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

10,044 prompts

Social Media Post - Suzhou Garden Bamboo Path Portrait — output

Social Media Post - Suzhou Garden Bamboo Path Portrait

ToroJushiAiGPT Image 2
Authentic daily life photography, natural candid moment. A full-length portrait of a {argument name="subject" default="young woman"} standing still in the middle of a narrow path cutting through a dense stand of {argument name="plant" default="moso bamboo"} in a classical {argument name="location" default="Suzhou garden"}. The bamboo culms rise vertically on both sides
Profile / Avatar - Cute Eco Robot Pixel Art Mascot — output

Profile / Avatar - Cute Eco Robot Pixel Art Mascot

JustinGPT Image 2
Create a centered full-body pixel art mascot robot on a clean {argument name="background" default="plain white background"}. The character is a cute chibi eco robot with a cream-colored rounded helmet-like head and body, dark black face screen, and green plant-themed accents in a retro 16-bit/32-bit pixel style with visible square pixels, crisp black outline, and soft shaded highlights. The robot has exactly 2 large glossy green eyes with white heart-shaped highlights, a tiny white smiling mouth, exactly 2 round green side ear pods, exactly 2 green sprout leaves growing from the top antenna, and a small green forehead panel decorated with a white sprout icon plus 2 white dots. The body is compact and rounded with a square green chest panel showing a white sprout emblem, exactly 2 arms with green joints, the left arm raised in a friendly wave with a small green segmented hand, the right arm hanging down with a chunkier cream forearm and green hand, exactly 2 short legs, and exactly 2 cream-and-green boots. Use a warm palette of {argument name="main body color" default="soft ivory cream"}, {argument name="accent color" default="leaf green"}, deep black for the face visor, and subtle yellow-green highlights. Make the mascot adorable, symmetrical, front-facing, isolated, no text, no watermark, no extra props, high-resolution pixel art with a slight grid-like pixel texture.
Social Media Post - Claymation Social Banner — output

Social Media Post - Claymation Social Banner

自学Ai的鱼头叔叔GPT Image 2
Create a wide 2:1 social media banner in a handcrafted claymation stop-motion style. Canvas size about 984×440 px, with a warm pastel yellow clay background and soft cinematic shadows. Center the composition around one huge raised clay fist emerging from behind a dark charcoal rounded rectangle strip across the bottom third. Behind the fist place one large red folded ribbon/banner shape spreading left and right, and one blue spiky burst shape directly behind the fist. At the top, add one oversized bold white 3D clay headline reading {argument name="headline text" default="懦夫救世!"}, with rounded puffy edges, slight red shadowing, and one red clay star used as the dot/accent near the final exclamation area. Add exactly one circular profile avatar at the lower left overlapping the dark bottom strip: a cute beige clay character with large googly eyes holding a small microphone, framed by a dark ring. Add exactly one rounded black button at the lower right with a thin light outline and small white Chinese text reading {argument name="button text" default="编辑个人资料"}. Use soft polymer-clay texture everywhere, visible fingerprints, rounded handmade edges, macaroon warm colors, shallow depth of field, and realistic stop-motion lighting. Keep the layout clean and graphic, with exactly 5 main visible components: the white headline, the red ribbon, the central fist, the left circular avatar, and the right edit-profile button. No photorealistic skin, no flat vector look, no extra characters, no watermark.
Social Media Post - Pearlescent Erlang Shen Bust — output

Social Media Post - Pearlescent Erlang Shen Bust

TWneseGPT Image 2
Using REFERENCE_0 as the visual style and composition reference, transform the pearlescent bead-and-jewel bust from Chang'e into {argument name="character name" default="Erlang Shen Yang Jian"}. Keep the same black background, half-length bust framing, luxury macro photography look, floating pearl/bubble particles, and intricate translucent beadwork. Change the subject into a masculine Chinese mythological warrior deity with a taller jeweled crown, darker blue-and-gold hair ornaments, sharper heroic silhouette, ornate shoulder armor, and richer red, blue, teal, pearl, and gold gemstones. Add exactly 1 prominent dragon-themed jeweled shoulder/armor motif on the viewer's right, made from beads and gems. Replace the left-side title graphic with an ornate circular bead-and-gold emblem containing exactly 4 elements: 1 small golden dragon icon at the top, 1 large main title text {argument name="main title text" default="二郎神"}, 1 smaller subtitle text {argument name="subtitle text" default="楊戩"}, and 1 thin horizontal divider line between the title and subtitle. Make all lettering look constructed from golden pearls and filigree, matching the jewelry material. Preserve the elegant dark negative space and high-contrast sparkling highlights; do not add modern objects, extra text, or a watermark.
Profile / Avatar - Professional Old Photo Restoration — output

Profile / Avatar - Professional Old Photo Restoration

自学Ai的鱼头叔叔GPT Image 2
Using the provided reference image, perform a professional old-photo restoration and realistic colorization. Keep the original composition, the single standing person, pose, clothing shape, body proportions, and background layout unchanged. Remove all visible aging damage: mold-like white patches, peeling emulsion, scratches, stains, cracks, creases, faded contrast, and worn borders. Reconstruct missing image areas naturally without inventing new objects or changing the subject's identity. Convert the black-and-white damaged photo into a clean natural-color photograph from the {argument name="photo era" default="1970s"}: green outdoor foliage, a white long-sleeve shirt, dark trousers, natural skin tones, and soft daylight. Improve clarity, exposure, and texture while preserving the original documentary look; do not beautify, slim, reshape facial features, alter clothing, change the pose, add modern effects, or make it look like a newly staged portrait.
Profile / Avatar - Photorealistic Selfie Upscale — output

