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

9,951 条提示词

Comic / Storyboard - Anime Tactical Operator Character Sheet — output

Comic / Storyboard - Anime Tactical Operator Character Sheet

ゴードン・シャムウェイ@AIスロップおじさんGPT Image 2
Goal: Create a clean anime character design sheet showing {argument name="character name" default="a young female tactical operator"} in two full-body views, emphasizing asymmetric tactical gear, realistic fabric folds, and detailed military equipment.

Canvas: Tall vertical canvas, plain off-white studio background, no scenery, no text, no watermark. The character appears twice side by side: exactly 2 full-body poses.

Layout: Left figure is a front three-quarter view, facing slightly right, standing with weight on one leg and holding a compact black assault rifle diagonally across her torso. Right figure is a rear three-quarter view, facing right in profile, showing the back of the vest, belt, pouches, and long ponytail while holding the same rifle low across the front of her body. Both figures are the same scale and aligned at the feet.

Subject details: A slim athletic anime woman with fair skin, sharp purple eyes, a calm serious expression, and very long {argument name="hair color" default="deep violet purple"} hair tied in a high ponytail with a dark scrunchie. Include long side bangs, layered fringe, glossy hair highlights, loose strands, and two upward ahoge-like hair wisps. Her face is youthful but mature, with crisp anime linework and subtle painterly shading.

Outfit and gear: Use {argument name="camouflage pattern" default="dark tiger-stripe camouflage"} on the jacket and pants. She wears exactly 8 visible tactical outfit/gear categories: 1 camo combat jacket, 2 matching camo cargo pants, 3 olive plate carrier vest, 4 olive tactical belt, 5 olive gloves, 6 tan lace-up combat boots, 7 thigh-mounted pouch/holster rig, 8 multiple small olive utility pouches on the vest and belt. Add ribbed dark green socks visible above the boots. The vest should have shoulder straps, MOLLE-like webbing, chest panels, rear pouches, and buckles. The pants should be tucked into socks/boots with realistic creases.

Weapons: Show exactly 2 visible instances of a black compact assault rifle, one per character view. Each rifle should have a short barrel, suppressor-like muzzle device, rail system, foregrip, magazine, tactical optic, and angular mechanical details. Keep the rifle believable but stylized for anime illustration.

Visual style: High-detail Japanese anime military illustration, polished digital painting, clean sharp outlines, soft cel shading mixed with semi-realistic rendering, muted olive, khaki, black, and tan palette contrasted with vivid purple hair. Full-body character sheet presentation, accurate proportions, detailed boots and tactical gear, crisp fabric texture and camouflage pattern.

Constraints: No extra characters, no background objects, no text labels, no logos, no cropped feet, no exaggerated fantasy armor. Keep both views consistent in hair, clothing, pouches, boots, and rifle design.
使用此提示词 ↗
Social Media Post - Pastel Miku Birthday Garden — output

Social Media Post - Pastel Miku Birthday Garden

しあぱんGPT Image 2
Create a dreamy pastel anime birthday illustration of {argument name="character name" default="Hatsune Miku"} seen from behind on a sunlit flower-filled balcony garden. She stands slightly left of center in a white off-shoulder ankle-length dress with a pale aqua ribbon sash, holding a large bouquet of pink, white, and lavender flowers close to her chest. Give her extremely long flowing twin-tails in {argument name="hair color" default="turquoise blue"}, trailing dramatically to the right in the breeze with many delicate strands and translucent highlights. The scene is soft, luminous, and ethereal, with a bright blue sky, warm sunlight from the upper right, lens flare, floating flower petals, bokeh, and a gentle celebratory mood. Include abundant pastel flowers: a floral arch or hanging blossoms in the upper left, a small round table at lower left with a white birthday cake topped by lit candles and tied with an aqua ribbon, a white balcony railing in the middle distance, and large blurred pink and purple foreground flowers on the right for depth. Use a high-detail polished 3D-anime / Vocaloid fan-art look, delicate fabric motion, soft bloom, rim lighting, shallow depth of field, and a pale pink-blue-white color palette. Add glowing handwritten pink text along the lower left and bottom center reading exactly: “Happy Birthday, Miku! You bring me more happiness than words can say, and I hope just as much finds its way back to you. — Aug. 31, 2026”. Keep the text integrated like a soft neon signature, legible but dreamy. Wide cinematic 16:9 composition, no watermark, no extra characters.
使用此提示词 ↗
E-commerce Main Image - 3D Footstool Mesh Comparison — output

E-commerce Main Image - 3D Footstool Mesh Comparison

KOBATAKA|Vibe ModelingGPT Image 2
Create a realistic architectural visualization comparing two 3D-generated ottoman stools in an interior showroom scene. Canvas: wide landscape 16:9 render, camera at low eye level looking slightly downward. Scene: two small Barcelona-style footstools placed side by side on a dark charcoal tiled floor, in front of a tall glossy stone wall with dramatic brown, cream, rust, and gold onyx/marble veining; a narrow strip of green-tinted glass or window appears at the far right edge, with pale exterior floor tiles visible beyond. Subjects: exactly 2 footstools: left stool is the HD version, slightly smaller and lower, with a white tufted square cushion divided into a 4-by-4 grid, rounded soft edges, light gray leather straps underneath, and reflective polished metal X-shaped legs; right stool is the Smart Mesh version, slightly larger and cleaner, with a thicker smooth white cushion, parallel white underside straps, and gray polished X-shaped metal legs. Add thin black 3D viewport guide lines or wireframe axis lines crossing through and around both stools, including vertical and horizontal alignment lines, as if shown in a 3D modeling comparison viewport. Materials: white leather cushions with subtle seams and soft highlights, chrome legs with reflections, dark matte floor tiles with faint grout lines, glossy veined stone wall. Lighting: soft indoor studio lighting, moderate shadows, realistic but slightly CGI-rendered look. Composition: left stool around one-third of the frame, right stool around two-thirds, both facing forward at a slight angle, enough floor visible in front, marble wall filling the background. No people, no labels, no UI panels, no watermark.
使用此提示词 ↗
Social Media Post - AI Thinking Before Drawing Banner — output

