GPT Image 2 prompts

Browse GPT Image 2 prompts for fashion portraits and graphic compositions. Compare how each prompt describes its subject and the way the image is arranged.

Explore prompts

10,026 prompts

Product Marketing - Avant-Garde Sportswear Editorial — output

Product Marketing - Avant-Garde Sportswear Editorial

PromptLabGPT Image 2
Avant-garde sportswear editorial ad, an oversized {argument name="prop" default="basketball"} treated as a sculptural monument, a female athlete lying back across its curved surface as though draped on designer furniture, the word "{argument name="headline" default="ELEVATE"}" in massive bold type filling the backdrop, rich {argument name="background color" default="burnt-orange"} studio background, high-gloss reflective floor, luxury high-end athletic editorial aesthetic, dramatic cinematic lighting, ultra-clean minimal composition, 1:1
Social Media Post - Tokyo Sidewalk Motion Blur Snapshot — output

Social Media Post - Tokyo Sidewalk Motion Blur Snapshot

WasifAIGPT Image 2
Ultra-realistic candid smartphone photo, 3:4 vertical. Original non-identifiable young woman inspired only by general proportions/vibe. {argument name="location" default="Night Tokyo sidewalk"} beside {argument name="background" default="white ceramic tile wall"}. Mid-walk, half-profile, turning back in surprise, {argument name="action" default="hand raised toward lens"}, shy restrained smile, continuing forward, imperfect framing, nearly exiting frame. Flash fired during motion. Strong handheld shake, heavy directional blur, smeared facial details, blurred hand and hair, ghosting, streaked tile wall, deep darkness around. Harsh flash highlights, blown exposure, hard shadows. Raw accidental snapshot, intrusive, playful, fleeting memory. Extremely noisy, heavily blurred, authentic phone-camera look.
Profile / Avatar - 提示词: — output

Profile / Avatar - 提示词:

sereinGPT Image 2
使用最后一张人物照片作为身份参考,保留人物五官、脸型、眼神、发型和整体辨识度,不要生成陌生人脸。保持每张原图的动作、姿势、镜头角度、构图、服装逻辑和表情一致。
Social Media Post - Cherry Blossom Portrait — output

Social Media Post - Cherry Blossom Portrait

LipstickGPT Image 2
Create a romantic spring portrait photograph of {argument name="character description" default="a young woman with long auburn-brown hair"} sitting sideways on a thick, curved cherry blossom tree branch in a blooming orchard. She wears {argument name="outfit" default="a flowing white ankle-length traditional dress with blue floral embroidery, patterned borders, and a matching draped scarf"}, with beige embroidered flats visible beneath the hem. Her face is softly anonymized with a natural skin-toned blur while the pose remains elegant and serene. She holds {argument name="bouquet" default="a bright bouquet of red flowers with green stems"} in her lap with both hands. Surround her with dense pink cherry blossoms on dark textured branches, using a shallow depth of field so the background becomes warm, dreamy bokeh with hints of green foliage and golden sunlight. Compose the image vertically but framed as a 4:3 portrait crop, subject centered slightly left, full seated figure visible, branch forming a natural diagonal cradle across the frame. Use realistic photography, soft natural light, gentle pastel colors, detailed fabric embroidery, crisp foreground blossoms, creamy blurred background, no text, no watermark, no extra people.
Profile / Avatar - Naive Digital Selfie Avatar — output

Profile / Avatar - Naive Digital Selfie Avatar

Eric Kang - Vogue AIGPT Image 2
Create a naive digital selfie avatar illustration of {argument name="character name" default="a young person"} from the shoulders up, centered and facing the viewer with a simple direct smile. Use a square 1:1 canvas with a plain pastel sky-blue background. The character has {argument name="hair color" default="messy black"} medium-length shaggy hair peeking out from under a {argument name="hat color" default="green"} backward baseball cap with a small metal ring detail on the left side. Dress them in a blue denim collared zip shirt layered under a beige-and-brown checkered jacket. Render in a playful naive digital painting style with rough crayon-like texture, visible brush strokes, soft flat colors, simplified anatomy, minimal facial detail, warm peach skin tones, and an everyday casual selfie composition. Keep the crop close, with the head large in frame and the shoulders filling the lower third. Avoid photorealism, hard outlines, text, logos, or complex background details.
Social Media Post - Romantic Garden Portrait with Cat — output