Profile / Avatar - Photorealistic Selfie Upscale

𝗦𝗮𝗻𝗶𝗮GPT Image 2
Using the provided reference image, upscale and enhance it into a crisp, high-resolution photorealistic version while preserving the original composition, selfie camera angle, pose, clothing, background, lighting, and color balance. Keep the same outdoor setting and all existing elements unchanged, including exactly 1 opaque square face-covering block, exactly 1 pair of sunglasses on top of the head, and the same white hoodie and light blue jeans. Improve fine detail, sharpness, texture clarity, and edge definition in the hair, fabric, foliage, pavement, and shadows without changing identity, adding new objects, removing objects, beautifying the scene, or altering the anonymizing block.
Comic / Storyboard - Anime Academy Girl with Popsicle — output

Comic / Storyboard - Anime Academy Girl with Popsicle

ないステGPT Image 2
Create a cinematic anime-style illustration in a wide 16:9 frame. In the left foreground, show {argument name="character name" default="a glamorous academy girl"} leaning in the shade of a grand stone arcade, seen from a slightly low, diagonal camera angle. She has long messy {argument name="hair color" default="blonde hair with a vivid pink front streak"}, tan skin, golden-brown eyes, glossy lips, and a confident teasing smile while holding a pale blue popsicle to her mouth. Her hair is windblown and highly detailed with loose curls, decorated with pearl-like beads, small flowers, and metallic hair ornaments. She wears an ornate fantasy school uniform: an oversized dusty pink cardigan with gold buttons, star pins, and a dark crest patch, a white open-collar blouse, a dark purple ribbon tie with gold compass-like motifs, a navy high-waisted skirt or corset waist with gold buttons and a small chain, plus delicate jewelry including hoop earrings, layered necklaces, rings, bracelets, and a black choker. The setting is an elegant fantasy academy courtyard with tall cathedral-like gothic buildings, spires, stone columns, arched walkways, and blue hanging banners decorated with gold compass emblems; include exactly 2 prominent hanging banners on the building side. In the right background, include small distant students walking through the sunlit courtyard, reduced to simple silhouettes and uniform shapes. Lighting should contrast deep cool shadow on the girl and arcade with bright midday sunlight in the courtyard, a saturated blue sky, scattered white clouds, and strong perspective lines leading toward the castle-like academy. Use a polished Japanese anime key visual style with reduced but precise linework, rich painterly background detail, dramatic composition, warm highlights, cool shadows, subtle lens-like cinematic depth, no text, no watermark.
Product Marketing - Tokyo Gothic Lolita Fashion Poster — output

Product Marketing - Tokyo Gothic Lolita Fashion Poster

AI Girl's Photo StudioGPT Image 2
Goal: Create a vertical fashion poster for a maximalist Tokyo gothic-lolita theme, styled like a glossy underground magazine cover and event flyer.

Canvas: Tall vertical 9:16 poster, high-contrast flash photography, dark black-red palette, ornate gothic textures, no clean white margins.

Main subject: A full-body young woman model stands slightly right of center in an elaborate black gothic lolita outfit. Her face is intentionally obscured by a soft square blur/mosaic censor block. She has long dark wavy hair with twin ornate hair accessories. Pose: legs crossed, one hand raised near her cheek, the other hand holding out the skirt edge. Expression is hidden; body language is cute, confident, and theatrical.

Outfit details: Use exactly 8 visible outfit/accessory groups: 1) black lace off-shoulder corset bodice with deep neckline, 2) short layered black ruffle skirt with glossy crinoline volume, 3) burgundy satin bows on the bodice and skirt, 4) black lace fingerless gloves and lace arm coverings, 5) silver cross necklace at the chest, 6) multiple dangling silver chains and cross charms around the waist, 7) black platform ankle boots with buckles, straps, metal rings, and cross ornaments, 8) two black-and-burgundy gothic hair ornaments with flowers, bows, lace, chains, and cross charms.

Background: A dense gothic interior with carved black frames, antique mirrors, red stained-glass panels, gold candle sconces, dim warm lights, and baroque decorative clutter. Make it look like a real flash photo taken in a dark gothic club or boutique, with wet glossy highlights and deep shadows.

Left-side poster typography: Fill the left third with ornate layered gothic lettering and decorative graphics. Include exactly 4 main text blocks: 1) large silver-black medieval blackletter headline reading “{argument name="main headline" default="Tokyo Gothic Style"}”, 2) hot-pink dripping graffiti brush text reading “{argument name="graffiti phrase" default="Curse Cute!"}”, 3) small elegant cursive footer reading “{argument name="event title" default="The Ultimate Tokyo Gothic Fashion & Pose Championship"}”, 4) small magenta footer line reading “{argument name="award line" default="2024 Grand Prix"}”.

Decorative elements: Use exactly 5 major gothic graphic motifs on the poster: 1) a large silver cross behind/overlapping the main headline, 2) spiderwebs in the upper-left corner, 3) black ornamental vines and filigree behind the text, 4) small pink hearts around the graffiti phrase, 5) an ornate silver cross near the lower-left area. Add red shadows, chrome highlights, distressed scratches, and layered embossing to the lettering.

Style: “Gorri-gori” heavy gothic, Japanese street fashion editorial, dramatic direct camera flash, glossy latex/leather shine, intricate lace detail, chrome silver jewelry, burgundy accents, punk-goth cute-but-cursed energy, extremely detailed, sharp focus on clothing and text, slightly chaotic but readable poster composition.