Social Media Post - AI Thinking Before Drawing Banner

john@macareno.netGPT Image 2
Goal: Create a cinematic horizontal website/social banner about AI image generation and accurate text rendering, with a moody studio atmosphere and clear Portuguese typography.

Canvas: Wide 2:1 banner, approximately 1200×600 px. Dark, soft-focus photographic background with strong depth of field. Leave the left half for text and the right half for the visual metaphor.

Background and subject: A dim artist studio with cool gray-blue lighting, a blurred window on the far left, concrete wall, soft shadows on the floor, and an easel on the right holding a blank white canvas. In the upper center-right, show a delicate mechanical drawing arm or robotic brush holder reaching toward the canvas, holding a thin brush that has a single dark blue paint droplet at its tip. The canvas should be bright white and mostly empty, slightly angled, with a subtle shadow. Use realistic lighting, shallow depth of field, and a premium editorial tech aesthetic.

Layout: Place all main text on the left side over a dark translucent gradient overlay for legibility. Use exactly 3 small rounded pill tags above the headline, aligned horizontally: “IA”, “MUSE IMAGE”, and “NOVIDADE”. Below them, place a large bold headline in two lines: “A IA aprendeu a pensar” on the first line and “antes de desenhar” on the second line. Highlight only the word “pensar” with a vivid purple rectangular label behind the word, white text on purple. Under the headline, add one smaller italic subtitle: “O texto deixou de ser o calcanhar de Aquiles. Falta todo o resto.” Add a small website URL in the bottom-right corner: “www.macareno.net”.

Text content: Use {argument name="headline text" default="A IA aprendeu a pensar antes de desenhar"}, with the word {argument name="highlighted word" default="pensar"} highlighted in purple. Use {argument name="subtitle text" default="O texto deixou de ser o calcanhar de Aquiles. Falta todo o resto."}. Use {argument name="tag labels" default="IA, MUSE IMAGE, NOVIDADE"}. Use {argument name="website URL" default="www.macareno.net"}.

Visual style: Minimal, polished, cinematic, realistic, high contrast between white typography and dark background, with a modern tech-news landing-page feel. Typography should be clean sans serif; headline extra bold, subtitle light italic, tags uppercase with letter spacing. Keep text sharp and perfectly spelled.

Constraints: Include exactly 3 pill tags, exactly 1 highlighted word, exactly 1 subtitle line, exactly 1 URL, exactly 1 easel canvas, exactly 1 robotic drawing arm, exactly 1 brush, and exactly 1 visible paint droplet. Do not add extra logos, people, icons, or additional text.
使用此提示词 ↗
Social Media Post - Underwater Red Dress Fantasy Portrait — output

Social Media Post - Underwater Red Dress Fantasy Portrait

SnehaGPT Image 2
Create a vertical cinematic underwater fantasy portrait of a young woman floating upright in a tropical ocean, centered in frame, wearing a flowing translucent {argument name="dress color" default="deep crimson red"} gown with long sheer sleeves, a wide swirling skirt, and a matching scarf or veil billowing dramatically to both sides. She has {argument name="hair color" default="dark brown"} loose wavy hair drifting in the water, a calm direct gaze, natural makeup, and bare feet visible beneath the dress. Add ornate gold embroidery along the neckline and sleeve edges, with realistic underwater caustic light patterns shimmering across the fabric and skin. The scene is lit from the bright sunlit water surface at the top, with strong shafts of light, bubbles rising around her, and deep teal-blue water fading into the background. Include exactly 3 large pink translucent jellyfish: one large jellyfish at upper left, one large jellyfish at upper right, and one smaller jellyfish behind her on the left; add a few tiny distant bubble-like jellyfish near the top. The bottom of the image shows a vibrant coral reef with many corals and exactly 7 clearly visible tropical fish: one large striped butterflyfish at lower left foreground, one orange-blue striped fish at far lower left, one black-white-yellow fish peeking from the far left edge, one clownfish at lower right foreground, one small yellow fish near the clownfish, one yellow-blue butterflyfish at mid-right, and one tiny orange fish at the right edge. Use photorealistic digital art, highly detailed fabric physics, realistic refraction, suspended particles, soft depth of field, dramatic fantasy editorial composition, 9:16 aspect ratio, no text, no watermark.
使用此提示词 ↗
Profile / Avatar - Anime Selfie on Smartphone — output

