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

Social Media Post - Long Exposure Motion Blur Photography — output

Social Media Post - Long Exposure Motion Blur Photography

Alexandra AislingGPT Image 2
A {argument name="mood" default="cinematic"} photograph of a {argument name="subject" default="fast moving subject"} tearing through a {argument name="environment" default="landscape"}, the environment captured in long exposure with {argument name="blur effect" default="light trails / motion blur / color streaks"}, while the subject remains tack-sharp and frozen in motion, with a color palette of [COLOR PALETTE].
Product Marketing - Futuristic Techwear Studio Portrait — output

Product Marketing - Futuristic Techwear Studio Portrait

AliGPT Image 2
Full-body studio portrait of a young man in futuristic high-fashion techwear. he is wearing a layered ensemble featuring an oversized {argument name="jacket color" default="olive-drab"} hooded bomber jacket with multiple tactical pockets and {argument name="accent color" default="mustard-yellow"} accents. Underneath, a cream-colored hoodie and a high-neck scarf. Hhe wears {argument name="pants style" default="charcoal-grey heavy-duty cargo joggers"} adorned with straps, buckles, and large modular utility pouches. Accessories include fingerless tacticalgloves, a utility belt with a quick-release buckle, and rugged tan combat boots with thick soles. The lighting is clean, high-key studio lighting with soft shadows against a neutral light grey background. 8k resolution, cinematic fashion photography, sharp focus, hyper-realistic textures of nylon, canvas, and leather. Aspect ratio is 9:16.
Profile / Avatar - Summer Balcony Film Portrait — output

Profile / Avatar - Summer Balcony Film Portrait

ShinningGPT Image 2
Create a hyper-realistic outdoor lifestyle portrait in a dreamy summer film photography style, 3:4 vertical composition. Use the uploaded portrait only as the identity and hairstyle reference. Preserve the person’s facial identity, face shape, natural facial features, skin tone, hairstyle, hair length, hair volume, and general body proportions accurately. Do not copy the original portrait’s clothes, background, lighting, pose, or image quality. The subject is standing on an upper-floor balcony, leaning slightly forward over a dark blue metal railing and looking directly into the camera with a calm, soft, gentle expression. The camera is positioned from a low angle below the balcony, creating a natural upward-looking perspective, as if someone is looking up at the subject from the courtyard. The scene takes place on a bright sunny day outside a quiet residential building. The background includes a pale white exterior wall, a slanted roof edge, clean blue sky, and soft green tree leaves in the foreground. Some leaves should be close to the lens and slightly blurred, creating natural depth and a candid summer atmosphere. The subject wears a {argument name="clothing" default="simple white summer dress or soft white blouse"} with a clean youthful look. The styling should feel fresh, natural, and realistic, not overly fashionable or staged. Hair should move slightly in the breeze, with fine hair strands catching the sunlight. Keep the pose relaxed and natural, with one or both hands lightly touching the balcony railing. Lighting should be bright natural daylight with strong backlight from the upper sky, soft lens flare, glowing rim light around the hair, slight overexposure in the highlights, and a warm nostalgic summer feeling. The image should look like a real photograph taken with a vintage digital camera or soft film camera: gentle grain, soft focus, natural skin texture, subtle haze, slightly washed colors, realistic shadows, and no heavy beauty retouching. Make the final image feel like a real-life candid summer portrait, romantic, clean, airy, youthful, and cinematic. No text, no logo, no watermark. Negative Prompt: anime, illustration, cartoon, 3D render, CGI, doll-like face, plastic skin, overly smooth skin, heavy retouching, fake face, distorted face, wrong facial identity, changed hairstyle, changed hair length, extra limbs, deformed hands, broken fingers, unnatural wrist, bad anatomy, strange body proportions, stiff pose,
Social Media Post - Creative Doodled Image Overlay — output

Social Media Post - Creative Doodled Image Overlay

OogieGPT Image 2
Edit this image while STRICTLY preserving the original subject, composition, background, lighting, and perspective. Do not replace or alter the environment — the real-world setting must remain fully intact and realistic. Slightly enhance the overall mood to feel more cinematic and calm — subtly deepen shadows, soften highlights, and introduce a gentle moody or rainy atmosphere while keeping it natural. Now transform the image by adding a highly expressive, hand-drawn overlay that creates a sense of a “living micro world” around the product. Use a {argument name="art style" default="crayon + marker + pencil sketch style"} that feels playful, slightly messy, but intentional and artistic — not overly chaotic, but rich and layered. Add controlled chaos elements: Hand-drawn arrows pointing to key product details (texture, shine, layers), Scribble circles and highlight marks to guide attention, Energetic lines, swirls, and sketch marks that enhance focus (not random placement). Introduce tiny doodle characters interacting with the product: Small illustrated figures climbing, sitting, sliding, or reacting to the product. Keep them minimal but expressive. Add small speech bubbles with playful reactions like “BRO???”, “THIS IS CRAZY”, or “WE STAYING HERE”. Ensure all doodles feel integrated into the scene and follow the product’s shape and perspective. Create a strong emotional contrast: Background remains calm, soft, and slightly moody; Overlays are lively, bright, and energetic. Add layered handwritten notes in a natural human style (not clean fonts), such as: “this part tho…”, “wait—look at this”, “why is this so good”. Optionally include a bold handwritten phrase like: “{argument name="slogan" default="LAST DANCE"}” as a statement element. The final result should feel like a real photograph infused with imagination — a quiet scene brought to life by chaotic, playful thoughts.
Profile / Avatar - Sunlit Butterfly Mirror Portrait — output

