GPT Image 2 prompts

Vertical healing-style home product advertising photography, soft light low contrast, off-white cream tone, macaron contrasting color textile texture as core visual tone. Blank space at the top for large handwritten brush-style Chinese title font, ochre brown.
Main visual is a close-up of [Subject], lying lazily prone, body wrapped in a fuzzy blanket exposing only head and upper body, chin resting on the edge of the blanket, eyelids slightly squinted, expression peaceful and relaxed.
Blanket has chenille/flannel texture, patchwork of pink, light blue, light green, yellow-brown color blocks, fluffy fibers. Background is solid off-white to cream yellow or simple home scene, sofa outline blurred and visible.
Light source is large-area soft diffuse light, no obvious hard shadows, highlights concentrated on subject and blanket fuzz creating a glowing feel. Medium-long telephoto lens compresses perspective, eye-level slightly downward angle, shallow depth of field blurs background. Overall high brightness low saturation, balanced warm and cool tones.
Small logo composed of letter V and cat pattern in top left corner; extremely small lowercase voxcat label at approximately 5% edge position on the right, ending with a lapis lazuli blue small dot; if the subject can wear accessories, add a small VOXCAT tag; logo positions fixed, sizes restrained, not overshadowing the main subject.

OUTFIT:

An oversized black and red tartan check flannel shirt, worn partially unbuttoned over a simple black slip dress.

Styled with layered necklaces and optional combat boots.

Cool-casual, edgy streetwear aesthetic.
Black, red, and black color scheme.
No extra decorations.

Output a single independent 4:5 vertical commercial poster, no collage, no multi-grid generation. Use an original fictional brand, do not imitate existing brand packaging, logos, or advertisements. Text should only use the brand name, main title, and category explicitly listed in the prompt, do not fabricate prices, award information, parameters, dates, origins, or slogans. The image should reach a level of completion suitable for creative direction review, with realistic structure, credible materials, clear layout, and no watermarks.

Theme: Dessert audit sample. Original brand "SOFT LAB", product "CARAMEL PUDDING", main title "MELT".

Use a warm cream-colored seamless background. Place a real caramel pudding slightly below center, with a smooth surface featuring fine air bubbles, caramel sauce flowing slowly down one side, and soft contact shadows on the white porcelain plate. In the middle background, stand four thick but not exaggerated cream-colored letters "MELT", with letter material resembling cooled fine sugar frost.

Core composition type: "Single Dessert - Temperature Direction - Typeface Gradient". Establish a clear temperature change from left to right: M remains dry crystalline; E edges show slight caramel wetness; L lower part slightly softens; T's bottom corner contacts caramel sauce and slowly melts. Changes must be continuous, restrained, and gravity-compliant, with all four letters remaining fully readable. Use two or three extremely thin hand-drawn arrows to annotate "Cooling / Contact / Softening", do not add long explanations. Write "SOFT LAB" at the top, "CARAMEL PUDDING" at the bottom.

Colors should only use cream white, caramel brown, and a small amount of dark coffee. Avoid multiple desserts, food splashes, exaggerated lava, cartoon expressions, recipe information, prices, promotional stickers, or children's illustrations.

A split-screen comparison of two train station scenes at dusk, separated by a thin vertical line. Left panel: A quiet, well-maintained train station named "RIVERDALE" with warm yellow lighting under the platform canopy, a wooden bench, a round clock on the wall, clean concrete platform with yellow tactile paving edge, curved railway tracks leading into the distance, signal lights glowing red, and a vibrant sunset sky with orange and purple hues over distant hills. Right panel: An abandoned, dilapidated train station named "RAVENSBROOK" with peeling paint, broken windows, overgrown weeds and grass between cracked concrete platforms, rusted railway tracks, weathered wooden support beams, dark and eerie atmosphere, similar sunset sky but more muted and cloudy. Bottom labels: "QUIET" in white text on dark background beneath left image, "ABANDONED" in white text on dark background beneath right image.