Product Marketing - Romantic Garden Portrait with Cat

Mehwish kiranGPT Image 2
Create a beautiful, high-quality professional portrait of a {argument name="subject" default="young South Asian woman"} seated gracefully on an ornate white carved stone garden bench in a lush, sun-drenched rose garden. She has long dark wavy hair parted slightly to one side, soft elegant styling, and wears traditional silver jhumka earrings. Her outfit is a flowing {argument name="dress color and style" default="pastel blush-pink floral A-line dress with sheer long sleeves"}, with delicate pink rose prints, a modest round neckline, gathered waist, and airy romantic fabric draping over the bench. A fluffy {argument name="pet" default="white Persian cat"} rests calmly on her lap while she gently places one hand on it. The setting includes exactly one ornate bench, one central tiered white fountain softly blurred in the background, one pale column behind the garden, clusters of bright pink roses on the right and left, and lavender flowers blurred in the foreground. Use warm golden-hour sunlight, dreamy shallow depth of field, creamy bokeh, soft backlighting, natural skin tones, realistic fabric texture, and an elegant editorial fashion photography look. Compose as a vertical 2:3 medium-full portrait, subject centered slightly right, seated posture, serene luxury garden atmosphere. Keep the face natural and unobstructed unless a privacy blur is requested; avoid extra people, text, logos, distorted hands, or unnatural anatomy.
Profile / Avatar - Formal Business Headshot Profile Poster — output

Profile / Avatar - Formal Business Headshot Profile Poster

Eric Kang - Vogue AIGPT Image 2
Goal: Create a formal business headshot profile poster for a fictional adult design professional, combining a clean studio portrait with an elegant nameplate area.

Canvas: Vertical 4:5 poster, approximately 768 x 960 px, crisp high-resolution corporate profile design.

Portrait area: Upper 70% of the image shows a centered waist-up studio headshot of an adult woman with {argument name="hair color" default="dark brown"} shoulder-length softly waved hair, wearing a dark navy tailored blazer over a white open-collar dress shirt. She has a professional upright pose with arms subtly crossed or held close, photographed against a pure white background with soft even studio lighting. Add a large centered rectangular anonymization block over the face, using a smooth skin-tone/brown gradient, covering the facial features from forehead to chin while leaving hair, ears, neck, and clothing visible.

Lower profile panel: Bottom 30% is a dark near-black navy rectangular information panel spanning the full width. Add a thin metallic gold line along the top edge of the panel, with a small angular gold accent notch near the upper-right edge. The design should feel premium, minimalist, and editorial.

Text content: Place exactly 3 text elements on the lower panel, aligned left with generous margins: 1) large white serif name text: {argument name="character name" default="Yuna Koizumi"}; 2) smaller gold sans-serif job title: {argument name="job title" default="Senior Graphic Designer"}; 3) smaller light gray sans-serif organization text: {argument name="organization" default="International Design Institute"}. Add exactly 1 short horizontal gold divider line beneath the name, aligned with the left text column.

Visual style: Professional LinkedIn-style executive portrait mixed with luxury editorial typography, realistic photography, clean composition, high contrast navy-white-gold palette, refined spacing, no clutter.