Profile / Avatar - Sunlit Butterfly Mirror Portrait

ShinningGPT Image 2
Use the uploaded portrait only as the identity and hairstyle reference. Preserve the subject’s facial identity, face shape, hairstyle, skin tone, natural body proportions, and realistic personal features. Do not copy the uploaded portrait’s background, outfit, lighting, or pose. Create a vertical hyper-realistic lifestyle portrait in a cozy indoor bedroom or dressing room, shot like a real casual smartphone photo with a dreamy warm sunlight atmosphere. The subject is sitting slightly forward, facing the camera with a soft natural expression, relaxed shoulders, and gentle eye contact. The composition should feel intimate, youthful, bright, and slightly nostalgic. The scene is lit by strong golden sunlight coming from the side and behind the subject, creating glowing rim light around the hair, bright sun flares, soft overexposed highlights, and natural shadow contrast. The hair should catch the sunlight with visible fine hair strands and realistic texture. The background includes a casual mirror, bookshelves, posters, soft room clutter, and a warm lived-in interior. In the mirror, show a partial rear reflection of the subject, naturally matching the main subject’s hairstyle and outfit from behind. Keep the reflection physically realistic and not distorted. The subject wears a casual light cream or pastel graphic T-shirt with tasteful English text on it, such as “SUNLIT BUTTERFLY DIARY” or “SOFT SUMMER MEMORY”, paired with a relaxed plaid shirt or light jacket slipping off the shoulders. The outfit should look natural, casual, cute, and realistic, not cosplay and not overly styled. Add many colorful realistic butterflies flying around the subject and filling the frame, with different sizes, colors, and depths. Some butterflies are close to the camera, some are mid-air, and some are farther in the background. The butterflies should look like realistic photo elements, softly integrated with the lighting, perspective, and depth of field. Use orange, blue, yellow, white, purple, and green butterflies. The lower foreground can have denser butterflies, while the upper area has scattered flying butterflies. Style: real-life photography, photorealistic, natural skin texture, realistic facial details, soft warm sunlight, dreamy indoor portrait, casual youthful mood, smartphone snapshot realism, high detail, realistic mirror reflection, cinematic golden-hour backlight, no anime style, no cartoon style, no illustration. Aspect ratio: 9:16 vertical portrait. Negative Prompt: watermark, logo, AI-generated text watermark, Korean text, Chinese text, unreadable text, messy typography, duplicated face, different face in mirror, wrong reflection, distorted mirror, deformed body, broken neck, unnatural skin, plastic skin, over-beautified face, fake doll face, anime style, cartoon style, illustration, 3D render, CGI, low quality, blurry face, bad anatomy, extra fingers, missing fingers, twisted hands, distorted arms, extra limbs, unnatural pose, harsh face smoothing, oversaturated colors, ugly lighting, dirty background, random letters, incorrect shadows, floating objects without perspective, fake butterflies, sticker-like butterflies, flat butterfly overlay, poor blending, compression artifacts, noise, cropped head, cropped face, duplicated subject, cloned person, bad composition
Comic / Storyboard - Custom Manga Panel Page Illustration — output

Comic / Storyboard - Custom Manga Panel Page Illustration

NexoraGPT Image 2
Create a full-page colored manga illustration based on the person in the reference photo. Use authentic {argument name="layout" default="manga page composition with 5–7 dynamic connected panels on a single page"}. Include cinematic storytelling, dramatic perspective, expressive emotions, dynamic poses, speed lines, impact frames, close-up shots, and intense visual flow. Add English comic-style sound effects, handwritten dialogue bubbles, ink splatter strokes, halftone textures, rough comic sketch lines, and graffiti-inspired brushwork. Blend vibrant colors with manga-style shading and strong contrast. Emphasize the character’s personality and aura with symbolic visual elements and atmospheric energy effects. Highly detailed, emotional, stylish, energetic, professional manga aesthetic with clean white negative space mixed with chaotic ink textures. Hand-drawn look, free-flowing lines, immersive composition, high quality.
Profile / Avatar - Seaside Street Photography with Neon Doodles — output

Profile / Avatar - Seaside Street Photography with Neon Doodles

ShinningGPT Image 2
Use the uploaded image strictly as a facial identity reference, hairstyle reference, and body proportions reference only. Do not copy clothing, pose, accessories, background, lighting, expression, composition, or any other elements from the uploaded portrait. Recreate a completely new photorealistic scene while preserving only the person's facial characteristics, hairstyle, and natural body proportions.

Ultra-realistic lifestyle action photography, candid {argument name="scene" default="seaside boardwalk scene during a bright sunny afternoon"}, immersive wide-angle lens shot with natural perspective, a young woman captured in a {argument name="activity" default="spontaneous joyful movement near the ocean railing"}, wind blowing through her hair dramatically, dynamic body posture full of freedom and energy, holding an iced drink casually, ocean waves sparkling under sunlight, distant city skyline and flying seabirds in the background, colorful glowing {argument name="doodle style" default="light-paint doodles and neon sketch effects"} swirling around the body, floating butterflies, vibrant flowing streaks of cyan, pink, yellow, and purple surrounding the subject like artistic energy trails, realistic outdoor lighting, cinematic sunlight, subtle lens flare, authentic skin texture, natural shadows, highly detailed environment, real-world photography aesthetic, editorial street-style mood, energetic summer atmosphere, premium travel photography, ultra-detailed, realistic textures, high dynamic range, DSLR quality, sharp focus, depth of field, no illustration style, no anime style

Negative Prompt:

anime, manga, cartoon, illustration, cel shading, painting, 3D render, CGI, stylized character art, artificial face, fantasy character, doll-like skin, exaggerated body proportions, duplicate people, extra limbs, extra fingers, distorted anatomy, blurry image, low quality, overprocessed skin, oversaturated colors, warped face, deformed hands, bad posture, cropped body, unnatural expression, floating objects, messy composition, text, logo, signature, watermark, AI watermark, Chinese characters, artifacts, low resolution, unrealistic lighting, plastic skin, heavy filters, flat lighting, duplicated birds, duplicated butterflies
Profile / Avatar - Cozy Korean Living Room Portrait — output