A hyper-realistic, cinematic horror scene featuring a massive werewolf crouching atop the rusted remains of an abandoned merry-go-round in a desolate playground. The creature is covered in matted dark fur and tattered rags, with glowing orange eyes and sharp teeth bared in a ferocious snarl. Its claws grip the metal bars tightly. To the left, a rusty swing set hangs empty; to the right, a weathered wooden sign reads "PLAYGROUND RULES: HAVE FUN BE SAFE SHARE SMILE". In the background, a dilapidated gazebo stands among bare trees under a stormy night sky illuminated by a full moon and jagged lightning bolts. Heavy rain falls throughout the scene, creating wet reflections on the ground and enhancing the ominous atmosphere.

A nostalgic, film-grain photograph of a young Japanese girl sitting on the polished wooden floor of a school gymnasium. She is wearing a classic red velour tracksuit with white piping along the sleeves and legs, typical of Showa-era retro fashion. Her dark hair is tied in low pigtails secured with small red beads, and she has a gentle, natural smile. A white towel is draped around her neck, suggesting she has just finished physical activity. She holds a worn, off-white leather volleyball against her chest with both arms crossed over it. The background features blurred details of a gym setting, including a metal rack holding more volleyballs and large windows letting in soft daylight. The image has a warm, slightly faded color palette characteristic of vintage 1980s photography.

Transform the uploaded photograph into a minimalist vintage travel-journal collage. Match the provided two-panel reference style as closely as possible.

CANVAS:
Vertical 3:4 composition with two horizontal sections of approximately equal height.

TOP HALF — ORIGINAL PHOTO:
Preserve the uploaded photograph and its original composition, location, architecture, objects, colors, and perspective. Keep the image photographic, natural, slightly soft-focused, and cinematic. Do not turn the top half into an illustration.

BOTTOM HALF — SIMPLE FLAT ILLUSTRATION:
Create a deliberately simplified, minimalist watercolor-and-ink illustration of the EXACT SAME SCENE.

IMPORTANT: The bottom illustration must look like the simple, faded illustration in the reference image, NOT a detailed watercolor painting.

Use:

Simplified architectural silhouettes

Flat, translucent watercolor washes

Minimal outlines and loose brushwork

Faded terracotta and dusty blue

Muted olive green and charcoal

Large areas of empty cream paper

Simple human silhouettes

Very few architectural details

Soft, imperfect edges

Do not reproduce every photographic detail. Preserve only the most recognizable elements of the original scene.

The illustration should feel understated, airy, slightly faded, and hand-painted on vintage paper.

LAYOUT:
Keep the photograph in the upper half and the illustration in the lower half. Separate them with a simple, subtly irregular horizontal edge. Avoid elaborate torn-paper effects.

PAPER:
Warm ivory paper with subtle natural grain and a matte finish. The lower section should have plenty of empty space.

TEXT:
At the bottom of the illustrated section, add a small lowercase typewriter-style heading followed by two short lines of reflective text relevant to the scene.

Use dark gray monospaced letters, slightly faded ink, and generous spacing. Keep the text small and understated.

STRICT STYLE RULES:
The top must remain a photograph.
The bottom must remain a simple flat illustration.
Do not make the bottom photorealistic.
Do not use highly detailed watercolor rendering.
Do not fill the entire paper with paint.
Do not add extra panels, borders, decorative elements, or random text.
Preserve the original scene's recognizable identity.

The final image must resemble a minimalist travel-journal page with a photograph above and a simple illustrated interpretation below.

Aspect ratio: 3:4 vertical.

A vintage-style photograph of two young East Asian women standing close together in front of a traditional Japanese shop. The woman on the left has dark hair styled in two braided buns and wears a navy blue dress with a small white floral pattern and ruffled collar, holding a bouquet of blue flowers. The woman on the right has short, choppy black hair and wears a maroon sweater over a white collared shirt, holding a bouquet of pink cosmos flowers. Both have natural skin textures with visible freckles and slight acne, looking directly at the camera with neutral expressions. The background features wooden shelves stocked with various bottles and boxes, including signage with Japanese characters like '清酒' (sake) and '白鶴'. The image has a warm, faded color palette typical of old film photography, with soft lighting and a slightly grainy texture.