Profile / Avatar - Anime Selfie on Smartphone

Aziz Cor XGPT Image 2
Create a stylized anime screenshot displayed inside a modern black smartphone screen, as if viewing a social media story or selfie. The phone fills the frame at a slight diagonal angle, with rounded black bezels, a top notch, small status icons, and the time shown as 11:27. Inside the phone screen is a close-up overhead selfie of a smiling young woman with {argument name="skin tone" default="deep warm brown skin"}, {argument name="hair color" default="long straight black hair"}, large glossy brown anime eyes, defined brows, and a bright friendly grin. She is holding the camera above herself with one arm extended, creating a dynamic wide-angle selfie perspective. She wears a fitted black cropped T-shirt with a white leaf spiral emblem on the chest and dark pants. The setting behind her is a softly blurred anime village street with tan stone pavement, wooden storefronts, greenery at the side, and an orange vertical sign bearing the Japanese character 「忍」. Add a translucent black caption bar across the lower third of the phone screen with centered white text reading {argument name="caption text" default="Good vibes only 🌿✨✨"}. Use polished high-quality anime rendering, clean linework, warm cinematic shading, expressive face details, subtle screen glare, shallow depth of field, and a vertical smartphone composition. Keep the image focused on exactly one person, one phone, one caption bar, one visible chest emblem, and one orange background sign.
使用此提示词 ↗
Infographic / Edu Visual - Vintage Macau Map Poster — output
Comic / Storyboard - Mars Colony Crisis Storyboard — output

Comic / Storyboard - Mars Colony Crisis Storyboard

Mystique Licorne 🇨🇦GPT Image 2
Goal: Create a cinematic sci-fi storyboard contact sheet showing a tense emergency inside a Mars colony during a red dust storm, in the style of high-budget film stills with anamorphic widescreen framing.

Canvas: A single 16:9 horizontal image divided into exactly 8 rectangular widescreen panels arranged in a clean 2-column by 4-row grid. Use thin white dividers between panels and no captions, no logos, no visible text except tiny incidental signage.

Setting and mood: {argument name="setting" default="a domed human colony on Mars during a violent red dust storm"}. The entire sheet should feel dark, dramatic, and cinematic, with heavy red-orange atmosphere, emergency lighting, volumetric dust, glass reflections, and realistic production design.

Panel count and layout: Include exactly 8 distinct frames:
1. Top-left: Wide exterior establishing shot of a futuristic city or research base protected by a transparent dome on a barren Martian landscape, distant towers glowing through a massive red storm cloud.
2. Top-right: Close-up silhouette of a large human hand pressing against or emerging behind a dusty translucent surface, almost abstract, surrounded by swirling red dust.
3. Second row left: Symmetrical view down a cylindrical metal tunnel or airlock corridor with circular ribs, red emergency lights, and small suited figures far in the distance.
4. Second row right: Close-up of {argument name="main character" default="a frightened astronaut"} inside a helmet visor, face lit by red alarms and a bright headlamp flare, with a blurred figure running in the background.
5. Third row left: Interior biodome or greenhouse walkway with trees and dense vegetation under orange emergency haze, railings in the foreground, a doorway at the right edge.
6. Third row right: Two astronauts in bulky suits running down a dark industrial corridor toward a group ahead, red alarm lights and motion blur emphasizing urgency.
7. Bottom-left: Low-angle view under a curved glass or transparent dome ceiling with red warning lights, flying debris, scratches, cracks, and storm particles battering the surface.
8. Bottom-right: Close-up from above or through glass of a dark human hand pressed flat against a dusty transparent panel, with faint cracks, condensation, and glowing red reflections around it.

Visual style: Photorealistic cinematic stills, gritty hard sci-fi, realistic astronauts and architecture, shallow depth of field in close-ups, strong contrast, smoky atmosphere, anamorphic lens flares, subtle film grain, deep blacks, saturated rust-red palette with occasional white headlamp highlights.

Constraints: Maintain exact 8-panel grid, no comic speech bubbles, no explanatory labels, no extra frames, no cartoon style. Make every panel feel like a different shot from the same suspenseful film sequence about {argument name="story event" default="a breach or crisis threatening the Mars colony"}.
使用此提示词 ↗
Comic / Storyboard - Manga Character Collaboration Scene — output

Comic / Storyboard - Manga Character Collaboration Scene

Strix8GPT Image 2
Image 1: {argument name="character reference" default="References for Azaty (blonde) & Mikado (black hair, one eye hidden)"}
Image 2: {argument name="guest character reference" default="Reference for the guest character entering the school"}

Maintain face, hair, eyes, skin, clothing, body type, and unique features faithfully to generate a vertical color 4-panel manga. No turning them into different people or adding/losing features. Use uniform, clean anime-style coloring, natural 5-finger hands, and accurate Japanese text.

[Personalities]
Azaty: A bright, decoration-loving gal. Pink round POP-style text.
Mikado: Quiet and skillful. Thin Mincho-style text.
Guest: Infer personality from Image 2. If the setting is non-verbal, use only inner thoughts and gestures.

[Sword Criteria]
Refer to Image 2 for sword shape, sheath, handle, guard, coloring, materials, and decorations.

- If Image 2's sword is already decorated:
In panels 1 and 2, draw it in a 'plain state' with decorations removed. Do not change the shape, color, handle, or guard. In panel 3, the two characters apply decorations to match the completed design in Image 2.

- If Image 2's sword is simple and undecorated:
Draw it as in Image 2 for panels 1 and 2, and add new decorations suited to the character in panel 3.

- If there is no sword in Image 2:
Automatically generate a simple sheathed sword that suits the guest and decorate it in panel 3.

Panels 2 and 3 must feature the exact same sword. Unify length, sheath, handle, and guard; do not show the blade.

[Panel 1]
A classroom on the first day. The guest sits small, holding a sword case. Other students are chatting around them.
Inner thought: "I wonder if I can make friends...?"