Profile / Avatar - Cozy Korean Living Room Portrait

JohnnGPT Image 2
Ultra-realistic cozy living room portrait of a beautiful {argument name="subject" default="Korean woman"} sitting sideways on a {argument name="furniture" default="soft cream-colored couch"} at night, turning gently over her shoulder toward the camera with a soft shy smile. Legs mostly hidden from the frame by the {argument name="outfit" default="oversized fluffy pastel pink hoodie"} and couch angle, creating a cozy modest composition. Long silky black hair flowing naturally down her back, glossy eyes, soft blush makeup, natural skin texture, dreamy youthful Korean beauty aesthetic.
Warm ambient apartment lighting mixed with subtle realistic iPhone flash photography for soft highlights and cinematic shadows. Cozy Korean-style living room with cream furniture, fluffy blankets, soft pillows, candles, warm table lamps, pastel flowers, cute photo cards on the wall, aesthetic coffee table decor, fashion books, soft curtains, and nighttime city lights glowing outside large windows.
Romantic slice-of-life atmosphere, intimate late-night mood, candid relaxed pose, cozy feminine aesthetic, soft pink and cream tones, shallow depth of field, cinematic composition, highly detailed, photorealistic Korean apartment vibe, soft glow, lifestyle editorial photography, calm comforting atmosphere.
Comic / Storyboard - Illustrated Character Personality Board — output

Comic / Storyboard - Illustrated Character Personality Board

AijazGPT Image 2
A stylish illustrated character profile sheet of a handsome young man named “{argument name="character name" default="SAMRONE"}”, sketchbook aesthetic, mixed media ink and watercolor style, highly detailed fashion portrait, {argument name="hair style" default="messy dark curly hair"}, sharp blue eyes, short beard, confident expression, wearing a long black trench coat, dark casual clothes, brown leather backpack, rugged boots.
The page layout includes multiple poses and close-up portraits, surrounded by handwritten notes, doodles, symbols, travel sketches, cameras, mountains, coding references, and adventure elements.
Personality board design with handwritten text such as traveler, coder, photographer, teacher, sound engineer, Scorpio zodiac sign, river rafting, creative ideas, life goals, motivational notes.
Includes small chibi version of the character, laptop working scene, rafting action scene, map sketches, stars, arrows, and artistic scribbles.
Modern concept art meets fashion illustration, dynamic ink strokes, expressive line art, textured watercolor shading, warm paper background, cinematic composition, Pinterest aesthetic, highly detailed, visually rich, creative journal vibe, character design sheet, semi-realistic comic illustration.
Profile / Avatar - Handcrafted Mixed Media Portrait Painting — output

Profile / Avatar - Handcrafted Mixed Media Portrait Painting

Maverick | AIGPT Image 2
Create a highly detailed mixed media portrait painting of the subject’s face, combining layered {argument name="artistic elements" default="acrylic brush strokes, charcoal smudges, pencil sketches, rough paint splashes, collage textures"}, paper grain, expressive ink marks, and imperfect handcrafted details to create an authentic human-made artwork appearance. The overall composition should feel raw, artistic, and naturally imperfect, avoiding overly clean or overly polished digital results.

Use bright, eye-catching, and dynamic colors with rich tonal variation so the artwork feels vibrant and visually engaging without looking monotonous. Add visible skin texture details, layered paint textures, subtle rough strokes, and organic imperfections to enhance realism and artistic depth. The mixed media style should also influence the entire visual presentation, including the outfit and surrounding artistic elements, so the final artwork feels cohesive and fully immersed in the mixed media aesthetic.

The subject’s pose and facial expression must appear relaxed, expressive, lively, and natural, avoiding stiff body language, awkward posing, or gloomy emotional tones. The pose, angle, expression, and overall composition must be clearly different from the attached reference photo while still maintaining the subject’s identity.

The {argument name="outfit" default="modern, stylish, casual, and fashion-forward"} with layered clothing details, trendy contemporary fashion elements, unique textures, and visually interesting accessories that enhance the artistic mood. Avoid plain, generic, repetitive, or standard clothing designs, and ensure the outfit is completely different from the clothing shown in the reference image.

Focus on creating a visually balanced artwork with an expressive artistic atmosphere, realistic handcrafted imperfections, rich mixed media layering, and strong emotional energy, while still maintaining an elegant and aesthetically pleasing portrait composition.
Social Media Post - Anime Streetwear Poster Design — output

Social Media Post - Anime Streetwear Poster Design

HarborisGPT Image 2
Create a cinematic anime inspired streetwear poster featuring a confident {argument name="subject" default="young man"} standing in the center in a relaxed pose with one hand in pocket. He wears {argument name="outfit" default="black sunglasses, a light blue oversized denim shirt layered over a white tank top, faded charcoal washed jeans, white sneakers, silver chain necklace"}, and a modern urban hairstyle with trimmed beard.

Background features a massive {argument name="background icon" default="vintage textured Straw Hat Pirates skull logo from One Piece"}, blending into a gritty brown parchment wall with distressed textures, scratches, dust, and dark vignette edges. Add hand drawn anime doodles of Monkey D. Luffy around the composition with rough sketch aesthetics.

Poster design should feel like a premium Gen Z fashion editorial mixed with anime street culture. Include cinematic shadows, moody warm brown tones, grunge overlays, faded ink textures, and dramatic contrast.

Typography elements:

Large distressed bold text saying “OUTFIT CHECK”

Small handwritten phrases and pirate themed quotes

Japanese vertical typography on the side

Barcode element at bottom corner

“Gear Fifth” graffiti text

Minimal fashion magazine layout styling

Style: ultra detailed, anime streetwear editorial, cinematic poster art, gritty urban fashion campaign, textured vintage paper aesthetic, modern hypebeast design, high contrast lighting, premium composition, realistic human proportions, sharp focus.