Constraints: Use exactly one portrait subject, exactly one face-obscuring rectangle, exactly one lower information panel, exactly one gold divider line under the name, and exactly three text lines. Do not add logos, icons, watermarks, extra captions, background patterns, or additional people.
Product Marketing - Cherry Blossom Couture Portrait — output

Product Marketing - Cherry Blossom Couture Portrait

NIKKOGPT Image 2
Create a vertical 2:3 high-fashion fantasy portrait of {argument name="character name" default="a young woman"} kneeling in an elaborate cherry-blossom installation, with a deliberately featureless face covered by a centered matte dusty-pink rectangular mask. She has very long, tousled {argument name="hair color" default="pale platinum-blonde hair"} and fair glowing skin, posed with one hand touching layered black necklaces at her chest and the other hand resting on the floor. Dress her in an ornate off-shoulder {argument name="dress color" default="pink and ivory"} couture gown with lace bodice, ruffled sleeves, oversized ribbon bows on both arms, sheer lace gloves or sleeves, black ribbon wrapping accents, dangling black bead details, and a voluminous skirt pooling around her knees. Surround her with a dense romantic background of cherry blossoms, petals, and circular hoop structures resembling mizuhiki cords: count exactly 14 visible circular hoops and disks, including 9 thin metallic white or rose-gold rings, 3 large pink floral disks, and 2 dark black floral disks. Add scattered pink petals on the floor, soft foreground blossom blur at the bottom edge, and a dreamy pastel palette of blush pink, cream, ivory, rose gold, and soft gray. Lighting should be bright, ethereal, backlit, and diffused, with shallow depth of field, glossy skin highlights, delicate fabric texture, cinematic editorial composition, ultra-detailed photorealistic rendering, no text, no watermark, no extra people.
Profile / Avatar - Photo to Sunlit Anime Character — output

Profile / Avatar - Photo to Sunlit Anime Character

AIAI.COMGPT Image 2
Using REFERENCE_0 as the base, transform the photo into a polished anime-style illustration while preserving the same overall pose, camera angle, white oversized shirt, shark plush being hugged, window-side indoor setting, and warm backlit mood. Replace the obscured face with a cute expressive anime face: soft blush, slightly shy eyes, and the hand still raised near the mouth. Keep the short black bob silhouette but make it glossy and stylized, with red-tinted hair tips and 3 visible hair accessories: 2 silver rectangular hair clips and 1 gray X-shaped pin. Render the scene with clean line art, detailed painterly shading, luminous sunlight, soft highlights on fabric, and a high-quality visual-novel/anime key art finish.

Add and clarify the room details that were only vague in the reference: a teal-green wall, a framed ocean-and-cloud picture on the wall, a wooden dresser on the right, a white vase with exactly 3 yellow flowers, and 1 small framed photo on the dresser. Add a few subtle floating dust motes/bubbles in the sunlight. Make the shark plush more cute and crisp, with a blue top, white belly, black eye, gill marks, fin, tail, and open mouth with triangular teeth.

Customizable details: anime character with {argument name="eye color" default="red"}, {argument name="hair accent color" default="red"}, wearing {argument name="shirt style" default="oversized white button-up shirt"}, holding {argument name="plush toy" default="blue shark plush"}, in a {argument name="background mood" default="warm sunlit bedroom by a window"}.

Constraints: do not crop out the plush, keep the composition vertical, keep the pose and outfit close to the reference, avoid photorealism, avoid extra characters, avoid text or watermarks.
Social Media Post - Kawaii Birthday Group Selfie — output

Social Media Post - Kawaii Birthday Group Selfie

小小东GPT Image 2
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.
YouTube Thumbnail - Japanese Developer Comment Banner — output

YouTube Thumbnail - Japanese Developer Comment Banner

kenta@AIVTuberテセウス•ノ•ハイネ開発者GPT Image 2
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.
Comic / Storyboard - Ancient Hanfu Sketch Comes Alive — output

Comic / Storyboard - Ancient Hanfu Sketch Comes Alive

Larus CanusGPT Image 2
Goal: Create a 4:5 vertical image titled {argument name="artwork concept" default="Paper Becomes Real · Ancient-Style Costume Draft"}, showing an elegant ancient Chinese clothing design sketch awakening into reality on paper.

Canvas and camera: Overhead view with a slight diagonal perspective, warm natural studio lighting, photorealistic details mixed with pencil fashion-sketch illustration. A large off-white garment-design sheet is clipped to a wooden drawing board on a warm brown wooden worktable.

