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.

探索提示词

10,044 条提示词

Product Marketing - Seaside Rock Festival Poster — output

Product Marketing - Seaside Rock Festival Poster

ai honobono | AI Art, Fiction, MusicGPT Image 2
Create a vertical 2:3 summer rock festival artist-promo poster with a dreamy Japanese seaside music-festival atmosphere. Use a bright, overexposed photo-realistic “follow me” composition: a young woman seen from behind and slightly to the side holds the viewer’s hand in the foreground, her arm reaching toward the camera with strong depth of field. She has long dark hair blown dramatically to the right by the ocean wind, wears a loose sheer white off-shoulder shirt over a white strap top and light blue denim jeans. Her face is intentionally covered by a centered soft gray square privacy block. In the background, place an outdoor concert stage on a beach or waterfront with visible truss, speakers, glowing stage lights, a small band performing, and a crowd dressed mostly in white facing the stage; show blue sea and sky with soft clouds. Color grade the whole image in airy cyan-blue and white tones, high-key sunlight, gentle lens haze, summery nostalgic film look, shallow depth of field, no harsh contrast. Add exactly 5 handwritten white text elements: 1) large top title “{argument name="festival title" default="天 ROCKFEST2026"}” in thin casual hand lettering, 2) smaller subtitle beneath it “{argument name="subtitle" default="The Summer Camp"}”, 3) left-middle handwritten Japanese quote “{argument name="quote" default="天の音がするほうへ、一緒に"}” with two simple underline strokes, 4) bottom-center artist name “{argument name="artist name" default="ai honobono"}”, and 5) bottom date line “{argument name="date line" default="DAY 1   7/25 SAT"}”. Keep the typography light, imperfect, and chalk/marker-like, integrated into the sky and lower foreground. Avoid extra logos, watermarks, borders, or additional text.
Product Marketing - Rainbow Magic Stone Board Game Flyer — output

Product Marketing - Rainbow Magic Stone Board Game Flyer

塚原 慧GPT Image 2
Goal: Create a richly illustrated anime-style one-page promotional explanation image for a fantasy tile placement board game titled {argument name="game title" default="虹の魔法石"}, with a dark medieval magic tavern atmosphere and readable Japanese game information.

Canvas: Wide horizontal 16:9 layout, approximately 1200×675 px. Use an ornate dark-brown and gold fantasy UI frame, parchment panels, glowing candlelight, jewel highlights, and premium board-game-ad styling.

Layout: Split the image into three main vertical areas plus a bottom information bar. Left area occupies about one third of the canvas and shows the narrative illustration. Center area contains the title header and two explanatory parchment sections. Right area contains a numbered game-flow panel with exactly 4 steps. Bottom bar spans the full width with player count, play time, age, price, publisher, and event booth information.

Left illustration: Inside a cozy medieval tavern or wizard workshop, show exactly 4 anime adventurers gathered around a wooden table covered with many small rainbow-colored gem or hex-tile pieces. The four visible characters are: 1) a blue-armored knight or soldier on the far left, 2) a dark-robed wizard-like figure near the back left, 3) a pale blond noble or mage seated at the center, 4) an orange-hooded adventurer on the right. Warm candles, hanging lanterns, shelves, bottles, magic tools, timber beams, and deep shadows fill the background. The table glows with scattered red, green, blue, yellow, purple, and clear gem fragments. Faces may be softly simplified, but the scene should feel detailed and cinematic.

Top header: Place a circular rainbow gemstone emblem on the upper center-left of the text area. Beside it, large gold-and-rainbow Japanese title text: {argument name="game title" default="虹の魔法石"}. Under the title, add the subtitle text: "Rainbow Magic Stone / タイル配置ゲーム". Add subtle sparkles around the title and gemstone.

Center parchment panel: Add the large tagline in Japanese: {argument name="tagline" default="砕かれた魔法石を修復せよ。虹の橋を渡るのは、ただひとり。"}. Below it, create exactly 2 labeled red-ribbon section headers: 1) "あらすじ" and 2) "ゲーム概要". Under "あらすじ", include a short Japanese fantasy story paragraph about repairing the shattered rainbow magic stone to build a rainbow bridge to the dragon king’s valley, and only one hero reaching it. Under "ゲーム概要", explain in Japanese that players place rainbow fragments, hexagonal tiles, on the board, score by aligning red, green, blue colors and circle/star marks in straight lines, and can aim for instant victory after arranging a required number.

Right parchment panel: Title ribbon reads "ゲームの流れ". Include exactly 4 numbered steps, each in its own parchment row with a colored shield badge number. The 4 steps are: 1) red shield, label "【準備】", text "ボードを並べ、各自の虹の欠片を山札にする"; 2) green shield, label "【配置】", text "手番プレイヤーが虹の欠片を1枚ボードに置く"; 3) blue shield, label "【得点】", text "カラー・マーク・無印が直線に揃うと得点獲得"; 4) red shield, label "【勝利】", text "全員の欠片がなくなったら最多得点のプレイヤーが勝利". Add a small note at the bottom: "※カラーやマークを一定数以上並べると即時勝利!".