Aspect ratio: 4:5 portrait.
Comic / Storyboard - Crimson Cloak in Snow Illustration — output

Comic / Storyboard - Crimson Cloak in Snow Illustration

Ayyoub AiGPT Image 2
Adult woman in a {argument name="cloak color" default="crimson"} hooded cloak kneeling in {argument name="environment" default="icy snow"}, hands clasped, staring upward, overhead centered composition, cloak spread in a circular splash around her, dark spears and rock shadows framing the scene, dramatic cold light, ink-and-gouache opaque flats on smooth matte illustration board, scratchy black pen line, dry-brush scumbling, scraped pale texture, aged art-book scan.
Social Media Post - Shopping Trolley POV Fashion Photography — output

Social Media Post - Shopping Trolley POV Fashion Photography

KGPT Image 2
Woman character inside {argument name="object" default="shopping trolley"}, POV angle from inside trolley looking upward toward her face, dramatic perspective distortion, one leg close to camera, relaxed Pose "Don't Care":  , {argument name="attire" default="oversized reddish-red Adidas cheongsam-inspired velvet tracksuit"} with beige stripes and knot closures, matching track pants, black Adidas Samba shoes, messy black bob hair moving in the wind, emotional calm expression, cloudy sky background, wet basketball court, cinematic youth photography, moody indie vibe, shallow depth of field, soft natural lighting, realistic skin texture, nostalgic film grain, premium fashion editorial, 4:5 portrait ratio.
Negative prompt legs position real normal no anomali legs.
Social Media Post - Street Side Bookstore Portrait — output

Social Media Post - Street Side Bookstore Portrait

Zar⭕onGPT Image 2
Generate a realistic portrait of a {argument name="subject" default="young man"} randomly stop at streetside front of the {argument name="location" default="book store which built by the red stone"}, wet weather, wet street, dim light, he stand slightly titled, a wide distance shot, slightly side angle shot, full body shot, hd image, he hold a old book in his one hand, look towards the camera, messy and trousled hairs, respective physics, he wear {argument name="outfit" default="dark maroon striped shirt with baggy white jeans and white shoes"}, strictly preserve respective face details, 4:5.
Social Media Post - Adorable Animals Luxury Fashion Group — output

Social Media Post - Adorable Animals Luxury Fashion Group

SairahGPT Image 2
A group of adorable anthropomorphic animals forming a circle and looking down into the camera, ultra wide-angle fisheye perspective, cute tiger, lion, bear, black cat, crocodile, panda, gorilla, bulldog, and red panda characters, luxury old-money fashion style, cream cable-knit sweaters, Ralph Lauren inspired outfits, sunglasses, watches, clean aesthetic, cinematic sunlight, blue sky with palm trees, highly detailed fur, glossy big eyes, realistic 3D Pixar-style render, vibrant colors, depth of field, symmetrical composition, playful yet premium vibe, ultra realistic textures, 8k. Cute gangster-style anthropomorphic animal squad standing in a circle and staring down at the camera, fisheye lens perspective, tiger, lion, bear, black cat, crocodile, panda, gorilla, bulldog, and red panda characters, varsity jackets, streetwear fashion, chains, luxury accessories, colorful outfits, confident expressions, sunny outdoor city street with palm trees, bright blue sky, cinematic lighting, ultra detailed fur and fabric textures, realistic 3D animated style, Pixar-inspired characters, vibrant urban aesthetic, dynamic composition, depth of field, 8k ultra detailed render.
Infographic / Edu Visual - Grok AI Editorial Infographic Poster — output

Infographic / Edu Visual - Grok AI Editorial Infographic Poster

Jahan ZaibGPT Image 2
Search the web for detailed information about {argument name="topic" default="Grok AI"} including its development history, underlying architecture, technical capabilities, training methodology, real-world applications, cultural impact, major breakthroughs, controversies, notable model versions, and surprising lesser-known facts. Then create a dramatic editorial infographic poster in the style of a {argument name="style" default="National Geographic feature"} — but for groundbreaking artificial intelligence. The most powerful and recognizable visual representation of Grok is the full-bleed atmospheric hero filling the entire poster top to bottom, dramatically lit and color-graded in deep black, electric blue, chrome silver, and cosmic neon tones matching Grok’s rebellious futuristic identity. Oversized bold typography with the word ‘GROK’ dominates the upper portion in a sleek sci-fi font inspired by space technology and cyberpunk interfaces. Text and graphic panels float naturally over the image as editorial overlays — not boxed magazine layouts. Include: release year and creator/company (xAI / Elon Musk), model architecture and parameter scale, training data scale and computing infrastructure, key capabilities and benchmark performance, integration with X (Twitter), adoption statistics and market influence, launch milestones, famous public demos, major achievements, cultural significance, and 3 surprising lesser-known facts. A technical specification panel on one side showing architecture, multimodal abilities, reasoning features, context window, API ecosystem, and compute power. A timeline of Grok’s major breakthroughs and releases along the bottom. Include references to real-time internet integration, humor-driven personality, open-source initiatives, and competition with other frontier AI models. Color palette and mood drawn entirely from Grok’s own visual identity — cinematic, edgy, intelligent, futuristic, slightly rebellious. No white background, no magazine grid, no boxed sections. Everything layered organically over the full-bleed hero image. Aspect ratio 4:5. Dense, tactile, cinematic, unmistakably Grok.
Infographic / Edu Visual - Luxury Patent Artifact Poster — output