[Panel 2]
Azaty on the left, guest in the center, Mikado on the right. The guest holds the plain sword taken from the case.
Azaty: "That sword looks like it could be totally cute!"
Mikado: "...Let me borrow it for a bit."

[Panel 3]
Decorating the sword at a desk. Azaty handles ribbons, hearts, and rhinestones; Mikado handles black and red braided cords. If a decorated sword is in the reference, accurately reproduce that finished design. The guest watches from the center with a sweat drop emoji.
Azaty: "Since it's plain, it's really worth decorating!"
Mikado: "If we tighten it with black and red, it won't be too sweet."

Attach a small mascot keychain themed after the guest to the sheath with a short gold chain. Ensure it doesn't look like a toy sitting on the desk.

[Panel 4]
A classroom at sunset. A smiling Azaty on the left, a joyful guest in the center, and a smiling Mikado on the right. The three are huddled together happily. For expressionless characters, show joy through blushing or gestures. Do not draw the physical sword.

At the bottom, place a POP-style friendship text logo combining the colors and traits of the three, with hearts, beads, and black ribbons.

"Gal Sword Motto!
Decorate your sword and you're already besties ♡"

Sticker-style with thick black and white outlines and gold edges. Do not cover faces; draw text accurately.

Point speech bubble tails toward the correct speaker. Unify characters, costumes, swords, and items across all panels. Prohibit extra dialogue, garbled text, giant swords, watermarks, or uneven coloring.
使用此提示词 ↗
Social Media Post - Pink Rose Garden Portrait — output

Social Media Post - Pink Rose Garden Portrait

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject:
{argument name="subject" default="Pink Rose Garden with Petals"}

Main Figure:
{argument name="main figure" default="A woman in her 20s smiling while holding a large bouquet in a pink rose garden with petals dancing in the air"}. Large thigh-up placement in the center of the frame, highlighting the pale pink sheer frilled dress and the rose bouquet.

Person and Expression:
Rounded oval face with a small chin, large brown eyes, thin arched eyebrows, small nose, and glossy coral pink lips. Head tilted slightly to the right, looking at the camera with a bright smile showing teeth. Brown hair in a high loose bun, thin bangs, fine wisps of hair, and pink floral ornaments on both sides.

Clothing and Pose:
{argument name="dress" default="Pale pink off-the-shoulder mini dress"}. Detailed frills at the neckline and shoulders, a lace-up corset-style bodice, transparent long puff sleeves, a large ribbon at the waist, and multi-layered lace hem. Facing the front, holding a bouquet of pink roses at the hip with the left hand and lightly pinching the skirt hem with the right hand.

Background and Light:
Pink roses in the foreground and on the sides, a fountain in the back left, and a pavilion with white pillars and a glass dome in the back right. Clear blue sky above with large and small pink petals dancing in the air. Soft, bright direct sunlight from the upper right, translucently illuminating the hair and sheer sleeves, enveloping everything in a soft warm tone.

Composition and Camera:
Vertical 3:4. Thigh-up portrait taken with a front-facing camera at eye level. Person positioned slightly left of center, bouquet at bottom right, fountain on the left, and pavilion at top right. Skirt and thighs cut off at the bottom edge. Focus on the face and bouquet with a shallow depth of field that softly blurs the background.

Texture and Style:
Bright, ethereal, photorealistic live-action photograph. Precise rendering of skin transparency, organza sheerness, lace and lacing, rose petals, and fluttering petals in the air, with soft color development of pink and sky blue.

Negative:
Major changes to facial features; omission of the sheer long sleeves and corset structure.
使用此提示词 ↗
Profile / Avatar - Woman Stretching in Sports Wear Under Blue Sky — output

Profile / Avatar - Woman Stretching in Sports Wear Under Blue Sky

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject:
White sleeves extending toward the blue sky

Main Subject:
A {argument name="age" default="20-something"} woman standing on a low concrete wall, stretching her arms overhead with interlaced hands. Place the full body prominently in the center of the frame, showing {argument name="sportswear" default="gray-blue sportswear and a white patterned windbreaker"} against a vast blue sky.

Character and Expression:
A small oval face, thin brown eyes, natural eyebrows, and pale pink lips. Facing the upper right of the frame, looking up at the sky with a gentle smile. Long dark brown hair worn down, flowing to the left of the frame in the wind.

Clothing and Pose:
A gray-blue thin-strap sports top and high-waisted short sports pants. Layered over it is a thin long-sleeved zip jacket with a black abstract pattern on a white background, worn open. White short socks and white platform sneakers. Standing with legs slightly crossed. Arms stretched straight up, hands interlaced overhead.

Background and Light:
Most of the background is a vivid blue sky, with white cumulus clouds at the bottom left and lower edge, and a gray concrete wall at the very bottom. Strong direct sunlight from above brightly illuminates the white sleeves and legs, creating short shadows in the clothing folds.

Composition and Camera:
Vertical 3:4 aspect ratio. A strong low-angle wide-angle camera shot looking up from near the ground, capturing the entire body from the soles of the shoes to the interlaced hands. Position the person tall and large in the center, leaving wide blue sky margins above and to the sides. Place the shoes and wall near the bottom edge, focusing on the person to keep them and the sky clear.

Texture and Style:
A photorealistic live-action photograph taken outdoors at midday. Precisely depict the stretch and luster of the sports fabric, the wrinkles of the thin jacket, the movement of the hair in the wind, and the texture of the white shoes.