Constraints: The subject must remain full-body and right-centered; the face must be covered by a square blur. Keep the palette mostly black, silver, burgundy, and hot pink. Do not add extra people. Do not add extra text beyond the 4 specified text blocks. Make the gothic typography large, dimensional, and legible.
Infographic / Edu Visual - Rainy City Bus Stop Commute — output

Infographic / Edu Visual - Rainy City Bus Stop Commute

しらき@パワポ図解GPT Image 2
Create a clean monochrome line-art illustration with soft blue accents showing a cute rainy-day city commute at a bus stop. Use a wide horizontal canvas, mostly white background, thin black outlines, minimal shading, gentle hand-drawn style, and selective pale sky-blue coloring on accessories, signs, and bus route numbers. The scene contains exactly 5 commuters at a covered bus shelter: 1 person walking left under a clear umbrella while holding a takeaway cup, wearing a hoodie and backpack; 1 woman standing under an umbrella with a blue canopy, looking at a smartphone, wearing a long coat and carrying a large blue tote bag; 1 seated person in the shelter reading an open blue book; 1 standing person facing away holding a coffee cup and a folded blue umbrella, wearing a long coat, scarf, and small crossbody bag; and 1 person on the far right under an umbrella looking at a smartphone, wearing a long coat and carrying a blue shoulder bag. Add light diagonal rain strokes across the whole scene and wet pavement with subtle puddle reflections. On the left, include a vertical bus stop sign with a blue bus icon and the visible English text “BUS STOP” plus exactly 4 route numbers: 101, 150, 202, and 505 in blue. In the center shelter, include a poster titled “CITY BUS MAP” with a simple transit-map diagram. On the right edge, show the front portion of a bus arriving, with route number {argument name="bus route number" default="150"} displayed in a blue destination panel. In the background, draw a leafy tree on the far left, low bushes, simple city skyline silhouettes, two small clouds, and a tower resembling Tokyo Tower behind the bus shelter roof. Keep all faces simplified and softly anonymized, no detailed facial features, no logos or watermark. Emphasize cozy, calm commuting mood with restrained blue accents only. Optional customization: make the overall mood {argument name="weather mood" default="gentle rainy morning"}, the accent color {argument name="accent color" default="pale sky blue"}, the city setting {argument name="city setting" default="Tokyo-inspired urban skyline"}, and the illustration style {argument name="illustration style" default="minimal cute Japanese line art"}.
Profile / Avatar - Realistic Selfie With Face Privacy Block — output

Profile / Avatar - Realistic Selfie With Face Privacy Block

kiriGPT Image 2
Create a realistic vertical smartphone selfie photo of a young East Asian woman sitting indoors in a softly lit clinic or office waiting-room setting. She is framed from the upper arms to the top of the head, holding the camera slightly above chest level with both arms extending toward the bottom corners. Her face is intentionally covered by one large centered opaque rectangular privacy block in a warm taupe-brown color, spanning from the lower forehead area down to below the mouth and hiding all facial features. She has long dark brown wavy hair with a half-up hairstyle, airy side strands, natural flyaways, and soft bangs parted near the center; the hair falls over both shoulders in loose curls. She wears a sheer light gray-white blouse with translucent puff short sleeves, a gathered ruffled neckline, and delicate thin fabric texture. The background is a plain off-white wall with a small posted paper sign in the upper right containing partial Japanese text, including visible characters like 「控えA…」 and smaller lines beneath, slightly out of focus. Use natural phone-camera realism, indoor ambient lighting, subtle skin texture on neck and arms, shallow depth of field, muted neutral colors, no glamour retouching, no added accessories, no watermark, and keep the image vertical 9:16. Customize: {argument name="hair color" default="dark brown"}, {argument name="blouse color" default="light gray-white"}, {argument name="privacy block color" default="warm taupe-brown"}, {argument name="background setting" default="clinic or office waiting-room"}.
Profile / Avatar - Silver-Haired Gothic Fantasy Reclining Man — output

Profile / Avatar - Silver-Haired Gothic Fantasy Reclining Man

ツキマチGPT Image 2
Create a highly detailed anime-style fantasy portrait of an elegant young man reclining diagonally on a luxurious black leather sofa, viewed from above at a close three-quarter angle. The character has {argument name="hair color" default="silver-white"} tousled layered hair and {argument name="eye color" default="cool blue"} eyes, with a calm, alluring expression. He is slender but athletic, with exposed shoulders and collarbones, wearing a sleeveless glossy black high-neck top with ornate dark embroidery, a translucent smoky-blue sheer robe draped loosely off both shoulders, fitted black trousers, and layered blue-black sash fabric at the waist. Add elaborate silver gothic jewelry: one large cross-and-crescent chest brooch with dangling chains and tassels, one crescent moon waist ornament, multiple fine chain strands, metallic filigree clasps, and dark bracelets on one wrist. His left arm bends up with his hand near his head, and his right hand rests relaxed near his thigh; emphasize graceful long fingers and soft skin highlights. The setting is a moody opulent lounge with dark pillows, reflective black surfaces, warm candlelike highlights, and a vase of small purple-blue flowers in the upper right background. Use a romantic dark palette of black, charcoal, deep navy, smoky lavender, silver, and pearly highlights. Lighting should be cinematic and soft, with glossy reflections, sparkling dust motes, delicate fabric transparency, intricate embroidery, and polished metal details. Composition should crop tightly around the reclining figure from head to upper legs, with the face area unobstructed and no text, watermark, or extra characters.
Game Asset - Censored Silver-Haired Anime Fashion Girl — output