Infographic / Edu Visual - Luxury Patent Artifact Poster

GadgetifyGPT Image 2
SYSTEM: Render the input as a {argument name="theme" default="luxury patent artifact poster"}. Treat the subject as both a historical intellectual object and a physical machine. Infer what the invention does, how it works, what problem it solves, what internal mechanisms matter, and what era it belongs to.  SEMANTIC SOLVE: PATENT_ARTIFACT =   (INFER(core_function FROM claims + mechanism + use_case) ::5) + (INFER(structure FROM components + assemblies + motion + logic) ::4) + (INFER(historical_aura FROM era + drafting conventions + industrial context) ::4) + (INFER(hidden_forces FROM failure_points + tolerances + constraints + manufacturing) ::3) - (generic blueprint cliché + sci-fi nonsense + clutter) ::-4  COMPOSITION: One central exploded patent object. Combine luxury rendering with precise technical callouts, sectional layers, hidden assemblies, schematic overlays, and elegant drafting fragments. Show how the invention thinks.  STYLE DNA: luxury technical poster ::0.30 museum invention plate ::0.25 engineering drawing elegance ::0.20 high-end object photography ::0.15 archival document texture ::0.10  OUTPUT: A premium white-background innovation poster with exacting detail, beautiful structure, and a sense of intellectual prestige.  NEGATIVE: no chaotic diagram overload, no cheap blueprint style, no random gears, no clutter, no low-end infographic look, no watermark.
Profile / Avatar - Low-Poly Geometric Celebrity Portraits — output

Profile / Avatar - Low-Poly Geometric Celebrity Portraits

Noor 🌸GPT Image 2
Ultra-detailed low-poly geometric portrait series featuring {argument name="celebrities" default="Leonardo DiCaprio, Michael Jordan, Conor McGregor, and Novak Djokovic"} in a modern polygonal digital illustration style, sharp angular facial planes, cinematic {argument name="color" default="teal"} gradient background, dramatic studio lighting, ultra-clean vector aesthetic, semi-realistic details, bold shadows, textured geometric shading, intense confident expressions, minimal composition, high contrast, premium poster design, ultra crisp edges, detailed eyes and skin facets, elegant modern art style, symmetrical framing, luxury editorial vibe, 4k ultra detailed, masterpiece, trending on artstation, stylized low-poly realism.
Social Media Post - Luxury Studio Ballet Portrait Grid — output

Social Media Post - Luxury Studio Ballet Portrait Grid

麻酱AI实验室GPT Image 2
# Light and Shadow Ballet Photo | 3x3 Grid Version | Open Source Stable Prompt Please generate one 3x3 grid puzzle photo based on 1-3 clear personal photos uploaded by the user. The final product should be a square puzzle composed of 9 different small images, suitable for direct posting on X. Each frame should look like a different shot from the same high-end photo set, with a unified theme, styling logic, color palette, and spatial atmosphere. Theme: "High-Luxury Light and Shadow Ballet Photo" Overall Reference Vibe: High-cost studio photography, ballet girl, swan princess, European court elegance, dreamy backlighting, crystalline snowflakes, gorgeous mirrors, crystal chandeliers, light ethereal aura, expensive custom styling, strong atmospheric lighting. --- ## I. Character Consistency Please strictly preserve the user's real identity characteristics, including but not limited to: - Face shape - Facial feature proportions - Eye shape - Nose shape - Lip shape - Skin tone - Apparent age - Hairline - Hairstyle foundation - Body type - Overall temperament The 9 small images in the grid must clearly look like the same real girl; they must not become a stranger, be westernized, become an influencer style, have excessive skin smoothing, or feature an AI fake face. --- ## II. Overall Style Positioning The set of images must present a high-end photo quality that looks like it "cost a lot of money to shoot." It is not an ordinary princess style or cheap cosplay, but rather: - High-end studio lighting - Expensive makeup and hair styling - Haute couture ballet dresses - Detailed hair accessories and crowns - High-end post-production atmosphere - Combining cinematic and jewelry-like quality Style Keywords: - Light and shadow ballet - White swan princess - European magnificence - Dark-field backlighting - Crystal lamps - Classic floor mirrors - Falling snow / Flying feathers / Sequins - Luxurious dreamscape - Ethereal spirit - High-end custom photos - Tang Vision style high-cost studio aesthetics, but without any text watermarks. --- ## III. Scene Setting The scene is a European dream space inside a dark-toned high-end photography studio. Spatial elements can include: - Black or dark background space - Golden vintage floor mirrors - European carved mirror frames - Crystal chandeliers - Ground glitter / sequins / snowflake reflections - Floating snowflakes / feathers / shimmering dust in the air - A few vintage clocks or European ornaments - Mirror-like reflections or sparkling textures on the floor The overall space must have a clear black-and-gold contrast, using strong backlighting and rim lighting to create an "expensive" atmosphere. Do not show ordinary home backgrounds. Do not show modern messy props. Do not show brand logos. Do not show "TANG VISION" text. Do not show Xiaohongshu watermarks. Do not show garbled text. --- ## IV. Clothing and Styling Clothing is haute couture light and shadow ballet dresses. Core Clothing Elements: - White, cream, champagne, or pearl white primary colors - Ballet short dress / tutu structure - Layers of tulle - Feather decorations - Beadwork, rhinestone embellishments, silver threads, fine chains - Corset bodice structure - Light fluffy skirts - Feather accents on shoulders or chest - Ballet slippers / satin ribbon shoes / pointe-style footwear The styling should look like a custom creation from a high-end studio—not cheap, not a simple wedding dress feel, and not a standard evening dress feel. Accessories include: - Exquisite crowns / light tiaras / crystal headpieces - Delicate shimmering earrings - Minimal facial sequins or tear rhinestones - Feather neckpieces or shoulder feather details --- ## V. Hairstyle Requirements (Very Important) Hairstyles must be extended based on the user's real hair; it is not allowed to forcibly unify them into the blonde hair from reference images. Please strictly follow these rules: - Retain the user's real hair color; do not bleach it blonde without permission. - Retain the user's original hairline and bangs type. - Retain the user's original hair length and overall volume. - Extend the "light and shadow ballet" styling on this basis. Flexibility based on original hairstyle: - Long hair: Elegant low bun / ballet bun / half-up style, retaining a few natural wisps. - Medium hair: Light bun or half-tied princess style, keeping it light. - With bangs: Retain and soften. - Without bangs: Use natural side wisps to frame the face. - Curly: Maintain curl base with more refined arrangement. - Straight: Smooth ballet bun or half-up style. Core Principle: Hairstyle is a "high-end upgrade based on the user's original hair," not a rigid template. --- ## VI. Makeup Requirements Makeup should reflect "wealth, royalty, photogenicity, and dreaminess," without looking like a fake face. Makeup Keywords: - Translucent cream skin - Cool-toned highlights - Exquisite eye makeup - Clear lashes - Lower lid shimmer - Pale pink, rose almond, champagne eye tones - Peach or cool pink blush - Hydrated yet restrained lips - Minimal facial sequins / tear diamonds / shimmer - Looking like high-end custom makeup that took a long time to complete Makeup Requirements: High completion, photogenic, noble, ethereal, but still retaining real skin texture; no plastic skin. --- ## VII. Grid Shot Design Please generate 9 different small images and combine them into a 3x3 grid puzzle. The 9 images must be clearly different, including full body, medium shot, half body, close-up, and extreme close-up, with varying poses and emotions. ### 1 Back to the camera turning slightly, looking at the lens, near a golden floor mirror, half-body close-up, graceful shoulder and back lines, falling snowflakes. ### 2 Standing half-body or 3/4 body, slightly angled, strong backlighting on the crown and shoulder feathers, cold and quiet aura. ### 3 Wide-angle full body sitting on the floor, next to a floor mirror, legs naturally tucked or to the side, looking up at the crystal chandelier, dreamy atmosphere. ### 4 Full body standing, ballet gesture, one hand raised like a stage still, crystal chandeliers and classic mirror frames in the frame. ### 5 Medium close-up sitting, skirt spread out, hand lightly on chest or shoulder, very strong backlighting, highlighting the layers of the expensive dress. ### 6 Half-body or medium shot looking at light, ethereal expression, like a white swan lit by stage lights. ### 7 Wide-angle sitting on the ground, head down, body slightly curled, arms around knees, quiet and fragile, like a captured rest moment. ### 8 Medium close-up sitting, hand on face, looking at lens, dreamy and soft expression, like a classic cover feel in high-end samples. ### 9 Medium or half body, looking down at the skirt or feather in hand, quiet and gentle movement, acting as a more relaxed shot in the grid. --- ## VIII. Photography and Lighting Photography quality must be very high-end, like expensive studio samples: - Strong backlighting - Rim lighting - Edge glow - Chandelier reflection highlights - Mirror refraction - Volumetric light - Shimmering particles - Floating snowflakes / sequins - Black-and-gold contrast - High-end dreamy film quality - Clear yet soft skin representation The image must have a clear "high-budget shoot" feel. --- ## IX. Puzzle Requirements Final output: one square 3x3 grid puzzle. All 9 images maintain unified theme, styling, and tone. --- ## X. Negative Requirements No identity changes. No mismatched hairstyles. No ordinary wedding styles. No cheap princess dresses. No plastic skin. No AI fake faces. No deformed hands. No extra fingers. No wrong proportions. Not all grids from the same angle. Not all grids with the same expression. Not all grids just standing. No repeated compositions in the puzzle. --- ## XI. Supplementary Stability The 9 images must look like the same high-end sample set: same person, same styling system, same spatial system, same color system, same luxury theme, but with obvious differences in distance, pose, expression, and emotion. The overall effect should be a professional studio sample ready for sale or display.
Profile / Avatar - Magic Academy Graduation Portrait System — output