Negative:
Changes to the interlaced arms overhead; omission of the white patterned jacket
使用此提示词 ↗
Social Media Post - Dark Chic Metallic Fashion Portrait — output

Social Media Post - Dark Chic Metallic Fashion Portrait

CHAseGPT Image 2
[Subject] Image 1 [Sign] The {argument name="logo" default="Meta Threads logo"} is located small in the bottom right, with "{argument name="signature" default="CHAse"}" written in white script above it like a signature.

[Subject Pose and Expression]

Pose and Composition: A medium full-length shot from a front-side angle capturing the full body and styling in a provocative yet controlled squatting pose, focusing on the silhouette.

Posture: Knees up, upper body resting on them, wrists overlapped and hanging down while leaning toward the camera. A dynamic angle emphasizing curves blends with the flow of chain decorations to create an overwhelming charisma.

Expression: A head slightly bowed with an intense, deep gaze staring directly at the camera over the shoulder. Clear eyes and a haughty mouth create the chic and modern aura characteristic of official concept pictorials.

[Clothing and Props]

Clothing:

Dress: A deep black mini sleeveless dress creating a glamorous and three-dimensional silhouette from the upper body to the hip line. Elaborate silver beads and multiple layers of metallic chain fringe are embroidered like a chandelier along the neckline and body lines.

Shoes: Deep black leather ankle steeple boots with a matte-glossy texture, featuring a high, clunky platform mold and thick heels.

Props & Nails:

Minimalist silver chain bracelets layered on the wrists, with deep black nail art adding weight and a chic finish to the fingertips.

[Hair and Makeup Details]

Hair: Sleek jet black full bangs covering the forehead with long, voluminous straight hair flowing over the shoulders and back. A light braid detail on the right crown adds a trendy element.

Makeup: A smooth, fair porcelain skin base with sharp deep black cat-eye eyeliner and deep smoky shaded eye makeup. Lips are finished with a calm nude rose pink over-lip for a sophisticated look.

[Lighting and Direction]

Light Source: Precise studio spot pin lighting (Studio Key Light) coming from the front upper side of the subject.

Effect: Creates intense and brilliant metallic specular highlights on the skin surface, collarbones, and silver chain particles of the dress against the dark emerald background.

[Texture and Color Mood]

Colors: A 'Dark Chic Metallic' color mood with clear contrast between the {argument name="background color" default="deep emerald/teal green"} background, the absolute jet black of the clothing and hair, and the cold silver of the chains.
使用此提示词 ↗
Social Media Post - Summer Stream Snorkeling Mask Portrait — output

Social Media Post - Summer Stream Snorkeling Mask Portrait

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: {argument name="theme" default="Forest waterside and floral swimsuit"} Main Focus: {argument name="person" default="woman in her 20s"} standing knee-deep in a clear river surrounded by greenery, holding a black snorkel mask in one hand. Large focus on the character placed to the right, featuring the white floral bikini and water ripples. Person/Expression: Slender oval face with small jaw, closed wide eyes, thin eyebrows, small nose, pale coral lips. Head tilted slightly down with a quiet smile and closed eyes. Long dark brown hair in a high ponytail, flowing over her right shoulder, with thin bangs. Clothing/Pose: {argument name="outfit" default="triangle bikini with orange floral patterns on white background"}. Clear view of strings at neck, back, and hips. Standing on both legs in water. Right arm bent with hand in hair, left arm extended diagonally down holding a black mask and orange-tipped snorkel. Background/Light: Foreground to center is transparent shallow water reflecting greenery, grey stones and ferns on the opposite bank, dense conifers in the back. Strong direct sunlight from the upper left illuminates skin and water surface, creating fine golden ripples and reflections of legs underwater. Composition/Camera: 3:4 vertical. Near full-body portrait taken from a front camera slightly above water level, from top of head to below knees in water. Character placed large on the right half with wide water space on the left, mask near the right edge. Legs shown through water surface and cut at the bottom. Focus on the character with deep depth of field showing the forest. Texture/Style: Photorealistic photo taken in midsummer natural light. High-definition depiction of water droplets on skin, floral pattern on cloth, water transparency, reflecting ripples, and details of leaves and rocks. Negative: Omission of snorkel gear in left hand; do not open closed eyes.
使用此提示词 ↗
Profile / Avatar - Morning Bedroom Sunlight and Lace — output

Profile / Avatar - Morning Bedroom Sunlight and Lace

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: {argument name="theme" default="Morning bedroom unwinding in light"} Main Focus: {argument name="person" default="one young adult female"} sitting on a white bed in a wood-toned bedroom with morning sun entering through shoji and a large window. Character placed large in the center, focusing from face to bent legs, ivory camisole with sheer wrap, and pale greige shorts. Person/Expression: Small rounded oval face, bright smooth skin, dark brown almond eyes, thin natural eyebrows, small nose, glossy pale rose slightly parted lips. Tilted slightly left looking sleepily at the camera with shoulders shrugged. Long lashes, thin eyeliner, faint redness on cheeks and around eyes. Loose straight dark brown hair reaching below chest, long thin bangs and strands over face, hair falling over left shoulder. Clothing/Pose: {argument name="outfit" default="ivory thin-strap camisole top, white sheer lace long-sleeve robe, pale greige pink soft shorts"}. Sitting on bed with left leg extended forward and right knee bent inward. Supporting body with hands on bedding, leaning slightly forward. Background/Light: Large black-framed glass window and white shoji on the left, soft blurred greenery outside. Low wooden storage, white round table lamp, small bottles and frames in back right. Foreground and right side show wrinkled white duvet and pillows. Warm direct morning light from the back-left window illuminates hair edges, cheeks, chest, left sleeve, and thighs diagonally. Soft diffused light inside leaves shadows. Composition/Camera: 2:3 vertical. Near sitting portrait from a camera at bed height. Head in top-center, torso in center, thighs and bent legs occupying the bottom half. Hands near bottom edges. Precise focus on face and chest with shallow depth of field blurring the room background. Texture/Style: High-definition photorealistic photo in natural morning light. Depicts natural skin texture and fine glow, back-lit hair strands, rib knit, lace transparency, soft shorts, wrinkled white bedding, wood, and glass. Quiet tones of white, beige, and pale pink, avoiding excessive smoothing or illustration style. Negative: Alteration of facial features, hairstyle, off-shoulder robe, or sitting posture on bed.
使用此提示词 ↗
Profile / Avatar - Elegant Windowside Portrait with White Rose — output