Game Asset - Censored Silver-Haired Anime Fashion Girl

Canário【カナリオ】GPT Image 2
Create a highly detailed vertical anime illustration of an elegant fantasy fashion woman named {argument name="character name" default="Lilva"} posing in a glittering nighttime city street. She is shown from upper thighs to head in a low-angle, glamorous three-quarter composition, leaning casually with one hand lifted near her lips and the other relaxed near her hip. Her face is intentionally obscured by a centered opaque square censor block in muted mauve-gray, covering the eyes, nose, and mouth while leaving the hair, ear, jawline, and jewelry visible. She has very long flowing {argument name="hair color" default="silvery ice-blue"} hair with luminous strands, wind-swept volume, and sharp anime linework. Her skin is pale and softly glowing. She wears a luxurious white and silver cropped outfit: ribbed sleeveless crop top, translucent draped sash across the chest, fitted high-waisted white trousers with a black belt and gold buckle, and a loose semi-transparent white jacket sliding off the shoulders. Add intricate celestial jewelry and accessories: exactly 9 visible jewelry groups consisting of a jeweled ear cuff and dangling blue gemstone earrings on one ear, a thin choker with blue gem, layered delicate necklaces with blue crystal pendants, bracelets on the raised wrist, rings on both hands, a gold chain belt detail, a dangling waist charm with crystal, a gold wristwatch bracelet on the lowered wrist, and a long hanging blue crystal tassel charm from that wrist. Include star, moon, and diamond motifs, tiny chains, faceted sapphire-blue crystals, metallic gold accents, and glossy manicured nails in pale blue. The background is a dreamy upscale urban night scene with ornate building facades, warm street lamps, illuminated shop windows, a dark hanging sign with a crest-like emblem, reflective stone and metal surfaces, falling sparkling particles, bokeh lights, and magical glowing shards. Use a cool lavender, silver, blue, and white palette with warm gold highlights, strong rim lighting from the right, soft bloom, lens sparkle effects, crisp clean line art, highly polished rendering, detailed fabric texture, translucent materials, and a luxurious ethereal atmosphere. Keep the composition unchanged: portrait 3:4 aspect ratio, character centered and filling most of the frame, hair sweeping across the left side, raised hand near the left side of the censor block, exposed shoulder on the right, no readable text, no watermark, no extra characters.
Profile / Avatar - Joyful Woman Raising Arms at Beach — output

Profile / Avatar - Joyful Woman Raising Arms at Beach

susumu_MindGPT Image 2
Create a realistic vertical lifestyle photo of a young woman standing on a sunny beach beside a sparkling blue ocean, arms raised high in a joyful V pose as if greeting the morning. She has {argument name="hair color" default="long black hair"} blowing naturally in the sea breeze, a slim figure, and her face is intentionally obscured by a soft square blur/mosaic privacy block centered over the face. She wears a fitted {argument name="top color and style" default="white short-sleeve cropped V-neck T-shirt"} and {argument name="bottom style" default="loose light blue drawstring beach pants"}, with subtle jewelry including bracelets on both wrists, a delicate necklace, and small dangling earrings. The scene is shot from waist-up to upper-thigh framing, slightly low/eye-level angle, with the ocean on the left, gentle foamy waves and pale sand on the right, distant green hills along the horizon, and a bright pale blue sky with a few soft clouds. Use natural daylight, warm summer atmosphere, airy morning mood, soft film-like colors, shallow depth of field, realistic skin and fabric texture, wind-tossed hair, sparkling highlights on the water, no text, no watermark, no extra people.
Infographic / Edu Visual - Japanese Carbonara Recipe Infographic — output

Infographic / Edu Visual - Japanese Carbonara Recipe Infographic

バナナ🍌GPT Image 2
Goal: Create a clean Japanese recipe infographic poster for {argument name="recipe title" default="本格カルボナーラの作り方"}, showing authentic Roman carbonara in a warm editorial style.

Canvas: Vertical poster, approximately 4:5 aspect ratio, soft cream paper background with subtle grain. Center everything with generous margins and thin divider lines. Use a modern Japanese layout with rounded cards, muted earth tones, and high-quality flat food illustration.

Header: Large bold black Japanese headline at the top: {argument name="recipe title" default="本格カルボナーラの作り方"}. Under it, smaller subtitle: {argument name="subtitle" default="本場ローマの味を自宅で"}. Below the title, place a large illustrated plate of spaghetti carbonara on an oval white plate, with golden pasta, creamy sauce, black pepper specks, and cubed browned guanciale. Put a pale yellow circular halo behind the plate.

Info badges: Add exactly 2 small rounded yellow badges beneath the hero plate, one on the left reading {argument name="cooking time" default="調理時間:20分"}, and one on the right reading {argument name="difficulty label" default="難易度:★★★★★"}.

Ingredients section: Create a section titled 「材料(2人分)」 with a small green bowl icon. Show exactly 8 ingredient rows in two columns, each with a small square ingredient illustration and Japanese text: 1) 「スパゲッティ 200g」 with pasta/bacon-like icon, 2) 「グアンチャーレ 100g」 with egg/guanciale-style icon, 3) 「卵黄 3個」 with cheese wedge icon, 4) 「全卵 1個」 with yellow pasta/cheese icon, 5) 「ペコリーノ・ロマーノ 50g」 with cured meat icon, 6) 「パルミジャーノ 20g」 with cheese block icon, 7) 「黒胡椒 適量」 with peppercorn icon, 8) 「塩 適量」 with salt icon.