Bottom bar: Use icon-and-text blocks separated by thin gold dividers. Include exactly 4 specification items on the left: 1) people icon with "3〜4人", 2) clock icon with "60分", 3) calendar or age icon with "8歳以上", 4) yen coin icon with "¥1,000". In the middle, show the publisher name {argument name="publisher name" default="まほろば"}. On the right, show event text {argument name="event booth text" default="ゲームマーケット2026春 F-18"}.

Visual style: High-detail Japanese anime fantasy illustration mixed with polished board-game flyer design, ornate gold borders, dark walnut wood, parchment texture, red ribbon labels, embossed typography, glowing magical rainbow gems, dramatic warm lighting, and dense decorative details. Keep all panels legible and balanced. Avoid modern flat design, avoid blank white space, avoid extra characters, avoid extra numbered steps, and avoid changing the exact counts.
Infographic / Edu Visual - Japanese Self-Care Flower Mark Card — output

Infographic / Edu Visual - Japanese Self-Care Flower Mark Card

chi OkeyGPT Image 2
Goal: Create a warm Japanese self-care affirmation card about a good-luck charm for someone who missed their chance to sleep, encouraging them to reward themselves with a big flower mark on the calendar.

Canvas: Horizontal 16:9 illustration, soft off-white textured paper background, watercolor and colored-pencil feel, gentle bedtime mood. Use muted cream, beige, dusty gold, sage gray, soft brown, faded denim blue, and coral red accents.

Layout: Place the main Japanese typography on the left two-thirds and a cute child illustration on the right third. The text is arranged in 6 visible lines: 1) large handwritten title at top left with a pink sparkling heart, 2) smaller line beneath it, 3) very large gold phrase in the center, 4) smaller line beneath, 5) medium line over a sage-gray horizontal brush stroke with a red circled stamp around one character, 6) final line at bottom with one word in coral red. Add a gold swoosh underline under the final line.

Text content: Use the exact Japanese text: {argument name="title text" default="ひつじのおまじない💕"}; second line: {argument name="encouragement line" default="がんばった自分に"}; large center line: {argument name="main reward phrase" default="大きな花丸"}; next line: "をあげるつもりで"; brush-stroke line: "カレンダーに印つけて" with the character "印" circled in coral red; final line: "自己肯定感アップするよ" with "アップ" colored coral red and the rest dark brown.

Subject details: On the right, draw exactly 1 cute young girl in a soft storybook watercolor style. She has light brown wavy shoulder-length hair, rosy cheeks, and a calm happy expression. She wears a fluffy white sheep hat with tiny ears, a pale yellow short-sleeve shirt, loose denim-blue overalls with rolled cuffs, white socks, and white sneakers. She is walking lightly to the right with one leg lifted, holding both arms above her head to make a big circle like a flower mark. She carries exactly 1 cream shoulder bag with a small sheep face emblem.

Decorative elements: Include exactly 1 small red ball near her feet, exactly 1 spiral-bound desktop calendar on the far right with a coral flower mark drawn on it, exactly 2 coral flower-mark doodles total, exactly 1 coral circled stamp around the word "印", exactly 1 pink sparkling heart near the title, and many tiny gold stars scattered around. Behind the girl, place approximately 5 imperfect concentric metallic-gold brush circles like a halo. Add loose gold brush arcs in the top-right and bottom-left corners, plus subtle gray-blue paint washes in the top-left and bottom-left corners.

Visual style: Gentle Japanese kawaii watercolor postcard, soft edges, hand-drawn imperfect lines, elegant brush-calligraphy typography mixed with neat handwritten text, airy composition, cozy night-time encouragement, no harsh outlines, no photorealism.

Constraints: Keep all text legible and positioned as described. Do not add extra characters, extra calendars, or extra major objects. No watermark, no logo, no English text.
Game Asset - Crystal Cat Familiar — output

Game Asset - Crystal Cat Familiar

ハスノ@核心ぶち抜きヒーラー/天空図書館夫婦GPT Image 2
Create a luminous fantasy anime illustration of {argument name="character name" default="Nico"}, a small elegant crystal cat familiar standing in a majestic cathedral made of glass and gemstones. The cat has a petite foxlike feline body, oversized pointed ears, a delicate smiling face, long white whiskers, and large sparkling {argument name="eye color" default="violet-blue"} eyes with glossy highlights. Its entire body appears made from translucent faceted crystal, primarily {argument name="crystal body color" default="icy blue and lavender"}, refracting rainbow prismatic colors across its face, torso, legs, paws, and ears. Add one diamond-shaped gem embedded on the forehead, plus a faceted jewel collar or chest ornament. The cat stands in a graceful three-quarter pose, front legs straight, back slightly arched, head turned toward the viewer, with an expression that feels magical, confident, and gentle. Include exactly two long flowing tails curling upward behind it, each ending in a bright crystal flame or gem tip: one glowing pink-magenta and one glowing cyan-blue, both surrounded by tiny star sparkles. The setting is an ethereal crystal palace interior with tall arched columns on both sides, stained-glass windows glowing softly in the background, a polished reflective floor, and distant gem formations. Use a pastel palette of blue, lavender, pink, cyan, and pearly white, with soft volumetric backlight, rainbow caustics, glittering particles, lens-like starbursts, and a dreamy high-detail gpt-image-2 fantasy art style. Keep the composition centered, full-body, square format, no text, no watermark, no extra animals.
Comic / Storyboard - Pombero Folklore Character Sheet — output