Profile / Avatar - Elegant Windowside Portrait with White Rose

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: {argument name="theme" default="Windowside gazing at a white rose"} Main Focus: {argument name="person" default="woman in her 20s"} kneeling with one knee on a beige sofa by a bright window, gazing at a single white rose. Large focus on character in center-right, featuring the white thin-strap mini dress and the hand holding the flower. Person/Expression: Slender oval profile, downcast dark brown eyes, thin eyebrows, small nose tip, pale coral lips. Head tilted down-left with a calm expression looking at the rose. Hair nearly black tied low, leaving thin bangs and long strands along cheeks. Clothing/Pose: {argument name="outfit" default="white thin-strap bodycon mini dress"}. Clear view of shallow V-neck, chest stitching, and fine horizontal wrinkles from waist to hem. Layered wide gold choker and thin necklace. Left leg bent on the sofa arm, standing on right leg. Right hand holding the rose stem, left arm lowered behind back. Background/Light: Large window with white lace curtains on the right, grey wall on the left with black geometric pendant and wood wall lamp, beige sofa below. Soft daylight from the right illuminates skin and white cloth, casting faint warm shadows on the left. Composition/Camera: 3:4 vertical. Thigh-up portrait from a front camera at eye level. Character large on the right half, rose in top-left, bent left leg extending bottom-left. Right leg and left knee cut at the bottom. Focus on face and rose with light background blur. Texture/Style: Photorealistic photo in natural window light. Detailed depiction of natural skin texture, thin wrinkles in white cloth, rose petals, metal accessories, and light through curtains. Negative: Alteration of posture with knee on sofa; omission of white rose.
使用此提示词 ↗
Profile / Avatar - Beach Resort Crochet Bikini — output

Profile / Avatar - Beach Resort Crochet Bikini

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: {argument name="subject" default="Shade with a cherry blossom pink bikini"}

Main: A woman in her 20s kneeling sideways on a white deck chair by the sea, touching her tied-up hair with one hand. Place her prominently in the center of the screen, making the {argument name="swimsuit" default="cherry blossom pink crochet bikini"} and the {argument name="parasol" default="navy and white striped parasol"} overhead the protagonists.

Character/Expression: Slender oval face with a small rounded jaw, horizontally elongated brown eyes, thin gentle eyebrows, a clean nose bridge, and pale coral lips. A calm expression looking into the distance with the face turned to the left, mouth slightly relaxed. Dark brown hair loosely tied at the back of the head, leaving thin bangs and wisps along the cheeks.

Clothing/Pose: Cherry blossom pink crochet triangle bikini. Tied with thin strings at the neck and back, with a small gold ring in the center of the chest, ties on the left and right of the waist, and a thin chain and star-shaped ornament on the hip bone. Sitting sideways on the chair with both knees bent, her right arm raised overhead touching her hair, and her left hand down by her thigh.

Background/Lighting: Wooden parasol pole and rattan bag in the left foreground, white deck chair in the center. Background features white sand, deep green palms and plants, and a white resort building. Warm direct sunlight enters from the back right, creating bright edging on the shoulders and waist, while providing soft shade under the parasol.

Composition/Camera: 3:4 aspect ratio. Knee-up portrait capturing from the parasol overhead to below the knees with the camera slightly lower than eye level and centered. The person is placed prominently slightly to the right, the raised right arm forms an arc to the top left, and both knees are cut off at the bottom. Focus on the person, with a medium depth of field lightly blurring the building and plants.

Texture/Style: Photorealistic live-action photo in warm natural light. Finely depicts natural skin texture, the gaps and threads of the crochet knit, metal ornaments, and the texture of sand and wood, softly harmonizing the cherry blossom pink with the tropical green.

Negative: Significant changes to facial features; omission of the crochet bikini structure.
使用此提示词 ↗
E-commerce Main Image - Blue Pinstripe Shirt Fashion Outfit — output

E-commerce Main Image - Blue Pinstripe Shirt Fashion Outfit

MoodLock|AI画像の惜しいを直すGPT Image 2
Wardrobe:

Wardrobe assembly rule:
Each FINAL WORN STATE overrides conflicting wearing-state wording only for the item descriptions it explicitly scopes or names, and only for wearing role or use, count, position, side, orientation, routing, fastening, layering, folds, tucking, and draping. Preserve those descriptions for identity, design, material, color, pattern, construction, and proportions.

