, goalkeeper diving, emotional atmosphere, ultra realistic, cinematic lighting, highly detailed, stadium lights glowing, epic sports photography, 4K

Goal: Create a wide Japanese promotional thumbnail/banner for {argument name="headline topic" default="GPT Image 2 cooking image prompts"}, designed like a cute handmade recipe-craft collage. Canvas: Horizontal 16:9-ish social media banner, warm cream paper background with subtle fabric texture, rounded sticker-like shapes, soft shadows, red and dark teal color palette, high readability. Main headline area: At the top, place a huge bold title reading “GPT Image 2” with “GPT Image” in dark teal and the number “2” in bright red, all with thick white sticker outlines and slight drop shadow. Below it, add a large Japanese headline reading “料理画像プロンプト32例”, with “料理画像” and “32” in red and the remaining text in dark teal, also outlined in white. Under the headline, add a dark teal rounded pill banner with white Japanese text: “写真・参考画像・オリキャラからレシピカードまで”. Mascot and decorations: Add exactly 3 appearances of a cute beige stitched cat mascot with red round glasses and a small red wand: one large mascot standing at the upper right, one tiny mascot near the bottom of the first recipe card, and one large mascot on the right movie-poster card. Add exactly 2 small red yarn-tree decorations near the top left and top right, plus small golden sparkle icons around the headline. Bottom layout: Create exactly 4 rounded rectangular showcase cards in a row across the lower half, separated by exactly 3 red triangular arrow markers pointing right between the cards. Each card has a dark teal label tab with white text at the top. Card 1: Label tab “レシピカード”. Show a clean recipe card layout with a photo of tomato basil pasta on the left and Japanese recipe text on the right. The title should read “トマトとバジルのパスタ”, followed by a small ingredient list with bullet-like lines. Include the tiny mascot at the lower right of this card. Card 2: Label tab “料理写真”. Show a warm professional food photograph of stacked pancakes topped with berries, whipped cream, mint, and syrup, centered in the card. Card 3: Label tab “写真集ページ”. Show an open photo book spread with exactly 2 visible food photos: a salad on the left page and a stew or soup bowl on the right page, with small Japanese captions beneath each photo. Card 4: Label tab “映画ポスター風”. Create a cinematic kitchen-poster style card with large Japanese copy “キッチンから、幸せを。” at the top left, a warm kitchen background, a pot of orange soup at the bottom, and the large cat mascot standing on the right. Visual style: Bright, cute, polished Japanese blog-thumbnail design; realistic food photos inside craft-paper frames; sticker outlines, rounded corners, subtle shadows, soft beige background, clean hierarchy. Keep the composition dense but readable, with no extra cards, no extra arrows, and no watermark.

Draw in a simple hand-drawn doodled style cartoon, as if sketched by a person with a black felt-tip pen. Use natural marker lines, slightly wobbly contours, subtle hand-drawn imperfections, slightly open edges, and a warm editorial doodle feeling. Avoid polished vector-clean results, avoid overly crisp digital perfection, and keep the image simple, airy, flat, and understated. Use {argument name="accent color" default="sky blue"} only as the accent color.

A journey looking down at the {argument name="landmark" default="Flame Towers"}. {argument name="city" default="Baku, the capital of Azerbaijan"}. A {argument name="flock" default="flock"} slowly crosses over the sea breeze. A single {argument name="hidden character" default="white goose"} mixed in the flock.

Create a {argument name="aspect" default="vertical"} cinematic tutorial comparison poster showing the same {argument name="subject" default="urban street scene"} repeated in five horizontal panels, each panel demonstrating a different film genre color grading style. The scene is a busy small city street at sunset, with cars, motorcycles, cyclists, shop buildings, electric poles, tangled overhead wires, and warm evening sky. Keep the camera angle exactly the same in every panel, like a fixed street photography shot. Each horizontal strip should have a different cinematic mood and color tone: Panel 1: SUSPENSE - cool blue shadows, dark contrast, slightly mysterious atmosphere, tense and moody lighting. Panel 2: ACTION - warm orange sunset tone, high contrast, energetic cinematic look, dramatic highlights. Panel 3: CRIME - cold teal blue grading, darker exposure, gritty urban feeling, realistic shadows. Panel 4: ROMANCE - soft golden warm tone, gentle contrast, nostalgic sunset mood, dreamy atmosphere. Panel 5: DRAMA - muted purple blue evening tone, emotional cinematic lighting, soft shadows and serious mood. Add bold white uppercase text on the right side of each panel: SUSPENSE, ACTION, CRIME, ROMANCE, DRAMA. Use clean modern sans serif typography with slight glow or subtle shadow. Separate each panel with thin white horizontal borders. Make it look like a professional filmmaking color grading tutorial thumbnail, cinematic, realistic, high detail, urban street photography, 9:16 vertical composition. Negative Prompt: cartoon, anime, painting, low quality, blurry, distorted buildings, extra text, wrong spelling, messy typography, overexposed, unrealistic colors, duplicated cars, warped street, bad perspective, random objects, low resolution, watermark

Goal: Create a dramatic live television news still of an anime typhoon report in Kobe, Japan, combining a realistic stormy waterfront background with a cute anime reporter composited in the foreground. Canvas: 16:9 horizontal broadcast frame, photorealistic rainy seaside city scene at dusk or early evening, grey-blue storm lighting, wet reflective pavement, rough ocean waves, heavy rain streaks, strong wind, dark clouds, distant skyline and harbor towers. Main subject: A cute blonde anime girl news reporter stands slightly right of center on a waterfront promenade beside a black metal railing. She has very long blonde hair blown violently to the left by the wind, pink flower hair accessories, a determined reporting pose, and one arm extended toward the stormy sea. She holds a black handheld microphone in her other hand. Her outfit is a pink idol-style ensemble: frilly pink jacket, bow details, pale blouse, pink plaid pleated skirt, ruffled trim, ribbons, and thigh accessory. Her face is intentionally covered by a centered square blur/censor block in a muted beige color. Environment details: The promenade is soaked and reflective, lined with street lamps and wind-bent palm trees. Show exactly 5 prominent palm trees leaning in the wind along the waterfront. The ocean on the right is churning with whitecaps and spray. The city skyline appears through rain and mist, with high-rise buildings and a lit tower in the distance. Broadcast graphics: Include exactly 4 on-screen TV overlay groups. 1) Top-left time/location stack: large white time {argument name="time display" default="7:42"}, smaller Japanese location text {argument name="location text" default="兵庫県・神戸市"}, and a white live-relay label {argument name="relay label" default="中継"}; add a short blue underline beneath the location. 2) Top-right typhoon alert banner: red-and-white Japanese breaking-news panel with the labels {argument name="typhoon headline" default="台風 情報 関西に最接近 暴風・高波に厳重警戒"} plus a red lower strip reading {argument name="broadcast place label" default="中継 兵庫県・神戸市"}. 3) Bottom-left weather radar box: a rectangular blue-framed rain-cloud radar map labeled 「雨雲レーダー」, with a vertical color scale on the right and small Japanese intensity labels 「強」「弱」 and 「現在」. 4) Bottom-right lower third: a blue slanted LIVE tag, white location bar reading 「神戸市中央区沿岸部」, and grey reporter credit bar reading 「記者：星川 さくら」. Visual style: Realistic disaster-news photography for the background, high-quality anime character rendering for the reporter, cohesive broadcast composition, rain and motion effects, dramatic wind-blown hair and clothing, sharp TV news graphics, no watermark, no extra logos, no extra overlay groups.