Tip card: To the right of the ingredients, place one muted green rounded rectangle card with a small chef-hat icon and heading 「ポイント」 in pale yellow. Body text: 「必ず火を止めてから卵液を加えること!余熱でとろみをつけます」.

Steps section: Below another thin divider, add the heading 「作り方」 in reddish brown with a short green underline. Use exactly 6 instruction cards arranged in two columns and three rows, each with a red-brown circular number badge, a small square cooking icon, and Japanese text. The visible number badges should be exactly: 01, 02, 03, 03, 04, 05. Step texts: 01 「グアンチャーレを1cm角に切る」, 02 「グアンを湯・カ丸切る」, 03 「卵黄・全卵・チーズ・黒胡椒をボウルで混ぜる」, duplicate 03 「たっぷりの湯に塩一入をスパゲッティを茹でる」, 04 「フライパンでグアンチャーレをカリカリになるまで炒める」, 05 「火を止めてパスタを加え卵液と素早く和える」. Keep some slight AI-generated awkwardness in the Japanese text to match the original look.

Footer: Add a dark charcoal horizontal footer band at the bottom. Center a small spaghetti/fork icon, the brand name “Cucina Roma” in elegant italic serif lettering, and the handle “@cucina_roma_jp” below it.

Visual style: Soft vector illustration mixed with polished infographic design, warm cream, ochre, caramel, muted green, and terracotta palette. Rounded corners, minimal shadows, clean Japanese typography, no photography, no extra sections, no watermark.
Game Asset - Anime Sci-Fi Character Sheet — output

Game Asset - Anime Sci-Fi Character Sheet

水姫(Mizuki)GPT Image 2
Goal: Create a polished anime character sheet for a futuristic schoolgirl angel character named {argument name="character name" default="Yuuka Kinaye"}, styled like a mobile game profile/spec sheet with clean UI panels, decorative typography, and intentionally pseudo-Japanese/pseudo-English microtext.

Canvas: Vertical poster, 2:3 aspect ratio, high resolution. White central background with a lavender left border and bright cyan geometric accents at the top and bottom. Use thin pale-gray framing lines, small triangles, dots, diamonds, and tech-interface decorations scattered around the page.

Main subject: One full-body anime girl dominates the left and center of the poster. She has {argument name="hair color" default="deep violet-purple"} long twin-tail hair with curled ends, side bangs, glossy highlights, and small cyan hair accessories. She has large purple eyes and a calm, slightly serious expression. A black-and-cyan halo floats above her head. She stands in a confident contrapposto pose, holding a purple smartphone in her left hand while her right hand rests near her waist. Outfit: black blazer with white lapels, white shirt, cyan necktie, short black pleated skirt, oversized white-and-cyan futuristic jacket draped around her shoulders, black gloves, black thigh-high tights, chunky black shoes, and a white utility pouch attached at her hip. Include crisp cel-shaded anime rendering, glossy highlights, clean line art, and a premium character-design-sheet finish.

Layout and visible elements: Use exactly 1 main full-body character illustration. Add exactly 1 large schedule/profile card in the upper right with a cyan header and exactly 5 horizontal table rows. The card title should resemble a character scanning/status panel and include mostly decorative pseudo-text. Add exactly 4 stacked rounded-rectangle portrait thumbnails down the right side, each showing a close crop of the same character with her halo and purple hair; the face area may be softly obscured or simplified like censored preview tiles. Add exactly 1 square logo panel near the center-right containing a large cyan stylized number-like emblem resembling “77”. Add exactly 1 paragraph block of tiny gray placeholder profile text beneath the logo. Add exactly 1 small chibi version of the character near the lower middle, lying or leaning forward with a grumpy expression, a small speech bubble, halo above, and a small keyboard or rectangular board beside her. Add exactly 1 large name label near the bottom reading {argument name="display name text" default="Yuuka Kinaye"} in widely spaced pale purple letters, with small decorative ID numbers underneath. Add exactly 1 bottom cyan-and-lavender footer bar with short pseudo-Japanese labels and a small purple emblem.

Text content: The upper-right heading should read approximately {argument name="header text" default="CHARACUR SCIANNINGT5 YUUKA"} with a smaller decorative line of pseudo-Japanese text below it. The schedule card should have a cyan header containing the character name and a status/date-like field, then 5 rows with left-side pseudo labels and right-side times: 19:30, 18:30, 18:30, 18:30, 18:30. Keep most small text intentionally nonsensical, faint, and UI-like rather than fully readable.

Visual style: Modern anime character sheet, Blue Archive-inspired sci-fi school uniform aesthetic, white/lavender/cyan palette, clean vector UI overlays, soft shadows under cards, high-detail polished illustration, sharp cel shading, attractive character design, no photorealism.