Comic / Storyboard - Pombero Folklore Character Sheet

Rafa.ioGPT Image 2
Goal: Create a dark fantasy creature-design character sheet for {argument name="character name" default="Pombero"}, alias {argument name="alias" default="Shadow of the Monte"}, presented as a professional concept-art turnaround on a clean off-white background.

Canvas: Wide horizontal 16:9 presentation sheet, museum-white paper texture, realistic high-detail digital painting with muted earthy colors, fine ink-like labels, thin divider lines, and a polished production-design layout.

Main subject: A short, muscular nocturnal forest spirit inspired by Paraguayan rural folklore. He has rough scarred bark-like skin, dirty bare feet, long clawed fingers, a hunched primal posture, and a faceless or obscured shadowed head under a ragged straw hat. His outfit is made from layered dry leaves, bark strips, vines, feathers, shredded fabric, rope ties, a shoulder mantle, ragged skirt-like wrappings, ankle bindings, and a round tobacco pouch at the waist. The mood is silent, elusive, territorial, cunning, and feral rather than heroic.

Layout: Place one large full-body hero figure on the left, occupying about one third of the canvas, standing barefoot with one arm extended and the other hanging, covered in ragged leaf-and-bark camouflage. At the top left, add the large title “{argument name="title text" default="POMBERO"}” and beneath it the subtitle “Alias: Shadow of the Monte”. Across the upper middle show exactly 4 labeled turnaround poses: “NEUTRAL FRONT VIEW”, “BACK VIEW”, “SIDE PROFILE”, and “SECONDARY ATTITUDE”. The secondary attitude pose should show him kneeling with one knee raised, holding a smoking hand-rolled tobacco roll.

Facial / expression studies: Under the turnaround row, include exactly 6 square portrait-study panels labeled, left to right: “SLY MENACE”, “EERIE CALM”, “PREDATORY FOCUS”, “CUNNING AMUSEMENT”, “WATCHFUL SILENCE”, and “WRATH”. Each portrait is mostly hidden in shadow beneath the hat, emphasizing a silent gaze rather than readable human features.

Details row: Along the lower center, include exactly 6 close-up detail panels labeled, left to right: “HAND & TOBACCO ROLL”, “TOBACCO POUCH”, “SHOULDER WRAP DETAIL”, “WAIST DETAILS”, “FOOT & ANKLE”, and “SKIN & SCAR DETAIL”. Show gnarled fingers gripping tobacco, a round leather pouch, tangled leaf wrappings, rope knots, cracked dirty feet, and scarred bark-like skin texture.

Right-side notes and props: On the right, add an “IDENTITY NOTES” text block with the following readable content: Role: Nocturnal forest spirit. Traits: silent, cunning, territorial, primal, elusive, feared. Core theme: Paraguayan rural folklore, night, superstition, nature, fear, respect. Signature feature: silent feet and piercing gaze. Below the attitude pose, add exactly 1 key prop section titled “KEY PROP – RITUAL TOBACCO”, showing a large hand-rolled tobacco bundle tied with fiber beside dry leaves, with a caption explaining that hand-rolled tobacco is his offering, companion, and mark of his presence.

Bottom-right studies: Include exactly 4 black silhouette thumbnails titled “SILHOUETTE STUDY”, showing different readable poses of the same ragged, hat-wearing creature. Include exactly 7 vertical color swatches titled “COLOR PALETTE”, labeled left to right: “BARK GRAY”, “MONTE GREEN”, “TOBACCO BROWN”, “EARTH UMBER”, “DRY LEAF BEIGE”, “NIGHT CHARCOAL”, and “BLACK”.

Visual style: Hyper-detailed cinematic creature concept art, earthy brown-gray-green palette, photorealistic anatomy mixed with folklore monster design, weathered textures, subtle shadows under each figure, crisp labels, thin black rules, organized like a studio character design board. Avoid bright colors, cute proportions, modern clothing, weapons, logos, watermarks, or extra panels beyond the specified counts.
Product Marketing - Axiom Cat-Eye Fashion Cover — output

Product Marketing - Axiom Cat-Eye Fashion Cover

VoxCatGPT Image 2
Goal: Create a vertical avant-garde fashion magazine cover featuring {argument name="character name" default="a mysterious cat-themed intelligence heroine"} as the central subject, blending minimalist high-fashion editorial portraiture, retro Japanese anime poster energy, and independent art-project key visual design.