A realistic illustration of the model in the attached image, preserving her face and expression. Retain the head from the background of the original image, with the {argument name="zodiac sign" default="Virgo"} symbol positioned at the top. The entire setting is suffused with a glow of {argument name="energy color" default="golden"} energy, luminous sacred geometry particles, and cosmic {argument name="nebula type" default="golden nebulae"}. Color palette: pure gold, neon amber, and deep black. High-resolution visual quality with a metallic finish.

A vintage-style photograph of a young woman at the beach, shot on film with visible grain and light leaks. She is looking back over her shoulder with a bright, genuine smile. She wears a large woven straw sun hat with a dark brown trim and a white eyelet lace top. Her hair is styled in two low braids or pigtails. The lighting is warm and sunny, casting dappled shadows from the hat onto her face. In the background, there are blurred figures swimming in the ocean and a green hillside.

Wildlife dynamic photography × decisive moment × minimalist environmental narrative.

The key for this type of image is placing the [interaction prop] on the action path, letting hand, prop, and animal form a clear action line, then using high-speed shutter to freeze the 'about to touch' moment, making the image more lively than simple posing.

Model: gpt-image2

Prompt:

Dynamic interaction photography between [Subject] and [Animal], person not fully shown, only partial hand, forearm, or body edge entering frame, establishing direct contact with [Animal] through [Interaction Prop].

Animal is in a moment of action: pouncing, reaching, biting, grabbing, lightly touching, or approaching the prop, forming a strong decisive moment.

Composition emphasizes action line, prominent subject, minimalist background, realistic environment, natural daylight, high-speed shutter freezing action, with real snapshot feel, life feel, and narrative feel. Background uses large area of simple natural environment color field, highlighting subject and interaction relationship.

Convert the entire uploaded photograph into a clean, colorful, hand-drawn 2D cartoon illustration. Preserve the original composition, camera angle, poses, facial expressions, recognizable identities, hairstyles, clothing, accessories, and all important background elements.

ILLUSTRATION STYLE:
•Clean black outlines with slightly varied, hand-drawn line weight.
•Flat pastel colors with minimal cel shading.
•Simplified characters with cute proportions, expressive faces, and natural poses.
•Smooth, slightly imperfect lines for an authentic illustrated feel.
•Simplify realistic textures into clean shapes while retaining recognizable details.
•Recreate the original environment, including buildings, trees, fountains, pathways, and other objects, in the same cartoon style.
•Use a soft pastel palette with light blue skies, muted greens, warm skin tones, and gentle pinks.
•Add subtle artistic details while maintaining the original scene’s atmosphere.

PRESERVE THE ORIGINAL:
Keep the same number of people, their positions, body proportions, clothing colors, interactions, and relative sizes. Maintain the original background layout and perspective. Do not add, remove, or reposition important objects.

OUTPUT REQUIREMENTS:
A polished, full-frame 2D illustration with consistent outlines, flat colors, and a playful, storybook aesthetic. The entire image must be illustrated, with zero photorealistic elements.

AVOID: 3D rendering, photorealism, excessive shading, watercolor textures, blurry lines, distorted faces, extra fingers or limbs, changed outfits, altered poses, missing background details, and unwanted text or watermarks.

Aspect ratio: Match the uploaded photo exactly.

Pro tip: Upload one photo at a time and use this same prompt to transform different photos into a consistent cartoon style.

A vertical medium-full shot of a young Asian woman sitting on a rocky pebble beach, looking directly at the camera with a calm, slightly melancholic expression. She has dark hair tied back in a low ponytail with loose bangs framing her face. She is wearing a traditional Japanese sailor school uniform (seifuku) consisting of a white short-sleeved blouse with a navy blue collar featuring white stripes and a red neckerchief, paired with a dark navy pleated skirt. On her feet, she wears white ankle socks and brown leather loafers. Her pose is relaxed, sitting with her legs extended forward and slightly bent, hands resting near her lap. The background features a bright blue sea with gentle waves crashing against the shore, scattered rocks, and a clear sky with faint clouds. The image has a strong retro film aesthetic, characterized by warm tones, visible grain, soft focus, and light leaks or color shifts along the edges typical of vintage instant film photography.