Constraints: Keep the page uncluttered but information-rich. Preserve the exact counts: 1 full-body character, 1 schedule card with 5 rows, 4 right-side portrait thumbnails, 1 logo panel, 1 chibi illustration, 1 bottom footer bar. Do not add extra characters, weapons, scenery, or readable branding.
Profile / Avatar - Faceless Anime Fashion Portrait — output

Profile / Avatar - Faceless Anime Fashion Portrait

ふぁりー💚💛GPT Image 2
Create a polished anime-style bust portrait of a fashionable young woman centered on a plain white background. She has {argument name="hair color" default="dark charcoal black with pale silver-white inner tips"} in a short layered bob, slightly messy and voluminous, with soft inward curls at the ends, loose side strands, and one visible ear on the right. Her face is intentionally obscured by a large vertical rectangular censor block filled with a smooth muted taupe-gray gradient, covering the eyes, nose, and mouth while leaving the hair, ear, chin edge, and clothing visible. She wears a fitted black ribbed turtleneck and an oversized glossy black leather trench coat with wide lapels, shoulder epaulets, visible seams, stitch details, and a single dark button; the coat has strong reflective highlights and deep shadow folds. Use clean high-resolution Japanese illustration rendering, crisp line art, subtle cel shading mixed with semi-realistic fabric highlights, cool monochrome palette, elegant street-fashion mood, front-facing three-quarter bust crop from chest upward, no text, no accessories, no background objects, no watermark.
YouTube Thumbnail - Correct Face in Existing Thumbnail — output

YouTube Thumbnail - Correct Face in Existing Thumbnail

無職51歳 初めて実験物語GPT Image 2
Using REFERENCE_1 as the current YouTube-style thumbnail and REFERENCE_0 as the identity reference, regenerate the thumbnail with the same layout and composition, but replace only the man's face so it matches the person in REFERENCE_0. Keep the existing pose, pointing finger, bass guitar, room background, lighting, high-contrast thumbnail style, and all Japanese typography unchanged. Preserve exactly 5 visible text/badge areas: the top blue banner, the large main headline, the bottom blue music/tempo line, the yellow circular age badge, and the bottom-right black square badge. Do not redesign the thumbnail; the layout is already good. The only intended correction is that the face should be {argument name="character identity" default="the same person as in REFERENCE_0"}.
Comic / Storyboard - Four-Panel Smartphone T-Rex Manga — output

Comic / Storyboard - Four-Panel Smartphone T-Rex Manga

RoyGPT Image 2
Goal: Create a vertical four-panel manga/comic page in a gritty hand-painted steampunk urban style. The story is about a frantic young coffee-stand worker riding a mechanical green tyrannosaurus while searching for his smartphone, then discovering it is ringing inside the dinosaur’s mouth.

Canvas: Tall vertical comic page, approximately 9:16 aspect ratio, with a thin black outer border and thick black horizontal gutters dividing exactly 4 stacked panels. Use textured ink, watercolor gouache shading, muted sepia city tones, and dynamic manga speed lines in the final panel.

Main characters: A young adult man with messy curly {argument name="hair color" default="black"} hair, face intentionally blurred/obscured, wearing a dark work shirt, black pants, boots, leather suspenders, and a stained apron with a small “COFFEE STAND” patch. He holds a metal gooseneck coffee kettle in his left hand throughout. Beside/under him is a large rugged mechanical T-rex with olive-green riveted armor plates, leather saddle and straps, visible seams, sharp white teeth, glowing orange cybernetic eye, and expressive cartoon body language.

Setting: Narrow old city street lined with brick buildings, shop signs, hanging lamps, and coffee-stand signage. Dust, debris, scratches, and paper flecks add motion and chaos. The atmosphere is cinematic, vintage, and slightly absurd.

Panel layout and count: Use exactly 4 panels.
1. Top panel: Wide action shot. The coffee worker rides the armored dinosaur down the street, looking confused and searching himself while holding the kettle. The dinosaur roars with mouth open. Add a small red-and-white mushroom floating above them as a whimsical detail. Include one large white speech bubble on the left with Japanese text: 「あれ?スマホどこだ?」
2. Second panel: The man leans forward and checks the saddle bags/pockets while still holding the kettle. The dinosaur looks sideways with a confused expression and a small question mark near its head. Include one white speech bubble on the left with Japanese text: 「ポケット?サドル?」
3. Third panel: The man raises one finger as if realizing something, still holding the kettle. The dinosaur faces right with its mouth closed, orange eye glowing, making a vibrating/ringing motion. Add Japanese sound effect text near the dinosaur: 「ブルル..」 Include one white speech bubble on the left with Japanese text: 「鳴っておる…?」
4. Bottom panel: Dramatic reveal with radial speed lines and warmer orange energy. The man points toward the dinosaur’s open mouth in shock while holding the kettle. Inside the dinosaur’s mouth, clearly show a smartphone standing upright among the teeth, screen lit with an incoming call interface and Japanese text {argument name="phone screen text" default="着信中"}. Include one large jagged speech bubble on the left with bold Japanese text: 「これじゃ!!」 Add a small sparkle near the dinosaur’s head.

Text and lettering: Preserve the Japanese dialogue exactly as specified, in bold black hand-lettered manga style inside white speech bubbles. Keep all text legible. Do not add extra dialogue beyond the listed bubbles and sound effect.