Main subject: One adult Chinese ancient-style beauty lies diagonally across the design paper as if emerging from the drawing. Her upper body is realistic and three-dimensional, while the lower skirt gradually transitions into pale watercolor fabric and pencil sketch lines. Her face is intentionally covered by a soft square mosaic blur. She has long black hair styled half-up with delicate silver floral hairpins, pearl tassels, and a small pale green ornament. Outfit: {argument name="costume palette" default="moon-white sheer gauze with pale mist-blue layers"}; translucent Hanfu-style robe, soft white inner wrap, pale blue embroidered collar and sash, layered gauze sleeves, flowing ribbon belt, pearl tassel accessory at the waist, and a long skirt fading into sketchy lines near the hem.

Layout and visible design labels: Include exactly 5 handwritten annotation groups on the paper: 1) top right title “LOOK 01” underlined; 2) near the head, “hair ornament” with Chinese-style notes and a small floral hairpin sketch; 3) left side, “sleeve volume” with a sleeve silhouette sketch; 4) right side, “ribbon flow” with arrows and ribbon movement lines; 5) lower left, “skirt movement” with arrows and skirt-flow sketch lines. Use neat pencil handwriting with small bilingual-looking notes, but keep the English labels readable.

Surrounding objects: Place exactly 7 studio props around the board: 1) a pale ceramic teacup at the upper left; 2) small paper color swatches at the left edge; 3) a pale jade pendant with tassel near the lower left; 4) two fabric swatches in white and mist blue at the bottom left; 5) a calligraphy paperweight or carved wooden block with a vertical calligraphy strip at the upper right; 6) two drawing pens on the right side; 7) one artist hand entering from the lower right holding a pencil and actively sketching the skirt hem.

Visual style: High-end ancient Chinese fashion portrait plus costume design manuscript, delicate fabric transparency, realistic skin and hands, graphite pencil construction lines, pale blue-gray watercolor washes, soft shadows, tactile paper texture, cinematic but quiet atelier atmosphere. Emphasize the illusion that a 2D/3D costume sketch is breaking through the page into real life.

Constraints: No extra characters, no modern clothing on the main subject, no clutter beyond the listed props, preserve the 4:5 vertical composition, keep the annotations subtle and hand-drawn, avoid logos or watermarks.
Social Media Post - Sunshine Day Travel Sticker Photo — output

Social Media Post - Sunshine Day Travel Sticker Photo

小小东GPT Image 2
Create a bright youthful travel-photo sticker aesthetic image: a candid outdoor lifestyle shot of {argument name="subject" default="a young woman traveler"} standing in a sunlit leafy cafe street, wearing {argument name="outfit" default="a white T-shirt, beige pants, and a large beige hiking backpack"}, holding a small black camera, with long dark hair and a softly blurred/covered face for privacy. The setting is a warm daytime sidewalk cafe with trees, dappled sunlight, wooden tables, a shop window, and an A-frame coffee sign in the background, shot like a natural wide-angle travel photograph with shallow depth of field and fresh summer colors. Overlay playful hand-drawn sticker graphics and handwritten typography in a cute Gen-Z scrapbook style, using white marker outlines, pastel pink, sky blue, yellow, and little hearts. Include exactly 18 visible doodle/sticker elements: 1 large headline reading {argument name="headline text" default="Sunshine Day"} in yellow bubble letters over a blue paint swash at upper left; 2 small pink hearts near the headline; 3 the handwritten word {argument name="small caption" default="happy"} with a smile at top center; 4 three white steam squiggles near the subject; 5 a pink camera icon above the subject; 6 a large white wavy outline surrounding the subject; 7 the handwritten phrase "Let's go!" at upper right with underline; 8 a pink scribbled heart at top right; 9 the word "Nice!" at left with yellow underline; 10 three pink exclamation strokes next to "Nice!"; 11 a white arrow curl pointing toward the subject; 12 a white cloud outline at lower left; 13 a pink brush stroke at bottom left; 14 the phrase "Enjoy every moment" at mid-right with a curved underline; 15 the word {argument name="bottom sticker text" default="Chill"} in blue bubble letters at lower right; 16 two small blue cloud puffs near "Chill"; 17 three tiny hearts around "Chill" in blue/yellow/white; 18 a pink fruit-like heart sticker at bottom right. Keep the doodles integrated over the photo like editable social-media stickers, energetic and optimistic, no extra text, no watermark.
Social Media Post - Cozy Bedtime Bedroom Portrait — output