Top:

One solid white cropped bustier top constructed from fine translucent floral lace and narrow opaque woven binding. The silhouette is close and compact, with a shallow rectangular front envelope, two broad straight straps, and a continuous narrow lower band. A low softly squared neckline spans between the straps. The front is divided into two shallow curved lace panels by a narrow center bridge, with a horizontal opaque binding defining the lower edge of the panels and a second binding completing the base. Dense floral-and-leaf lace fills every framed panel, retaining small irregular transparent cells between the embroidered threads. The broad straps continue into the rear band, where two narrow white tie extensions form one compact bow with two balanced loops and two short finished tails. Keep the lace panels flat and distinct, the binding widths consistent, the strap paths independent, the rear bow compact, and the open lacework crisp rather than blurred into solid white fabric.

One pale sky-blue and white pinstripe longline button-front shirt in fine, lightly crinkled cotton shirting, with dense narrow white stripes running longitudinally across the main panels and sleeves and following the cut direction of the collar, placket, and cuffs. The cut is relaxed and spacious, with dropped shoulders, a straight upper section, and softly expanding volume through the lower section. A broad pointed collar frames a short exposed upper placket that continues into a full center-front opening. Long sleeves follow a generous straight line from low armscyes and narrow into deep rectangular cuffs. Clean side seams, a narrow stitched front placket, and an extended shirttail hem with curved front and rear edges and short side separations complete the panels. Small matte cream four-hole buttons are evenly spaced along the placket. Preserve the fine stripe spacing, crisp pattern alignment, airy shirt volume, clean collar points, controlled cuff structure, and dry woven surface without swollen seams, melted stripes, or fuzzy edges.

Final worn state:
For the second item description above only:
- Front closure: Engage every existing front closure from the top edge to the bottom edge and align the two front edges continuously on the anatomical center front. Leave no open interval, offset panel, or invented fastening. FINAL FASTENING STATE: every selected closed position is physically connected to its matching receiver; every selected open position is visibly disconnected, and panel overlap does not imitate a fastening.
- Sleeves: Release both cuff and sleeve-placket closures, then turn each sleeve outward in broad folds. Position both rolled bands immediately below the elbow points. Preserve both cuffs and keep every fold below the elbows. Do not push either roll above an elbow, hide a cuff, or leave one sleeve at the wrist.
- Layering: Layer this garment underneath the first item description under Top across the upper body regardless of whether either garment has a front opening. Preserve each garment's selected or source-default closure state and expose only the inner garment's neckline, collar, cuffs, sleeves, or hem where the outer garment does not physically cover them. Change only the inside-outside order; do not invent an opening or fastening, merge the garments, or route either garment through the other's neckline, armhole, sleeve, seam, or panel.
- Finish — Natural: Use a few broad soft folds with modest overlap and small large-scale asymmetry while preserving every cuff, hem, seam, and hardware component.

For the second item description under Top and the first item description under Top together only:
- Finish — Relaxed: At the real upper-body overlap at the neckline, armholes, sides, and hems, compress mainly this covered inner garment and let the outer target yield only locally; let only a few broad folds leave those boundaries and descend within each garment's own panels. Preserve both garments' necklines, armholes, straps, sleeves, seams, closures, and hems. Use light stable contact and broader material-correct ease without changing the selected layer order.

Bottom: One charcoal-grey two-tier pleated mini skirt made from lightweight opaque woven fabric with a smooth matte-to-soft-luster face, crisp fold memory, and fluid movement. A clean fitted waistband supports a full upper tier formed from broad regular knife pleats that open into a short circular flare. A second pleated tier is joined beneath the first tier and expands to a slightly deeper lower flare, producing two clearly separated horizontal levels without merging their pleat systems. Every tier edge is finished with one narrow white gathered ruffle tape that remains continuously visible as a bright border against the grey ground. One small muted lavender-grey rectangular woven label is centered on a lower-front outer section, framed by a narrow white scalloped border and carrying the exact two-line inscription “Moonlit” in large white cursive above “Daydream” in smaller white cursive. Keep both tiers evenly distributed, the pleat ridges sharply legible, the white edge trims narrow and continuous, and the rear construction consistent with the same two-tier pleated assembly.

Legwear: One matching pair of warm-ivory knee-high socks, displayed side by side as complete footed products. Opaque fine-rib knit with a soft matte finish, straight tubular shafts, gently shaped heels and toes, and wide plain elasticized cuffs with a subtle folded appearance. Uniform warm-ivory color throughout, tonal linking at the toes, no logos, no stripes, no lace, no graphics.

Footwear: Create one anatomically mirrored matching pair of white low-top lace-up sneakers with rounded shell toes and flat white rubber cupsoles. Build each upper from smooth white leather panels with a low padded collar, seven paired lace positions, flat white laces, a straight central tongue, and a ribbed molded rubber shell covering the complete toe cap. Apply three parallel black stepped bands to the outer quarter, each band formed from two solid diagonal rectangular segments separated by one narrow vertical gap; keep the three bands equal in width and evenly spaced. Add one compact white rectangular tongue label printed with the clearly legible uppercase word “SOLSTICE” in black condensed lettering. Finish the sole with a fine textured sidewall and shallow dark outsole edge, keeping the complete pair identical except for anatomical mirroring, with precise panel stitching, balanced lacing, clean black-white separation, and no graphics outside the confirmed quarter and tongue zones.

Bag:

One compact metallic-silver mini tote bag with a softly structured rectangular body, narrow side gussets, rounded lower corners, and a flat reinforced base. Smooth reflective synthetic leather covers the body with crisp silver edge binding and tonal stitching. Two short rolled top handles attach through rectangular stitched tabs, while one long narrow adjustable strap connects to small silver swivel clasps at the upper side rings and includes a matching slider buckle. A single silver zipper closes the straight top opening and terminates at compact end tabs. Keep the broad faces plain and uninterrupted, with pale-gray lining and no text, patches, appliqués, or decorative ornaments.