Canvas: Vertical magazine cover, 4:5 aspect ratio, full bleed. Deep teal-blue sky background with warm orange clouds at the lower left and lower right edges. Use a strong teal-and-orange color grade, dramatic rim lighting, hard graphic shadows, and a vintage print-poster feel.

Main subject: One anime-style female character, shown from an extreme low-angle perspective so she feels monumental and presses outward from the page. She stands in a tense, stylish JoJo-like fashion pose with one hand in her trouser pocket, torso leaning slightly back, shoulders angled, and head tilted downward toward the viewer. Her face is intentionally obscured by a soft rectangular blur or shadow block, keeping her identity mysterious. She has short black bobbed hair with orange rim highlights, visible earrings, and a cool, intelligent, slightly dangerous aura.

Outfit and symbols: Include exactly 5 main outfit pieces: an oversized cream trench coat with dramatic lapels, a black fitted top, high-waisted cream trousers, a black belt with a rectangular buckle, and small dark earrings. Add exactly 1 signature accessory system: a cat-eye bracelet on her wrist, designed as the character’s identifying visual symbol rather than simple jewelry. The bracelet should have exactly 2 visible dangling charms: one orange cat-eye amulet charm and one small dark spherical charm, connected by a gold chain.

Cover layout: Place a huge elegant serif masthead across the top reading {argument name="magazine title" default="AXIOM"} in warm orange, partially behind the character’s head and hair. Keep the typography sparse, refined, and fashion-magazine-like, with plenty of negative space. The character should overlap the masthead, reinforcing cover logic.

Visible text content: Include exactly 8 text groups on the cover: 1) top-left tiny text: “美学・反逆・未来”; 2) top-right tiny text: “VOL.09 / AUTUMN 2025”; 3) giant masthead: “AXIOM”; 4) left mid large Japanese character: “猫”; 5) left mid small English/Japanese lines: “WIT. INSTINCT. REVOLUTION.” and “本能は、真実を見抜く。”; 6) right mid small English/Japanese lines: “INTELLIGENCE IS THE ULTIMATE WEAPON.” and “知性こそ、最強の武器。”; 7) bottom-left issue block: “THE CLEVER ISSUE” and “妙案は、静かに仕掛ける。”; 8) bottom-right large issue number and caption: {argument name="issue number" default="09"} with “SHADOW GAME” and “BRIGHT MIND”.

Visual style: Japanese old-anime-inspired character design, fashion editorial cover composition, cinematic low-angle lighting, sharp inked contours, painterly cel-shaded folds, warm orange highlights, deep teal shadows, elegant serif typography, clean luxury-magazine spacing, retro poster grain.

Constraints: Use exactly one central character, exactly one cat-eye bracelet, exactly two bracelet charms, and exactly eight visible text groups. Do not add extra cover lines, logos, barcodes, watermarks, or additional characters. Keep the image stylish and symbolic rather than overly detailed or realistic.
Comic / Storyboard - Horror Wraith Character Sheet — output

Comic / Storyboard - Horror Wraith Character Sheet

Rafa.ioGPT Image 2
Goal: Create a horror character concept sheet for a ghostly female creature named {argument name="character name" default="unnamed wraith"}, designed for film or game production.

Canvas: Wide horizontal character turnaround sheet on an off-white studio background, realistic dark fantasy concept art, high-detail texture study, evenly lit with soft shadows and no environment.

Layout: Show exactly 4 labeled views arranged left to right: 1) FRONT full-body view, 2) SIDE full-body profile view, 3) BACK full-body view, 4) CLOSE-UP portrait view occupying the right third of the canvas. Place small uppercase serif labels above each view: “FRONT”, “SIDE”, “BACK”, and “CLOSE-UP”.

Subject details: The character is a tall, emaciated, corpse-like female apparition with very long tangled {argument name="hair color" default="wet black hair"} falling past the chest and down the back. Her skin is gray, deathly, and dirty, with thin blackened hands and bare feet. She wears a floor-length shredded {argument name="garment color" default="ash-gray"} burial gown made of rotting gauze, torn lace, and ragged strips, full of holes, frayed edges, dangling fibers, grime, stains, and smoky wisps. The silhouette should feel fragile, decayed, damp, and supernatural.

Face treatment: Obscure the face in every visible view with a plain dark rectangular censor block, creating an anonymous unsettling look. The close-up should show the head, hair, neck, and shoulders in extreme detail, with a large dark rectangle covering the face.

Visual style: Photorealistic cinematic creature-design sheet, monochrome gray-black palette, high-resolution fabric decay, wet hair strands, fine scratchy linework, subtle misty edges around the torn gown, eerie but clean production-art presentation.