Profile / Avatar - Magic Academy Graduation Portrait System

麻酱AI实验室GPT Image 2
Please generate **9 different individually refined graduation portrait photos** based on the 1-3 clear personal photos uploaded by the user, all in **{argument name="composition ratio" default="3:4 vertical composition"}**. 
The theme is:

**"{argument name="portrait theme" default="British Magic Academy Graduation Portrait"}"**

---

## I. Character Consistency

Please strictly preserve the user's true identity features, including but not limited to:

- Face shape
- Facial feature proportions
- Eye shape
- Nose shape
- Lip shape
- Skin tone
- Age feeling
- Hairstyle basis
- Body type
- Overall temperament

In all 9 images, it must clearly look like the **same real girl**. 
Do not turn into a stranger, do not westernize, do not use internet celebrity aesthetics, do not over-smooth skin, and do not use AI fake faces.

Please only use the uploaded photos as **identity references**, do not keep the realistic background, daily clothing, or original composition from the source photos. 
Please regenerate a completely new set of **British Magic Academy Graduation Portraits**.

---

## II. Overall Positioning

The focus of this set is not a realistic portrait, but rather:

**"British Magic Academy Graduation Theme + Surreal AI Magic Scenes + Movie Poster Visual Impact."**

The entire set should resemble:

- A set of high-concept magic academy movie promotional posters
- Clearly filled with AI spectacle visuals
- Featuring exaggerated space, grand narratives, and a strong magical atmosphere
- Preserving both the graduation identity and a refined portrait feel

Overall Keywords:

- British
- Academy
- Graduation
- Magic
- Movie poster feel
- Floating pages
- Flying books
- Giant owl
- Starry dome
- Magic light trails
- Gothic library
- Mysterious candlelight
- Surreal space
- High-end feel
- Strong narrative feel

---

## III. Clothing Requirements

Clothing should follow the logic of **"Academy Graduation Robe + Magic Elements."**

Requirements:

- Preserve graduation robe / academic gown silhouette
- Preserve academic cap / mortarboard system
- Add academy scarves, badges, crests, trimmings, ties, and embroidery details
- Inner wear can include white shirts, ties, knit vests, pleated skirts, or academy long skirts
- Overall looks like a commemorative portrait style for an "outstanding magic academy graduate"
- Clothing is unified, complete, and has a cinematic quality
- Do not make it look like a stage costume
- Avoid cheap cosplay feel

---

## IV. Scenes and Surreal Spectacle Requirements

The 9 images should revolve around the magic academy world-building and be significantly more exaggerated, dreamlike, and cinematic than ordinary portraits.