Final worn state:
- Bag carry: VIEW: FRONT/NEAR-FRONT SUPPO
使用此提示词 ↗
Comic / Storyboard - Cartoon Ants Carrying a Golden Seed — output

Comic / Storyboard - Cartoon Ants Carrying a Golden Seed

CencenGPT Image 2
Create a charming cinematic 3D animated macro scene in a Pixar-like style: {argument name="number of ants" default="5"} glossy red cartoon ants with oversized expressive eyes march in a line down a narrow stone channel like a miniature canyon path. The front ant is closest to the camera, smiling and pushing a tiny wooden sled made from two parallel twig runners and cross sticks, carrying one large shiny {argument name="carried object" default="golden yellow berry or seed"}. Behind it are exactly 4 more red ants following along the same path: one directly behind the leader, one behind that, one farther up the slope, and one tiny ant barely visible in the upper-right background. The ants have rounded segmented bodies, thin legs, curved antennae, friendly faces, and toy-like orange-red shells. The environment is a sunlit garden floor with weathered gray stone walls on both sides of the channel, small patches of moss, tall green blades of grass, and soft natural shadows. Use a vertical {argument name="aspect ratio" default="9:16"} composition, low macro camera angle looking slightly downhill toward the sled, shallow depth of field with the foreground ant and yellow object sharp while background ants and plants blur gently. Lighting should be warm midday sunlight with cinematic highlights, vibrant colors, whimsical adventure mood, high-quality 3D render, no text, no watermark, no humans.
使用此提示词 ↗
Infographic / Edu Visual - 2D Floor Plan to 3D Home Render — output
Social Media Post - Editorial Outfit Breakdown — output

Social Media Post - Editorial Outfit Breakdown

MagnificGPT Image 2
Using REFERENCE_0 as the source outfit photo, turn the mirror-style fit check into an editorial magazine outfit breakdown while keeping the original person, pose, room photo, and worn outfit visible as the central base image. Add a bold fashion-layout overlay with the headline {argument name="headline text" default="TODAY’S FIT"} at the top and a small handwritten subtitle {argument name="subtitle text" default="what I’m wearing"} beneath it.

Extract and isolate exactly 6 clothing/accessory items from what the subject is wearing, presenting them as cutout product callouts with off-white sticker outlines and thin white leader lines pointing back to the matching item on the subject: 1) jacket, 2) pants, 3) sneakers, 4) beanie, 5) sunglasses, 6) necklace. Arrange the cutouts around the subject: jacket upper left, pants lower left, sneakers bottom left, beanie upper right, sunglasses mid-right, necklace lower right.

Add compact uppercase labels next to each item with brand and price in an editorial catalog style: JACKET — KOVI — $128; PANTS — SENKO — $94; SNEAKERS — VYRO — $145; BEANIE — NOMA — $38; SUNGLASSES — AERI — $72; NECKLACE — MOTIF — $46. Use {argument name="label style" default="condensed black uppercase fashion typography"} for the labels and preserve the gritty high-contrast cutout collage look.

Keep the background photo recognizable, slightly darkened where needed for readability, and do not invent extra garments beyond the 6 visible outfit pieces. Maintain a vertical social-media editorial poster composition with clean spacing and no additional logos or watermarks beyond what is already present in the reference.
使用此提示词 ↗
Social Media Post - Cinematic Holographic Music Player Edit — output

Social Media Post - Cinematic Holographic Music Player Edit

MagnificGPT Image 2
Using the provided reference image as the base street-busker photo, keep the person, guitar, outfit, sidewalk location, guitar case, and overall camera angle intact, but transform it into a viral cinematic music-edit scene. Add floating translucent glass music-player UI panels around the performer, with glowing rounded edges, soft reflections, and slight perspective depth as if they are holographic cards suspended in the street. Apply warm golden-hour color grading, stronger contrast, subtle motion/radial blur streaks around the edges, and a polished editorial look while keeping the subject relatively sharp.

Add exactly 6 visible music-player cards:
1. Large foreground card at lower left: album art with a red/orange blurred silhouette, title “SOFT NOISE,” artist “LUMA VENN,” timestamp 1:52, duration 3:48, and playback controls.
2. Large foreground card at lower right: misty blue-pink album art, title “SLOW SIGNAL,” artist “ELIORA,” timestamp 1:30, duration 3:35, and playback controls.
3. Small upper-left card: title “SIDEWAYS,” artist “NAMI REY,” with album art and controls.
4. Small top-center card: title “NOA VEL,” artist “AFTER MIDNIGHT,” with album art and controls.
5. Small upper-middle-right card: title “HALF AWAKE,” artist “MONO JUNE,” with album art and controls.
6. Small upper-right card: title “STATIC HEART,” artist “KIRO MAVE,” with album art and controls.

Style constraints: make the cards semi-transparent so the street remains visible through them; use white typography and minimalist player icons; add luminous edge highlights and realistic glass distortion; preserve the small black “AI” badge in the bottom-left corner; avoid changing the performer’s identity or pose. Optional customization: use {argument name="main left song title" default="SOFT NOISE"}, {argument name="main right song title" default="SLOW SIGNAL"}, {argument name="color grade" default="warm golden-hour cinematic"}, {argument name="UI material" default="translucent glowing glass"}, and {argument name="overall effect" default="viral music edit with edge blur"}.
使用此提示词 ↗
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 ↗