Constraints: Include exactly 4 views and exactly 4 labels. Keep the background blank and neutral. Do not add extra props, weapons, scenery, logos, watermark, or additional text.
E-commerce Main Image - Mythic Yerba Mate Product Shot — output

E-commerce Main Image - Mythic Yerba Mate Product Shot

Rafa.ioGPT Image 2
Create a cinematic premium product advertising image for {argument name="brand name" default="MyThos"}, a luxury yerba mate package design inspired by Paraguayan forest legends. Scene: a dark rustic wooden tabletop against a smoky warm brown studio backdrop, dramatic low-key lighting, soft shadows, amber highlights, hyper-realistic commercial photography. Arrange exactly 7 visible product objects from left to right: 1 large front-facing black matte yerba mate bag, 1 narrow side-facing black bag showing side branding, 1 rear-facing black bag showing nutrition facts and barcode, 1 three-quarter angled black bag showing the front and right side, 1 tall green Stanley-style thermos with stainless steel cap on the far right, 1 dark worn mate gourd in front of the thermos filled with loose yerba, and 1 curved metal bombilla straw emerging from the gourd. Packaging details: all bags are glossy crinkled black kraft pouches with gold foil typography, ornate botanical borders, small tree emblem at the top, subtle forest leaves near the bottom, and mythical guardian character art. The main front bag should prominently read "{argument name="product name" default="MyThos"}" in large elegant serif gold lettering, with Spanish text "YERBA MATE", "ELABORADA CON PALO", and the slogan "{argument name="slogan" default="LEYENDAS QUE VIVEN"}". The front artwork shows exactly 2 legendary forest guardian figures, one straw-covered Pombero-like figure and one long-haired pale Pora-like figure, standing in a dark jungle scene; include a round gold seal near the bottom and the label "POMBERO & PÓRA" plus "GUARDIANES DEL MONTE". The back-facing bag includes a centered nutrition information panel, exactly 3 small circular claim icons above it, paragraph text, a barcode at the bottom, and gold ornamental trim. The side-facing bag includes vertical layout branding, a small tree icon, short descriptive Spanish copy, and "POMBERO & PÓRA" near the bottom. Color palette: black, deep forest green, aged bronze, antique gold, warm brown wood, subtle cream text. Style: ultra-detailed, photorealistic, premium beverage packaging mockup, sharp focus on products, shallow depth of field, no people, no hands, no extra props, no watermark. Use {argument name="package weight" default="1kg"} on the front lower corner and make the thermos label read {argument name="thermos label" default="STANLEY SINCE 1913"}.
Social Media Post - Rainy Night Event Street Fashion — output

Social Media Post - Rainy Night Event Street Fashion

AnionAIGPT Image 2
Create a candid full-body street-event photograph at night of a fashionable {argument name="character description" default="young East Asian woman"} walking past a registration or security table outside a modern glass building. She is captured in profile with her body moving to the right while her head turns back toward the camera, giving a slightly surprised, neutral expression. She has long dark hair tied loosely in a low ponytail with stray strands around her face, natural makeup, and a slim silhouette. Outfit: a fitted {argument name="outfit color" default="white"} blazer, matching short pencil skirt, sheer pale tights, and glossy white pointed-toe stiletto heels; include a small light shoulder bag with a chain strap partly visible at her side. Environment: rainy urban night, wet reflective stone pavement, metal crowd-control barriers, a black check-in table on the right, event staff wearing dark clothing and blue lanyards handling wristbands, and a blue screen/sign in the background showing an AI-company-style logo and the text {argument name="screen text" default="AnionAI"}. Add a crowd of people waiting in the dim background, blurred city lights and signage, strong on-camera flash lighting on the woman, high ISO documentary realism, vertical smartphone-like composition, 2:3 aspect ratio, slightly imperfect candid framing, realistic skin texture, realistic fabric folds, no glamour retouching, no studio lighting, no illustration style.
YouTube Thumbnail - Night Train Sleeper Selfie Thumbnail — output

YouTube Thumbnail - Night Train Sleeper Selfie Thumbnail

Marco Schneekluth ⚡🧡GPT Image 2
Create a photorealistic 16:9 YouTube thumbnail-style wide-angle selfie taken inside an overnight train sleeper compartment at night. In the right foreground, show one young adult man with {argument name="hair color" default="light blond-brown"} messy hair wearing a plain black T-shirt and one black smartwatch, holding his hand on top of his head in a stressed or surprised pose; his face is deliberately hidden by one large opaque vertical rectangle in a warm medium-brown censor color, covering the entire face area from forehead to chin. The left half of the image shows the open doorway and window of a compact sleeper cabin with beige wood paneling, metal door frame, bright white interior lighting, a red/orange bed or blanket, white pillows, and small travel items on the bunk. Outside the train on the far left, show exactly one red ÖBB passenger train speeding past on parallel tracks, with motion blur, glowing carriage windows, the white “ÖBB” logo visible, dark mountains in the background, and a bright full moon in a cloudy deep-blue night sky. Use dramatic ultra-wide perspective, strong contrast between warm cabin light and cool night exterior, realistic lens distortion, crisp details on the interior and person, motion blur only on the passing train and tracks, no extra text overlays, no watermark.
Profile / Avatar - Night Street Fashion Selfie — output