Style constraints: Highly detailed comic illustration, expressive poses, visible brush texture, black ink outlines, rough paper grain, cinematic lighting, no photorealism, no watermark, no captions outside the comic panels.
Profile / Avatar - Pastel Cyberpunk Retro Electronics Portrait — output

Profile / Avatar - Pastel Cyberpunk Retro Electronics Portrait

こっちに〜るGPT Image 2
Create a cinematic photorealistic high-angle portrait of a cute young woman with {argument name="hair color" default="pastel pink"} hair styled into two large space buns, crouching on the floor in a dark retro electronics room. Her face is covered by a flat opaque square privacy blur matching muted skin tones, with no facial details visible. She wears a pastel sailor-style school uniform in {argument name="outfit colors" default="pink, cream, and sky blue"}: oversized cream sleeves, blue striped cuffs and collar, a short pleated skirt, and soft kawaii styling. A small round crossbody bag hangs at her side, designed like a smiling cartoon face with rosy cheeks and two fluffy blue pom-poms. Surround her with vintage analog machines, tangled cables, metallic control panels, knobs, switches, and old broadcast equipment; place a bulky CRT television on the right with a dim greenish screen and a second rectangular blur over part of the screen. Use moody low-key lighting with teal shadows, warm amber highlights, dusty film grain, slight lens distortion, and an 1980s cyberpunk/kawaii atmosphere. The camera looks down from above at a three-quarter angle, framing her from head to knees, with the equipment filling the background. Make it feel like a real photograph shot on high ISO film, richly textured, dark, nostalgic, and slightly surreal.
Social Media Post - Japanese AI Consulting OGP Banner — output

Social Media Post - Japanese AI Consulting OGP Banner

はな🌸"あなたの色"で輝くお手伝いGPT Image 2
Goal: Create a polished Japanese OGP/social media banner for a B2B generative AI implementation support service, emphasizing AI becoming part of daily work.

Canvas: Wide horizontal 1200×630 px banner, clean corporate technology style, bright and trustworthy. Use a pale sky-blue to white gradient background with a soft sunrise glow on the right.

Layout: Place a semi-transparent frosted white rectangular panel slightly left of center, occupying about 55% of the width and 82% of the height, with a subtle white border and soft shadow. All main text sits inside this panel, left aligned. The right half shows a futuristic city skyline fading into the background, with sweeping highway-like light trails curving from the lower right toward the center.

Text content: At the top of the panel, add a small blue brand line with thin horizontal decorative lines on both sides: {argument name="brand text" default="Lightkeeper AI伴走"}. Beneath it, add a smaller dark navy subtitle: {argument name="subtitle text" default="中小企業の生成AI内製化・定着支援"}. The main headline is very large, bold, dark navy Japanese type on two lines: {argument name="headline text" default="AIを、\n日常の仕事に。"}. Near the bottom left, add a short gold horizontal accent line, then a smaller bold navy supporting message on two lines: {argument name="supporting text" default="研修で終わらせず、\n現場定着まで伴走します。"}.

Visual style: Use modern Japanese sans-serif typography, heavy weight for the headline, crisp spacing, high contrast navy text. Color palette: deep navy, medium corporate blue, soft white, pale cyan, and warm gold highlights. Add translucent floating UI panels in the background: exactly 5 faint panels, including 1 line chart card on the left, 1 donut chart card on the lower left, 1 bar chart card upper right, 1 organization/network icon card mid-right, and 1 small abstract data card far right. Add subtle hexagon patterns and small glowing particles, but keep the design uncluttered.

Constraints: Preserve the Japanese text exactly as written in the default values. No people, no logo icon, no watermark, no mockup frame, no extra text. Ensure the final banner feels like a professional landing-page hero or OGP image for an AI consulting service.
Comic / Storyboard - Android Spaceship Rover Exploration — output

Comic / Storyboard - Android Spaceship Rover Exploration

カーブミラーGPT Image 2
Create a highly detailed cinematic sci-fi concept art scene of {argument name="exploration rover label" default="EXR-17"} exploring an alien Earth-like valley. The main subject is one large white-and-gray mechanical quadruped rover in the foreground, walking through a shallow rocky stream; it has four heavy articulated legs, one long front-right manipulator arm touching the water with a sampling tool, a boxy armored chassis covered in panels, bolts, antennas, cables, blue indicator lights, and a tall sensor mast with three prominent camera lenses. Surround it with lush green moss, wet stones, tiny glowing blue flowers, splashing water, waterfalls, steep cliffs, distant mountains, and a bright blue sky with dramatic clouds. Add a colossal green-and-blue planet or moon dominating the upper-right sky with a bright sun flare near its edge. Include exactly 3 small survey drones in the sky: one large drone at upper right projecting a blue scanning beam downward, one small drone at upper left over the mountains also projecting a blue beam, and one distant drone near the center-right valley projecting a faint blue beam. Add translucent holographic AR mapping overlays on the right side of the landscape with exactly 3 visible labels: “MINERAL DEPOSIT”, “WATER SOURCE”, and “CONSERVATION ZONE”, plus blue wireframe scan grids and triangulation lines. In the upper-left corner, overlay a rectangular blue holographic display panel showing {argument name="android description" default="a female android aboard a spaceship"}: a reclining silver-white female android torso with exposed cybernetic joints, long pale hair, tubes and cables around her head and shoulders, glowing blue circuitry, and medical diagnostic UI graphs; her face is covered by a soft opaque square blur. The inset panel should have a neon cyan frame and contain exactly 4 main UI diagnostic regions: a waveform graph, a small data table, a circular targeting reticle, and thin technical readout lines. Overall style: ultra-detailed Japanese sci-fi key visual, photorealistic 3D concept art mixed with anime-inspired cyberpunk aesthetics, sharp focus, high dynamic range, vivid cyan-blue holograms, intricate hard-surface robotics, volumetric light, sparkling water, dense environmental detail, epic scale, no watermark, no extra readable text beyond the specified labels and rover marking.
Profile / Avatar - Surreal Schoolgirl and Blob Creature — output