Create a modern sneaker promotional poster featuring a [shoe type] with [shoe appearance] as the main foreground visual. Use a fixed visual system of warm white, graphite black, and burning orange. Realistic commercial product photography with clear details of fabric, foam, rubber, and unified lighting. The shoe is complete and occupies about half of the frame; the scene serves the shoe's purpose.

Execute corresponding scenes based on [shoe type]:
Road Running Shoes: Early morning city elevated track, low-angle three-quarter view, toe pointing upper right; ground lines converge into the distance, combined with moderate motion blur to express propulsion.
Trail Shoes: Rock layers, forest slopes, and cold gray sky, wide angle close to the ground side-front; deep-toothed sole crosses gravel, kicking up a little dust, burning orange path climbs along the slope.
City Training Shoes: Warm white concrete training space, skylights and side light, low-angle upward shot, toe pointing upper left; short orange folded lines around the sole to express turning and stable support.

Integrate large [English action words] into the selected environment: Road models become perspective-thick track components; Trail models are embedded in cut rock layers; Training models fold along walls, columns, and floors. The font uses warm white front faces and burning orange cuts, with realistic occlusion, thickness, and shadows, maintaining recognizability.

Place [Product Name] and "CONCEPT FOOTWEAR" at the top; place [Slogan] and [Structure Labels] at the bottom. Small text should be concise and clear, using modern sans-serif fonts, ensuring hierarchy between product, spatial title, and auxiliary information.

Output only one independent poster each time. Avoid real brand logos, people, fictional performance numbers, irrelevant text, shoe deformities, plastic textures, and excessive decoration.

Create a 4:5 vertical alcohol promotional poster, outputting only one independent finished piece at a time.

Fixed visual language: Realistic commercial still life photography, warm ivory negative space, huge high-contrast serif titles, restrained editorial typography. Use the wine bottle and transparent medium to refract text, combined with bright channels formed by materials on both sides; no neon, glowing outlines, generic e-commerce cards, or golden luxury decorations.

Arrange [product] according to [scene and lens]. Let [materials on both sides] enter the frame in close-up, partially cropped, forming asymmetric vertical negative space. The opening can slowly expand and contract with the lens, do not make it a mirrored doorframe. The bottle occupies about half to two-thirds of the screen height, fully recognizable in the negative space, with slight occlusion from one side's material; the upper part should still have breathing room. The materials on both sides must come from the storage, packaging, or drinking context of this liquor, without adding irrelevant caves, flower fields, or uniformly replicated photographic backgrounds.

Place the large title [title] across layers in the upper middle section, letting a small segment of its strokes pass behind the wine bottle or transparent medium, undergoing local magnification, displacement, or color filtering in the [optical event]. Parts of the text not passing through the medium remain normally clear; deformation is limited to the light-transmitting area, cannot twist the entire title into waves. Use [primary color] as the only strong color, assisted by warm ivory and charcoal black; light transmission through the liquid, glass thickness, contact shadows, condensation, and projections must be credible, do not draw optical effects as lasers.

[Brand] is placed at the top in a small, sparse sans-serif font; [Category] and "Drink Responsibly" are at the bottom as two lines of small text. The bottle label uses only the brand and category, leaving clean information density. Only the above text is allowed in the entire image, do not automatically add slogans, vintages, alcohol content, awards, origins, prices, or unprovided product claims. No people in the scene, do not depict excessive drinking.

Preserve delicate natural material textures and precise glass highlights, avoid heavy CGI, floating products, extra bottles, random letters, glare stacking, collages, and watermarks. Scene, lens, bottle shape, and optical events change with variables, but the visual language of warm ivory negative space, serif main titles, and local refraction remains consistent.