Profile / Avatar - Night Street Fashion Selfie

500美金炒美股GPT Image 2
Create a realistic nighttime smartphone selfie photo of an adult woman leaning against a red brick wall on a city sidewalk in autumn. She has long wavy black hair, light-to-medium skin, a slim build, and her face is intentionally covered by a centered opaque square privacy block in a dark brown skin-tone color. She is wearing a dark charcoal ribbed sleeveless cropped tank top, high-waisted black denim shorts, and a glossy black shoulder bag with the strap over one shoulder. Pose her at a slight diagonal angle with one arm extended toward the camera as if taking the selfie and the other hand resting near her shorts pocket, torso cropped from upper head to mid-thigh. The background shows a nighttime urban street with large trees, scattered yellow fallen leaves on the pavement, warm streetlights, blurred car headlights, and distant city lights with shallow depth of field. Use realistic phone-camera flash lighting, natural shadows, detailed brick texture, candid street-fashion mood, vertical 2:3 composition, no text, no watermark. Customize the outfit color as {argument name="top color" default="dark charcoal"}, the hairstyle as {argument name="hair style" default="long wavy black hair"}, the setting as {argument name="street setting" default="autumn nighttime city sidewalk"}, and the privacy block color as {argument name="face block color" default="dark brown skin-tone"}.
Profile / Avatar - Rain Glass Fashion Portrait — output

Profile / Avatar - Rain Glass Fashion Portrait

sereinGPT Image 2
Create a moody fine-art fashion portrait of an adult woman photographed through rain-streaked glass, with a soft gray concrete wall behind her and the feeling of being enclosed by rain, glass, and time. The subject has {argument name="hair color" default="wet black hair"} hanging messily around her face and shoulders, a centered seated pose, bare shoulders, a {argument name="top color" default="black"} deep V camisole, loose white trousers, and an oversized {argument name="shirt color" default="white"} button-up shirt slipped off both shoulders and bunched around her arms. Her face is intentionally obscured by a soft rectangular blur/censor block, while the body and clothing remain sharp enough to read. Shoot vertically in a realistic editorial photography style, muted monochrome palette, cool gray and white tones, soft diffused daylight from the left, high contrast between wet dark hair and pale wall, translucent water trails and droplets running down the glass in front of the camera, subtle reflections, hazy atmosphere, delicate skin highlights, quiet melancholic mood, like a photograph slowly turning into a memory. Use a 4:5 portrait composition, subject centered from thighs to head, minimal background, no text, no watermark, no extra people.
Comic / Storyboard - Cybernetic Woman in Cryo Tank — output

Comic / Storyboard - Cybernetic Woman in Cryo Tank

ĴƦƱƓƓƑƱȻĦŦGPT Image 2
Create a cinematic vertical sci-fi scene of {argument name="character name" default="LATVIA-01"}, a female humanoid bio-synthesis android suspended in a transparent cylindrical cryogenic tank filled with glowing aqua liquid. She is curled in a fetal-like floating pose with legs bent, one hand raised against the glass and the other arm crossing her chest, long pale blond hair drifting weightlessly in the fluid. Her body is mostly human but partially biomechanical: visible synthetic muscles, exposed cybernetic shoulder joint, metallic chest structure, cables, and translucent skin details, with elegant anatomy and no explicit nudity. The tank is covered with condensation, bubbles, scratches, and refracted light, capped by heavy black metal rings at top and bottom, with hoses and connectors attached. Bright cyan energy ribbons spiral upward around her body inside the chamber, casting teal highlights. Surround the tank with a dark futuristic laboratory: robotic arms on both sides, hanging cables, control consoles, glowing monitors, and one left-side holographic status screen reading “{argument name="status label" default="LATVIA-01 // BIO SYNTHESIS // STATUS: ACTIVE"}”. Use a moody high-contrast cyberpunk color palette of black, steel gray, turquoise, and cyan; hyper-detailed photorealistic digital art, dramatic rim lighting, volumetric glow, water caustics, cinematic depth, sharp focus on the glass pod, vertical 9:16 composition. The subject’s face should be hidden or obscured by a neutral rectangular censor-like blur, preserving the rest of the image in high detail. Avoid gore, avoid extra characters, avoid logos or watermarks.
Comic / Storyboard - Anime Boy at Night Window — output