Social Media Post - Cozy Bedtime Bedroom Portrait

ゆーじ/AI生成見習いGPT Image 2
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.
Social Media Post - Xiamen Bougainvillea Morning Poster — output

Social Media Post - Xiamen Bougainvillea Morning Poster

小小东GPT Image 2
Goal: Create a refined morning greeting poster for Xiamen featuring bright pink bougainvillea flowers, soft city-flower aesthetics, and elegant Chinese typography.

Canvas: Horizontal 16:9 poster, 1200×675 style composition, airy high-key lighting, pastel white-blue-pink background, soft bokeh, clean editorial design.

Main image: Use a close-up macro photograph style of vivid pink bougainvillea branches across the center and right side, with translucent veined petals, tiny cream-white blossoms, dew drops on petals, green leaves, sunlit fresh morning atmosphere, shallow depth of field, and a gentle sky-blue background blur. Include exactly 3 prominent bougainvillea flower clusters: one large cluster centered, one large cropped cluster on the right edge, and one softer blurred cluster in the rear right.

Typography and layout: Add a very large, thin, pale pink handwritten English phrase “GOOD MORNING” stretched across the top as a subtle background overlay. On the upper left place a small greeting block with Chinese text: “早安,厦门” and below it “海风轻拂,花色正好”, with a small pink flower doodle and thin underline. Along the far left edge add vertical micro text “XIAMEN | AMOY” with a tiny pink map-pin or heart dot between the words.

Bottom information card: Place one large frosted-glass rounded rectangle card across the lower left and middle, with a pale blue rounded decorative block behind it. The card contains the main title “厦门早安” in large black Chinese serif calligraphy style, with spaced romanization below: “XIAMEN ZAO AN”. Under that show the date line “{argument name="date" default="2026.06.25"} / 星期四”. Add a small pink circular accent near the title/date.

Right side of bottom card: Include a vertical divider and exactly 3 information rows with simple pastel icons. Row 1 has a small rainy weather icon and text: “今日天气 | 小雨转多云 28-32°C” plus smaller text “午后有雷阵雨,体感闷热”. Row 2 has a pink flower outline icon and text: “城市花意 | 三角梅 Bougainvillea”. Row 3 has a light green T-shirt icon and text: “穿衣建议 | 轻薄短袖 + 透气下装” plus smaller text “随身带伞,注意防晒,午后防雨 空调房可备薄外搭”.

Top-right badge: Add one vertical rounded pink translucent label card with a small flower icon at top, Chinese text “城市花意”, larger “三角梅”, and cursive English “Bougainvillea”.

Style: Modern Chinese lifestyle poster, soft feminine botanical aesthetic, premium magazine layout, translucent glassmorphism cards, delicate line icons, pastel gradients, crisp readable text, elegant spacing, no clutter. Use {argument name="city name" default="厦门"} as the city theme, {argument name="main flower" default="三角梅"} as the city flower, {argument name="weather summary" default="小雨转多云 28-32°C"} for the forecast, and {argument name="morning message" default="海风轻拂,花色正好"} for the greeting subtitle.