High Priority Visual Elements:

- Towering bookshelves
- Gothic library
- Magical book mountain
- Floating pages
- Flying books
- Giant owl
- Candlelit hall
- Glowing scrolls
- Starry dome
- Deep blue cosmic sky
- Glowing magic lines
- Spell light trails surrounding the character
- Blue-white magic particles
- Backlit smoke
- Giant ancient book covers
- Ritualistic academy certificates
- Light beams through arched windows
- Large magical creature guardian composition

Overall should feel like:
**Magical academy movie poster spectacle scenes that only AI can create.**

---

## V. Lighting and Image Quality Requirements

Lighting Requirements:

- Cool blue moonlight / starlight
- Warm golden candlelight
- Partial backlighting
- Magic glowing particles
- Atmospheric smoke
- Rich layers of light and dark
- Clear character face
- Clear facial and hand details
- Cinematic lighting

Quality Requirements:

- Realistic photography texture
- High precision
- High-end feel
- Cinematic feel
- No ordinary photo studio feel
- No dull backgrounds
- No plastic skin
- No messy or cheap CG feel

---

## VI. Fixed Allocation of 9 Individual Photos

Please output the following **9 different individually refined photos** at once:

### 1st | Main Visual Poster
- Full body or 3/4 body
- Character standing in the center of a towering magic library
- Surrounded by a large number of floating pages
- Holding a wand or graduation scroll
- Very strong main movie poster visual feel

### 2nd | Spell-casting Close-up
- Half-body close-up
- Character holding a wand
- Pages flying in front and behind the camera
- Background has a glowing starry dome or magic canopy
- Emphasize face, hands, and spell-casting feel

### 3rd | Giant Owl Guardian
- Medium shot or full body
- A giant owl appears behind or to the side
- Character holding a book, scroll, or lightly raising a hand for a spell
- Clear narrative sense of guardianship and companionship

### 4th | Surrounding Light Trail Spell-casting
- Full body
- Character standing and casting a spell
- Glowing magical rings circling around
- Books pages floating in the air
- Stronger atmosphere, more dramatic

### 5th | Starry Library Contemplation
- Half-body or medium close-up
- Character set in a Gothic library
- Background is a deep blue starry sky, candlelight, glowing pages
- Holding an ancient book or magic scroll
- Expression is quiet, smart, and mysterious

### 6th | Book Mountain and Flying Books
- Medium shot
- Character standing among stacked mountains of books
- Flying books circling overhead and around
- Exaggerated composition, large spatial scale
- Clear surreal spectacle effect

### 7th | Graduation Identity Reinforcement
- 3/4 body
- Character clearly displaying the graduation robe, academy scarf, badges, and cap system
- Holding a graduation certificate or academy scroll
- Background is academy hall / candlelit auditorium / bookshelves

### 8th | Low-angle Heroic Poster
- Full body or medium shot
- Low-angle shot
- Character posture has more presence
- Pages, light effects, and particles extending upwards from around the character
- Like a final protagonist poster after graduation

### 9th | Emotional Conclusion
- Half-body close-up
- Character in a quiet mysterious library corner or in front of an arched window
- Can include candlelight, feathers, glowing dust, slightly floating pages
- One hand lightly touching an ancient book or scarf
- Serving as the most emotional and story-driven image in the set

---

## VII. Unified Requirements

The 9 images must be unified in:

- The same real face
- The same British magic academy graduation styling system
- The same academy color scheme
- The same world-building system
- The same movie poster atmosphere
- The same surreal magic spectacle style

The 9 images must have clear variations in:

- Camera distance
- Perspective height
- Actions
- Expressions
- Prop interaction
- Focus of magic spectacle
- Composition methods

---

## VIII. Explicit Avoidance

- No dull portraits
- No cheap cosplay feel
- No modern campus daily style
- No cheap studio magic style
- No messy background stacking without a focus
- No character face changes
- No westernization
- No internet celebrity faces
- No over-smoothing of skin
- No AI fake faces
- No incorrect fingers
- No incorrect body structures
Profile / Avatar - Luxury Ballet Couture Portrait Series — output

Profile / Avatar - Luxury Ballet Couture Portrait Series

麻酱AI实验室GPT Image 2
Please generate **9 different individually refined portrait photos** based on the 1-3 clear personal photos uploaded by the user, all in **{argument name="composition" default="3:4 vertical composition"}**.

The theme is:

**"{argument name="theme name" default="Luxury Light and Shadow Ballet Style Portrait"}"**

Atmosphere reference: High-budget sample photos, European dreamlike, swan princess, stage backlighting, crystal chandeliers, vintage gold mirrors, floating snowflakes and feathers, white gold haute couture ballet dress, ethereal yet noble, resembling expensive photo studio sample shots.

---

## I. Character Consistency

Please strictly preserve the user's true identity features, including but not limited to:

- Face shape
- facial proportions
- Eye shape
- Nose shape
- Lip shape
- Skin tone
- Age feeling
- Hairline
- Hairstyle basis
- Body type
- Overall temperament

The same real girl must appear in all 9 photos. Do not turn her into a stranger, do not westernize, do not use internet celebrity styles, do not over-smooth the skin, and do not use AI fake faces.

---

## II. Overall Style Positioning

This is not an ordinary fairy portrait or a standard wedding studio shoot. It is:

**High-cost Studio Ballet Samples + European Mirror Hall Dream + Magnificent Light and Shadow Epic**

The overall feel must satisfy:
- Looks like the hair and makeup were expensive
- Looks like the lighting was expensive
- Looks like the clothing was expensive
- Looks like the post-processing was expensive
- Every image can serve as a commercial cover

Keywords: Light and shadow ballet, White swan / Snow feather princess, European vintage, Mirror hall, Crystal chandeliers, Classic carved mirror, Snow effect, Haute couture feather ballet dress, Fragile elegance, Noble, Ethereal, Stage presence, Cinematic poster feel.