Comic / Storyboard - Anime Boy at Night Window

薛元煌GPT Image 2
Create a cinematic anime-style night scene in a horizontal 3:2 composition. A quiet teenage boy with {argument name="hair color" default="messy dark brown hair"} sits in profile on a wide wooden window ledge inside a dim room, wearing a loose dark hoodie and pants, holding a warm mug with both hands and looking out through an open wooden window. His face is softly shadowed and not detailed. On the ledge near him are exactly 3 visible objects: an open book lying flat in the foreground, a small glowing lantern on the left, and the mug in his hands. Outside, show a sprawling hillside town at night with old stone houses and red-tiled roofs, warm lit windows, low garden walls, trees, and a wet winding road that starts near the foreground and snakes down through the town toward a distant bay or lake between dark mountains. Include exactly 5 prominent street lamps: one tall glowing lamp on the far right foreground, three smaller lamps following the road through the middle distance, and one faint lamp farther down the road. The sky is deep navy blue with scattered stars, layered clouds, and a thin orange glow on the horizon after sunset. Use rich warm-and-cool contrast: golden lamplight and window glow reflecting on the rain-slick road and polished wood, against cool blue moonlit mountains and sky. Highly detailed painterly anime background art, cozy melancholic atmosphere, realistic perspective depth, soft bloom around lights, no text, no watermark, no modern vehicles or crowds.
Product Marketing - Glass Shard Dance Action Portrait — output

Product Marketing - Glass Shard Dance Action Portrait

AijazGPT Image 2
Ultra-realistic dynamic dance action portrait of a female street dancer performing an energetic hip-hop move, captured mid-motion in a minimalist studio with a clean white-gray backdrop. Use the exact same face and facial features from the reference image without alteration. She wears a {argument name="outfit color" default="vibrant green"} cropped athletic top, loose black sweatpants, colorful high-top sneakers, and blue wristbands. Powerful urban dance aesthetic with dramatic body posture, intense movement energy, and freeze-frame action photography style. Parts of her body explosively disintegrate into thousands of floating {argument name="shards material" default="glass-like shards"}, reflective metallic fragments, and crystal particles scattering through the air with realistic debris physics and cinematic motion blur. Sharp focus on the face and torso, ultra-detailed skin pores, realistic fabric textures, dynamic lighting with soft studio shadows, high contrast, modern VFX-inspired composition, reflective floating fragments, photorealistic 8K masterpiece, cinematic realism, hyper-detailed action scene, depth of field, premium fashion sports editorial look.
Comic / Storyboard - Dreamlike Fantasy Forest Boy — output

Comic / Storyboard - Dreamlike Fantasy Forest Boy

SnowGPT Image 2
Dreamlike fantasy world with giant glowing flowers and floating particles, soft pastel clouds, cinematic sunlight rays, magical atmosphere, {argument name="character" default="ethereal boy wearing elegant flowing clothes"}, emotional peaceful mood, surreal realism, ultra detailed fantasy art, volumetric lighting, cinematic composition
Profile / Avatar - Perler Bead Character Transformation — output

Profile / Avatar - Perler Bead Character Transformation

Mr GafishGPT Image 2
Draw the main character from the image in a {argument name=\"style\" default=\"Perler bead style\"}, with colors close to the effect in the picture. Change the background to {argument name=\"background\" default=\"relatively simple decorations\"}, making the overall image feel warm.
Product Marketing - Rebel Urban Skateboard Advertisement Poster — output

Product Marketing - Rebel Urban Skateboard Advertisement Poster

𝐌GPT Image 2
A photorealistic youth streetwear advertisement poster featuring a teenage boy wearing baggy cargo pants and a graphic tee while performing a dynamic kickflip beside a giant vertical skateboard nearly 3x his height, with the word “REBEL” boldly painted in graffiti style across the skateboard deck. The scene is set in a gritty urban environment with rough concrete textures, spray paint splatter art, and an underground street culture vibe. Massive distressed graffiti typography reading “REBEL” dominates the background walls, adding a rebellious energy to the composition. At the bottom, a bold tagline reads “Rules are made to break.” while small grey text in the top-right corner says “Designed with GPT Image 2”. Shot in a raw street photography style with harsh shadow lighting, cinematic contrast, realistic textures, and ultra-detailed photorealism.
Social Media Post - Funny Ugly Doodle Drawing Transformation — output

Social Media Post - Funny Ugly Doodle Drawing Transformation

mmtvr (studying docs arc)GPT Image 2
Turn this photo into a funny ugly doodle drawing. Make it look like: a quick sketch using a cheap marker or crayon messy, rough, childlike style bad perspective and awkward proportions slightly exaggerated facial features Add: simple cartoon background (like buildings, trees, street) random sketchy lines and details uneven coloring and visible strokes Style: looks like a lazy drawing, not polished humorous and a bit stupid-looking meme-like, casual, internet style Do NOT: make it realistic
Comic / Storyboard - Neo-Noir Character Design Board — output

Comic / Storyboard - Neo-Noir Character Design Board

Cherry 2.OGPT Image 2
Create a cinematic realistic character design board for a high-budget neo-noir film production set in a rain-soaked futuristic city.