Profile / Avatar - Surreal Schoolgirl and Blob Creature

シオナイGPT Image 2
Create a surreal cinematic analog-film photograph in a dreamy white architectural fantasy interior under an open painted sky. A {argument name="character description" default="Japanese schoolgirl in a navy sailor uniform with a black pleated skirt, white long-sleeve top, dark neckerchief, black socks, black loafers, and a dark backpack"} crouches on the left, leaning forward with her hands clasped near her knees as if gently greeting a strange companion. Beside her, centered and slightly to the right, sits a large pale pink, soft, rounded mascot-like blob creature with a smooth rubbery texture and multiple stubby tentacle-like feet visible at the bottom; keep the creature non-threatening, cute, and mysterious. Add exactly two opaque censorship blocks as visible graphic overlays: one large warm gray square covering the creature’s upper center and much of the middle of the image, and one smaller warm gray square covering the girl’s face. The environment is an ethereal white plaster space with curved walls, glossy pale floor reflections, and exactly three black oval openings: one high oval opening on the far left wall, one tall vertical oval opening lower on the left wall, and one large half-dome black opening on the right. Include exactly two sets of white stairs: a prominent staircase rising along the left wall and a smaller stepped platform leading toward the right opening. The background sky is blue-gray with soft clouds and a faint full moon in the upper right. Use muted colors, soft grain, slight vignette, low-contrast 1990s film still mood, horizontal 3:2 composition, camera at seated eye level, no text, no watermark.
Social Media Post - VEYA Anime Canyon Action Poster — output

Social Media Post - VEYA Anime Canyon Action Poster

42GPT Image 2
Create a dynamic cinematic anime-style action poster of {argument name="character name" default="VEYA"}, a young warrior sprinting and lunging low across a rocky desert canyon at sunset. The composition is wide 16:9, with an extreme low-angle perspective and strong motion blur: the character is centered, hair whipping wildly backward, one huge stone-and-crystal armored hand reaching toward the viewer in foreshortened perspective, the other arm pulled behind, legs spread in a powerful running leap. The outfit is dark tactical clothing in charcoal and black, with organic jagged pale stone armor wrapping the forearm, hand, shin guards, and rugged shoes; the armor looks like layered shards of bone, crystal, and rock. Add flying gravel, dust explosions at both feet, scattered debris suspended in the air, and canyon walls receding into a bright hazy sunlit background. Use warm beige, sand, sepia, and smoky gray tones, high contrast lighting, detailed painterly linework, and an epic manga-meets-concept-art finish. Across the top, place the large thin uppercase title text {argument name="headline text" default="VEYA"} in a minimalist widely spaced geometric font, dark gray, spanning nearly the full width. Keep the face hidden by a centered opaque square censor block in muted brown, covering the facial features while leaving the hair, ears, body pose, and action readable. No additional text, no watermark.
Comic / Storyboard - VEYA Character Design Sheet — output

Comic / Storyboard - VEYA Character Design Sheet

42GPT Image 2
Goal: Create a polished anime-style character design sheet for {argument name="character name" default="VEYA"}, a futuristic streetwear heroine with organic ivory armor growths.

Canvas: Wide horizontal concept art board, 16:9 aspect ratio, warm off-white background, clean margins, premium game/animation production sheet presentation. Put the name “VEYA” in thin uppercase letters at the top left with a short horizontal line beneath it.

Layout: Show exactly 4 full-body character views across the upper and left area: 1 large dynamic hero pose on the far left, 1 smaller front view, 1 smaller back view, and 1 smaller side view. On the upper right, include exactly 6 cropped head bust studies arranged in a 3-by-2 grid, all showing the same hairstyle from slightly varied angles/expressions. Along the bottom, include exactly 4 labeled detail panels under the label “DETAILS”: close-up of the organic armored hand, close-up of the vine-like leg armor, close-up of the cropped jacket and neck area, and close-up of cargo pants pocket construction. To the lower middle-right, include exactly 1 object under the label “PROP”: a seed-pod-shaped ivory sculptural artifact with layered leaf forms and a glowing round core. To its right, include the label “SILHOUETTE” with exactly 2 black character silhouettes, one front-facing and one back-facing. On the far lower right, include the label “PALETTE” with exactly 4 vertical color swatches: near-black, charcoal gray, warm beige ivory, and off-white.

Subject details: The character is a young adult woman with tan skin, athletic slim build, asymmetrical long {argument name="hair color" default="black"} hair swept dramatically over one side, shaved/undercut on the opposite side, and a confident urban attitude. She wears a cropped black sleeveless top, a short oversized dark cropped jacket or vest with angular shoulders, high-waisted charcoal cargo jogger pants with seams and pockets, a slim belt, and chunky cream-and-black sneakers. Add distinctive organic ivory armor shaped like twisting leaves or branches on one arm and one lower leg, with layered petal-like plates, subtle gray shadows, and elegant botanical curves.

Visual style: High-end 2D concept art, clean ink lines, soft cel shading, subtle watercolor texture, muted charcoal and ivory palette, fashionable sci-fi fantasy streetwear, crisp production-design sheet, no photorealism. Faces should be simplified or softly obscured rather than highly detailed, while clothing, hair, armor, and props remain sharp and readable.

Constraints: Keep the composition uncluttered and balanced; preserve all visible labels in English: “VEYA”, “DETAILS”, “PROP”, “SILHOUETTE”, and “PALETTE”. Use exactly 4 full-body views, 6 head busts, 4 detail panels, 1 prop, 2 silhouettes, and 4 palette swatches. Do not add extra characters, weapons, logos, watermarks, or background scenery.
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 ↗