---

## III. Scene Setting

The scene is a **high-end European studio in a dark setting**.
Spatial elements can be combined flexibly:
- Vintage gold floor mirror / European carved mirror
- Black or dark background space
- Crystal chandeliers
- Sequins, glitter, icy reflections on the floor
- Floating snowflakes / feathers / fine glitter fragments in the air
- Minimal European decoration
- Moderate mirror reflections
- Local strong light beams / light spots / lens flare

The space must look like a high-end sample studio, not a home setting.

---

## IV. Clothing and Styling

Clothing is **Haute Couture Ballet Style**.
Structure suggestions:
- White / Cream white / Pearl white / Champagne white
- Feathers, tulle, sequins, silver threads, diamond details
- Corset-style top
- Voluminous tutu or short ballet skirt hem
- Lightweight feather shoulder decorations
- Ballet shoes / ribbon-tied dance shoes / pointe-style footwear
- Small touches of pearls, chains, or diamond embellishments

It must be: Magnificent, dreamlike, expensive, detailed, and looking like custom-made sample clothing.

---

## V. Hairstyle Requirements (Must follow user's original hairstyle)

Hairstyles must be an advanced ballet-style extension of the user's actual hair, strictly adhering to:
- Original hair color
- Original hairline
- Original bang type
- Original length and overall volume trend
- No forced blonde hair or specific bun templates

Upgrades can include: Elegant buns, French half-updos, Princess updos, or soft semi-tied styles decorated with crystal crowns, light tiaras, feather hairpieces, or pearl accessories.

---

## VI. Makeup Requirements

Studio-level refined makeup: Flawless cream skin, sophisticated highlighting and contouring, delicate lashes, under-eye glitter, peach/rose/champagne eye tones, cool or neutral highlights, slightly flushed cheeks, and polished lips. Must be dreamlike and expensive but not a fake face.

---

## VII. Camera Design for 9 Individual Photos

1. **Mirror Reflection Close-up**: Half-body near the mirror, looking back at the camera. Focus on shoulder line, crown, and eyes with floating snowflakes.
2. **Backlit Semi-profile**: 3/4 body shot with strong backlighting on the bun and dress edges like a stage light.
3. **Grand Scene Sitting**: Sitting on the floor near the mirror, skirt spread out, looking up at the chandelier. Full-body shot.
4. **Full Body Ballet Pose**: Elegant ballet arm movements, long leg lines, voluminous skirt. Main visual poster style.
5. **Cover Style Sitting**: Sitting on the floor/platform, hand on chest/collarbone, mid-close-up shot.
6. **Looking Up at Light**: Standing or sitting, looking up at the light source with a quiet, dreamlike expression. High detail on crown and feathers.
7. **Quiet Curled Pose**: Sitting on the floor holding knees, soft and relaxed, like a backstage candid shot. Atmospheric and fragile.
8. **Dreamy Portrait with Hand on Face**: Sitting by the mirror, leaning face on one hand, soft expression. Commercial cover style.
9. **Looking Down at Skirt**: Gentle and introspective shot looking at dress details or feathers, a quiet concluding image.

---

## VIII. Expressions and Emotions

Distribute different emotions: Quiet gaze, gentle look back, ethereal looking up, noble distance, soft dreaminess, thoughtful, fragile, slight melancholy, and relaxed stillness.

---

## IX. Photography and Quality

Must have high-cost texture: Realistic photography, rim lighting, atmospheric lighting, crystal reflections, soft skin, high-contrast dark gold background, dynamic floating elements, and lens flare. Cinematic lighting quality.

---

## X. Negative Requirements

Do not change identity. No cheap princess dresses. No standard wedding styles. No internet celebrity filters. No plastic skin. No fake faces. No extra/deformed fingers or body structures. No repetitive poses or expressions.
Infographic / Edu Visual - Manga Recipe Illustration Generator — output

Infographic / Edu Visual - Manga Recipe Illustration Generator

Eris Create LabGPT Image 2
Please create a "{argument name="recipe type" default="easy recipe introduction illustration page"}" using the character from the reference image, with a humorous and relaxed atmosphere.

Use the provided image only as a character reference. Do not reuse the composition, pose, background, camera angle, or layout of the original image.

The AI should freely devise the recipe content.

The cooking genre should be simple dishes within a realistic range, such as {argument name="cooking genre" default="lazy meals, simple snacks, late-night snacks, budget meals, junk-style snacks, leftovers cooking, convenience store arrangements, or quick side dishes"}.

The overall screen should have the atmosphere of a Japanese online recipe article, a magazine bonus page, a casual personal blog, an SNS meme image, or a relaxed manga.

Structure: - Include multiple step-by-step panels - Put an SD/chibi version of the character in each step - Include reactions or small jokes during cooking - Include flustered faces, smug faces, a sense of messy work, small accidents, mysterious momentum, etc. - Include hand-drawn style arrows, notes, supplementary comments, and rough warnings - Make the layout slightly cluttered - Ensure high information density but keep it readable and organized.

Even if the presentation is gag-oriented, make the food look appetizing enough to make someone think, "Maybe I'll try making this."

At the bottom of the image, place a large main cut showing the finished dish.

Include a scene where the character themselves is eating the dish, with a speech bubble for their impressions.

The impressions should be surprisingly persuasive, silly, weirdly high-energy, but making one want to eat it.

Important: - Do not make the food look unappetizing - Do not go in the direction of high-end cuisine or formal restaurants - Do not use an overly clean minimal design - Do not make it a lifeless infographic - Emphasize manga-style reactions and a sense of daily life - Balance a relaxed atmosphere with practicality - Maintain the character's face and identity.
Social Media Post - Gothic Lolita Furisode Outfit — output

Social Media Post - Gothic Lolita Furisode Outfit

白亜@AIチャット民GPT Image 2
{argument name="japanese style" default="furisode lolita"}, lace, frills, {argument name="belt" default="leather belt"}, tights, {argument name="boots" default="lace-up long boots"}
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 ↗