Create a vertical 3:4 “Second World” editorial photography poster based on the uploaded photo.

### COMPOSITION
Divide the image into two perfectly equal horizontal halves.

Upper Half: Preserve the original photograph exactly, allowing only proportional cropping. Keep the original subject, colors, composition, natural lighting, and realistic textures. No redraws or illustrations.

Lower Half: Use a warm ivory-white paper background with generous negative space.

### CORE CONCEPT
Identify the photograph’s most recognizable visual structure and transform it into a tangible, three-dimensional photographic object in the lower half.

The object must retain the original photographic textures and look physically real, with visible thickness, subtle folds or curves, natural shadows, perspective, and realistic contact with the paper.

Subject preservation is essential: If the photo contains a recognizable person, building, animal, landmark, or distinct object, make it the centerpiece of the lower half. Preserve its identity, realistic texture, and most of its complete silhouette. Never replace the main subject with background elements.

The lower half must clearly display a substantial photographic object, not merely torn paper or a tiny character.

### CHARACTERS & CAPTION
Include 0–2 minimalist black line-art characters interacting meaningfully with the photographic object, such as lifting, unfolding, pushing, repairing, or entering it.

Add one short, casual, slightly humorous handwritten English caption. Avoid motivational quotes and clichés.

### VISUAL STYLE
Realistic photography + warm ivory paper + minimalist black doodles + subtle handmade collage + independent magazine aesthetics.

Keep the design clean, bright, minimal, and visually striking.

### AVOID
Full-image illustration, generic torn-paper templates, tiny decorative characters, unrelated ornaments, clutter, complex illustrations, and commercial aesthetics.

FINAL RULE: The lower half must feel like a real, three-dimensional piece of the original photograph brought into a second world. The main subject must remain immediately recognizable at first glance, even if its physical properties are creatively transformed.

Create a creative minimalist 3D scrapbook composition featuring a realistic photograph transformed into a thick, floating paper cutout on a warm ivory textured background. For the first scene, showcase a beautiful sunlit modern bedroom with an olive-green bed, walnut furniture, large windows, lush indoor plants, a round mirror, and soft natural sunlight. For the second scene, depict dramatic white coastal cliffs covered in vibrant green grass, deep blue ocean waves, and a striking red-and-white lighthouse beneath a cloudy sky. Make the photograph appear as a curved, folded sheet of thick paper with realistic depth, visible layered edges, soft shadows, and a floating effect. Add tiny hand-drawn black stick-figure characters interacting with the paper, pushing, lifting, or holding its edges, alongside playful handwritten captions such as "{argument name="caption1" default="Another version of my room"}" and "{argument name="caption2" default="Same place, different layer."} Elegant editorial photography, realistic textures, subtle imperfections, minimalist art direction, soft beige palette, creative visual storytelling, premium design magazine aesthetic, crisp details, balanced composition, vertical poster layout.

the abbot
has neglected the wine cellar
and the rats
content to eat rancid grapes
have the run of the place

A vertical digital collage poster featuring a central young woman with messy purple hair, freckles, and a mustard-yellow sleeveless top. She is looking over her shoulder at the viewer with a confident expression, raising one hand to point her index finger upward. Surrounding her are five distinct character portraits arranged in a fan-like layout: top-left shows a punk figure with blue skin, pink mohawk, and sharp teeth; top-center features an anthropomorphic orange cat with yellow eyes wearing a green jacket and black pants; top-right displays a fluffy white creature resembling a sheep or cloud made of petals; middle-left is a grayscale image of a person with dreadlocks, sunglasses, and a gas mask making a peace sign; middle-right shows a surreal baby-like figure with a large head, red sweater, and green pants carrying a bag shaped like another face. The background is a textured beige paper style filled with scattered autumn maple leaves and clusters of red rowan berries. At the bottom, elegant black cursive text reads "Мои планы на осень — Творить!".

Create one premium editorial poster for each uploaded photo, outputting each separately without combining images or reusing other photos’ designs.