Constraints: Keep exactly 2 rounded translucent information cards total: the large bottom card and the small top-right badge. Keep exactly 3 information rows in the bottom card. Preserve the bright pink bougainvillea as the dominant visual subject, maintain legible Chinese text, avoid people, avoid logos, avoid watermarks.
Infographic / Edu Visual - Claude Code Breaking News Infographic — output

Infographic / Edu Visual - Claude Code Breaking News Infographic

甲斐甲GPT Image 2
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.
Social Media Post - Night Market Dinner Portrait — output

Social Media Post - Night Market Dinner Portrait

HarshithGPT Image 2
Create a realistic candid nighttime food-market photo of a young woman seated at a table under a dark tent canopy with metal poles and bright overhead lights, shallow depth of field and a busy blurred crowd behind her. She has long wavy brown hair and wears a vivid royal-blue hoodie, with her face intentionally anonymized by one large vertical opaque muted taupe rectangle centered over the face; also include two smaller blurred face blocks on background people. She is holding one smartphone in her left hand and one clear plastic cup of water in her right hand. In the foreground on a red-and-white checkered plastic tablecloth, show exactly two white disposable paper plates: the left plate contains sliced grilled glazed chicken with dark sauce, and the right plate contains white rice topped with one sunny-side-up egg with a bright yellow yolk. Include exactly one clear plastic cup near her right hand, one visible phone, two delicate necklaces with small round pendants, and casual market diners in the background. Use direct flash or bright market lighting, natural colors, documentary travel-photo realism, vertical 4:3 composition, eye-level framing, slight background bokeh, no added text, no watermark. Customizable details: {argument name="hoodie color" default="royal blue"}, {argument name="tablecloth pattern" default="red-and-white checkered"}, {argument name="main dish" default="sliced grilled glazed chicken"}, {argument name="side dish" default="white rice topped with one sunny-side-up egg"}, {argument name="face anonymization color" default="muted taupe"}.
Infographic / Edu Visual - Exploded Tennis Racquet Blueprint — output

Infographic / Edu Visual - Exploded Tennis Racquet Blueprint

RTGPT Image 2
Goal: Create a clean technical blueprint-style exploded diagram of {argument name="sports equipment" default="a tennis racquet and ball"}, labeled like an instruction manual figure.

Canvas: Portrait 3:4 composition on an off-white paper background with lots of negative space. Use thin cobalt-blue line art, monoline drafting strokes, pale blue translucent fills only where needed, and subtle print texture. No shadows except very light flat blue fill on the ball core.

Layout: Place the exploded tennis racquet vertically on the left two-thirds of the canvas, with parts separated along a central vertical axis. Place the exploded tennis ball cross-section on the right-middle. Add dashed vertical guide lines framing each exploded assembly and dashed horizontal callout lines leading to labels on the right of each part. Use small all-caps monospaced blueprint lettering.

Header and footer text: Put "FIG.002" in the top-left corner. Put "[ RACQUET AND BALL ]" in the top-right corner. Put "COPYRIGHT 2026" in the bottom-right corner.

Discrete elements: Show exactly 13 main visible components total: 7 racquet components and 6 ball/callout-related components. The 7 racquet components, from top to bottom, are: 1) FRAME, an oval racquet head rim with small evenly spaced holes around the inner edge and a partial yoke connection; 2) STRING BED, a separate oval mesh insert made of dense diagonal crisscross strings with tiny perimeter nodes; 3) THROAT, a forked triangular bridge piece with open cutouts; 4) SHAFT, a short cylindrical handle connector with a rectangular inset detail and a circular end opening; 5) GRIP, a longer wrapped cylindrical grip with spiral bands; 6) BUTT CAP, a small hexagonal end cap below the grip; 7) dashed alignment box and callout construction lines surrounding the racquet assembly. The 6 ball/callout-related components are: 1) top white curved FELT shell piece; 2) central blue spherical CORE with a circular ring contour on top; 3) bottom white curved FELT shell piece; 4) dashed vertical guide lines around the ball assembly; 5) dashed horizontal label line to FELT; 6) dashed horizontal label line to CORE.