Use a dark charcoal and electric cyan color palette with neon reflections in the background.
Avoid generic grids or symmetrical layouts; composition should feel like a stylized director’s pitch board.
Design a grounded human character with realistic anatomy, subtle imperfections, and strong emotional presence.
Include full-body turnarounds, expressive head angles, cinematic portrait, wardrobe breakdown, fabric texture detail, and production notes.
Background: blurred cyberpunk city lights, wet glass reflections, moody atmosphere, soft neon glow.
Style: semi-realistic cinematic realism, high contrast lighting, shallow depth of field, film grain, emotional intensity.
Product Marketing - Premium Graphic Design Portfolio Mockup — output

Product Marketing - Premium Graphic Design Portfolio Mockup

Abkr SadiqGPT Image 2
Create a premium graphic design portfolio mockup in a professional creative studio style. Show a clean, elegant workspace presentation featuring multiple graphic design project cards arranged in a refined grid layout, a modern laptop screen displaying a portfolio homepage, and several printed posters laid out neatly on the desk and mounted on the wall.

Use a polished creative-director aesthetic with soft shadows, subtle depth, realistic materials, and a high-end studio atmosphere. Include sleek typography placeholders, branding samples, editorial-style poster designs, business cards, stationery, notebook, pen, coffee cup, desk lamp, and small plant for a realistic studio setup.

The design should have a clean grid layout, balanced spacing, soft realistic shadows, premium lighting, and neutral luxury tones such as black, white, warm beige, gray, champagne, and cream. Make the laptop screen show a minimal portfolio interface with project thumbnails, navigation menu, and bold hero text.

Add multiple portfolio project cards such as logo design, brand identity, poster design, packaging design, social media design, UI/UX design, and advertising campaign mockups. Arrange everything professionally with a modern minimalist style, sharp details, clean typography, elegant composition, and a premium creative agency presentation look.

Style: high-end graphic design portfolio, modern creative studio, luxury branding mockup, clean desk setup, realistic shadows, editorial layout, minimalist premium design, professional presentation, 4K, ultra-detailed.
Profile / Avatar - Vintage Classroom Cinematic Portrait — output

Profile / Avatar - Vintage Classroom Cinematic Portrait

KarlosGPT Image 2
Cinematic portrait of the man in the picture (preserve 100% facial details of this man), vintage classroom. The camera is shot from a distance, with a small section of the chair's back visible in front, as if the photographer was taking the portrait discreetly or in a quiet moment, Soft, gol-den-yellow lighting reflects from a small window or gap in the right wall, forming diagonal lines of light that fall directly on the man's face, creating a dramatic and emotional contrast.The man sits alone in a chair, his feet resting on the desk, with a relaxed and cool expression, as if lost in thought.
Profile / Avatar - 1990s Flash Camera Snapshot Portrait — output

Profile / Avatar - 1990s Flash Camera Snapshot Portrait

BMXGPT Image 2
Use the uploaded image as the main reference. 

Transform the uploaded photo into a realistic candid portrait with an early 1990s digital camera aesthetic. Preserve the subject’s identity, facial features, pose, outfit, and overall composition, but restyle the image with harsh blown-out flash highlights, subtle red-eye effect, low-resolution image quality, raw snapshot imperfections, nostalgic flash-filter styling, and a vintage timestamp look.

The final image should feel candid, imperfect, and authentic, like an old retro party or personal snapshot taken with an early consumer digital camera. Keep the background dark or naturally subdued when appropriate, maintain a direct-flash look, and give the image a raw, unpolished, memory-like atmosphere.

Include:
- harsh direct camera flash
- overexposed or blown-out highlights
- subtle red-eye effect
- low-resolution / soft digital detail
- slight grain or noisy texture
- authentic retro snapshot feeling
- vintage date/timestamp aesthetic in one corner
- candid, natural, imperfect energy

Avoid:
- cartoon or anime style
- overly polished beauty retouching
- studio lighting
- ultra-sharp modern DSLR look
- glossy AI skin
- text, logos, watermarks, or graphic overlays other than the timestamp aesthetic
- distorted anatomy or altered identity

Make the aspect ratio 3:4
Comic / Storyboard - Retro Heisei Purikura Photo Strip — output

Comic / Storyboard - Retro Heisei Purikura Photo Strip

AniroleGPT Image 2
Generate a Heisei-era retro {argument name="style" default="purikura-style strip image"} of these characters making {argument name="expression" default="funny faces"}. Create a vertical 4-panel layout with {argument name="expression" default="funny faces"} in every frame. Include different expressions in each frame, with a slightly overexposed flash effect, nostalgic grainy quality, flashy graffiti, heart and star stamps, decorative text, and short Japanese messages to capture the 'uncool-but-cute' atmosphere of old photo booths.
按模型、风格或主题浏览

寻找下一个方向

ASTORIE CANVAS

让灵感,继续生长。

在同一个创作空间,连接提示词、参考素材和生成结果。

打开创作画布 ↗