FORMAT: Strict vertical 3:4 composition, divided into two equal horizontal halves. The upper 50% preserves the original photograph’s subject, identity, pose, colors, textures, and natural lighting, with only subtle editorial color grading. Never distort the subject.

LOWER HALF: Transform the main subject into a minimalist cut-paper collage using simplified silhouettes, geometric shapes, layered paper, and subtle grain. Remove unnecessary background details while preserving the subject’s recognizable identity and story.

COMPOSITION: Use a split color-field layout with generous near-white negative space above and a solid-color base spanning the entire bottom width. Position the subject near the boundary, optionally adding a simple geometric background panel. Maintain clean horizontal edges, asymmetrical balance, and strong visual hierarchy.

COLORS & TEXTURES: Extract 2–4 colors from the original photo and refine them into a soft, bright pastel palette. Use subtle paper grain, handcrafted edges, and delicate layering. Create depth through overlapping shapes rather than realistic shadows or perspective.

TYPOGRAPHY: Use minimal, elegant editorial text only when appropriate, placed discreetly within the negative space or near the composition’s edges.

STYLE: Sophisticated art-magazine aesthetic combining authentic photography, minimalist paper-art collage, clean color blocks, generous negative space, and refined editorial design.

AVOID: Clutter, complete scene redraws, gradients, heavy shadows, excessive details, distorted subjects, cartoon styles, 3D effects, and generic template-like layouts.

Goal: Create a distinctive, gallery-worthy poster that preserves the original photograph above and artistically reinterprets its essence below.

A vertical portrait photograph in a retro Showa-era style, featuring a young Asian woman looking back over her shoulder directly at the camera. She has long dark brown hair with wispy bangs and wears an elaborate floral kanzashi hair ornament on the left side of her head, consisting of light blue and white fabric flowers with dangling pearl strands. Her skin is fair with visible natural texture, including freckles and small blemishes, giving a raw, unretouched look. She is wearing a traditional yukata (summer kimono) patterned with large blue morning glory flowers on a white background, secured with a solid light blue obi sash. In her hands, she holds a clear glass Ramune soda bottle with a marble stopper near her chest. The background is slightly out of focus but clearly shows a red noren curtain with white Japanese calligraphy and a drawing of a maneki-neko (beckoning cat), suggesting a traditional shop or festival setting. The image has a vintage film aesthetic with warm tones, slight grain, and distressed white borders around the edges.

A ruined cathedral at night. Moonlight through the rose window lays its tracery down the aisle as a path, and one hooded traveller stands on it.

A hyper-realistic, full-body portrait of a female alien with pale, smooth skin and an elongated head, sitting elegantly in a worn, dark leather armchair. She is wearing a voluminous, sheer white tulle wedding dress that cascades around her legs and onto the floor, paired with high-heeled platform shoes. Large silver over-ear headphones rest on her head, with one hand gently touching the ear cup as if listening to music. Her expression is calm and slightly melancholic, with large, dark almond-shaped eyes typical of extraterrestrial depictions. The setting is a dilapidated, post-apocalyptic laboratory or abandoned facility, featuring broken computer monitors with cracked screens, scattered glass beakers, debris-strewn floors, and crumbling concrete walls. Through a large hole in the background wall, a desolate cityscape with ruined skyscrapers is visible under a grey, overcast sky. The lighting is soft and diffused, highlighting the texture of the tulle fabric against the gritty, industrial surroundings.

Soft pastel anime style. A girl wearing a Trick-or-Treat themed Halloween dress.
The outfit is an off-shoulder mini dress in natural beige. On one side of the skirt, there is a large pocket resembling a snack bag.
Candies and small cookies peek out from the pocket opening.
A thin orange ribbon around the waist.
A short cape on the shoulders, with the ends of the cape tied like bag strings.
No hat on the head. Holding a snack bag in one hand and showing the pocket with the other.
Background is in front of a house at night, with lights.
Colors are natural beige, honey, pale orange, and night blue.
Full body, trying to share sweets. A happy smile.