Labels: Add exactly 7 callout labels in blue uppercase text: FRAME, STRING BED, THROAT, SHAFT, GRIP, BUTT CAP, FELT, and CORE. Place racquet labels to the right of the racquet parts and ball labels to the right of the ball parts. Keep labels aligned and minimal.

Visual style: Minimal futuristic technical illustration, exploded product schematic, precise vector-like geometry, thin blue outlines, light blue hatching/mesh for the string bed, no photorealism, no people, no court, no extra decorations.

Constraints: Keep the racquet head slightly tilted in perspective at the top, the string bed floating below it, and all handle parts stacked downward with visible gaps. The ball should appear as a simple cutaway with two separated felt shell halves and one blue core. Use {argument name="line color" default="cobalt blue"} linework on {argument name="background color" default="warm off-white"} paper, with {argument name="figure number" default="FIG.002"} and {argument name="copyright text" default="COPYRIGHT 2026"}.
Social Media Post - Miniature Girl in Giant Palm — output

Social Media Post - Miniature Girl in Giant Palm

Alina AiGPT Image 2
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.
Profile / Avatar - Yokai Streetwear Identity Portrait — output

Profile / Avatar - Yokai Streetwear Identity Portrait

Cia0GPT Image 2
Create a vertical full-body anime streetwear identity portrait of {argument name="character name" default="a mysterious young woman"} standing centered and facing forward, with a deliberately soft blurred rectangular face for anonymity. She has very long straight {argument name="hair color" default="black hair with deep teal highlights"} falling past her chest, and a large flat red circular halo/disc directly behind her head. Outfit: oversized black long-sleeve sweatshirt with a red square chest patch containing bold black Japanese calligraphy, loose cropped floral-pattern pants in muted green with red and beige flowers, black socks, and chunky white sneakers. Her hands hang together in front, holding one orange-red oni demon mask by its black horns; the handheld mask has angry eyes, black brows, tusks, and a grimacing mouth. Background: a dense Japanese yokai/oni mask mural in a hand-painted graphic style, using muted red, cream, teal-green, black, and tan. Include exactly eight large visible background mask/face motifs: 1 teal-green snarling mask at upper right, 2 cream cyclops mask at middle left, 3 pale rabbit-like figure with red ring at lower left, 4 cream square grinning face at left lower middle, 5 gray demon head at bottom left, 6 red skull-like panel face at middle right, 7 huge red oni face at bottom right, and 8 a crowded top strip of small red and cream grotesque masks. Add a horizontal cream sign with red Japanese-style characters behind the subject on the right, and rough black brush marks on red panels on the left. Visual style: moody editorial fashion illustration mixed with cel-shaded anime, thick ink outlines, painterly texture, slightly distressed poster surface, high detail, subdued vintage colors, symmetrical composition, low-angle full-body framing, no photorealism, no extra people, no watermark. Use {argument name="canvas ratio" default="9:16 vertical"} composition and a gritty Japanese folklore streetwear aesthetic.
Social Media Post - Dreamy Chibi Mirror Selfie Room — output

Social Media Post - Dreamy Chibi Mirror Selfie Room

Alina AiGPT Image 2
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.
Comic / Storyboard - Magical Girl Transformation Storyboard — output

Comic / Storyboard - Magical Girl Transformation Storyboard

MikisunGPT Image 2
Using the provided reference image as the character base, create a horizontal magical-girl transformation storyboard for a {argument name="video length" default="15 seconds"} sequence. Analyze and preserve the character’s face, hairstyle, hair color, eyes, body type, outfit details, accessories, motifs, and overall pink magical world aesthetic from the reference while showing the transformation as a cinematic sequence.

Format: Wide storyboard sheet with exactly 10 panels arranged in a 2-row by 5-column grid. Each panel must have a small storyboard table layout with Japanese labels, including Scene number, time range, split image, scene description, camera work, and video generation prompt. Use a clean pink-bordered production storyboard style.

Panel sequence and timing: 1: 0.0–1.5s, the character stands before transforming on a glowing magic circle in a dreamy pink sky. 2: 1.5–3.0s, close-up of feet as magic circles and light particles rise. 3: 3.0–4.5s, shoes, socks, and leg decorations appear in sparkling light. 4: 4.5–6.0s, skirt forms as ribbons and luminous fabric wrap around the waist. 5: 6.0–7.5s, upper-body outfit forms with blouse, bow, and shoulder details. 6: 7.5–9.0s, arm accessories, sleeves, gloves, and wrist decorations appear. 7: 9.0–10.5s, close-up of the face and upper body as hair ornaments, ribbons, ears/head accessories, and a small crown-like ornament sparkle into place. 8: 10.5–12.0s, the character receives or summons a signature magical item/weapon, with face partially obscured by bright transformation light if needed. 9: 12.0–13.5s, full-body reveal with hair, skirt, and ribbons swirling dramatically in radiant pink light. 10: 13.5–15.0s, final heroic pose with a large magic circle behind her and cute mascot-like elements around the scene.

Style: pastel pink anime magical-girl aesthetic, soft clouds, hearts, sparkles, glowing runes, ribbons, luminous particles, dreamy fantasy lighting, delicate line art, high saturation pinks and whites. Keep the storyboard readable and production-oriented, with Japanese text blocks in each panel. Add a title-free grid only; no watermark.

Customization: The transformation theme should be {argument name="transformation theme" default="pink heart-and-ribbon magical girl"}, and the final magical item should be {argument name="magical item" default="a cute pink wand-like weapon"}.
Profile / Avatar - Anime Ninja Girl on Rooftop — output

Profile / Avatar - Anime Ninja Girl on Rooftop

AIAI.COMGPT Image 2
Create a polished anime-style illustration of a young female ninja sitting on a sloped traditional Japanese tiled rooftop at dusk, shown in a vertical 2:3 composition. She has {argument name="hair color" default="bright orange coral"} hair in two high twin ponytails tied with scrunchies, long bangs framing the face, and a playful pose with one hand raised making a peace sign; add a small yellow sparkle beside the gesture. Her face is intentionally hidden by a centered square censor block with a soft pink-to-gray gradient, covering the eyes, nose, and mouth completely. She wears an orange sleeveless kunoichi outfit with a deep wrap-style neckline, black trim, black belt and hanging ties, visible fishnet mesh side panels, black tight shorts under a short orange skirt panel, black fingerless gloves with wrist guards, black shin wraps, and open-toe black ninja sandals with red ties. The camera is close and slightly above, emphasizing her seated pose with bent legs in the foreground and one arm braced behind her on the roof. Background: dark blue rooftops with ceramic tiles, shadowy trees and chimney silhouettes, and a clear teal-blue evening sky. Use clean crisp linework, cel-shaded anime rendering, glossy skin highlights, saturated orange accents against cool blue shadows, cinematic lighting, high detail, no text, no watermark, no extra characters.
Social Media Post - Lo-fi Editorial Bedroom Portrait — output

Social Media Post - Lo-fi Editorial Bedroom Portrait

PromptLabGPT Image 2
An intimate bedroom portrait of a {argument name="subject" default="young man in his early 20s"}, calm and slightly guarded expression, direct eye contact, standing relaxed. He wears a {argument name="clothing" default="textured knit sweater with a soft purple gradient fade"} and muted casual trousers. The background is a personal room wall covered in taped polaroids, handwritten sketches, and concert ticket stubs. Shot at eye level on a 35mm lens with harsh on-camera flash, hard shadows behind subject, deep depth of field, slight vignetting and flash falloff. Real skin texture, visible blemishes, no retouching, noticeable film grain, subtle date stamp in corner. Muted neutral palette with soft lavender emphasis, natural warm skin tones, moderate contrast. Lo-fi editorial realism, late 90s / early 2000s candid diary aesthetic. No studio lighting, no beauty retouching, no clean minimal background.
Browse by model, style or topic

Find your next direction

ASTORIE CANVAS

Take the idea further.

Bring your prompts, references and results together in one creative space.

Open your canvas ↗