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

10,043 条提示词

Product Marketing - Retro JRPG Magazine Ad — output

Product Marketing - Retro JRPG Magazine Ad

レティシア・ノエルGPT Image 2
Goal: Create a busy 1990s Japanese fantasy RPG magazine advertisement / game box promotional poster for {argument name="game title" default="SynClub SaGa"}, with a dramatic anime schoolgirl warrior and retro console marketing layout.

Canvas: Vertical full-page ad, 2:3 aspect ratio, aged cream paper texture, dense print design, slightly distressed ink, high saturation, black and gold borders, nostalgic 16-bit RPG era aesthetic.

Main layout: The top two-thirds is a fantasy key visual. At the very top, add a narrow Japanese tagline banner using {argument name="top tagline" default="異世界に召喚された女子高生(きなこ)の壮大な冒険が今、始まる!!"}. Beneath it on the upper left, place a huge ornate logo reading {argument name="game title" default="SynClub SaGa"}: blue-and-gold serif letters for “SynClub,” red-and-gold letters for “SaGa,” with decorative fantasy filigree and a small subtitle plaque. The background is split between a bright blue-sky castle kingdom on the left and a dark purple lightning demon castle on the right. On the right side, add vertical glowing Japanese copy using {argument name="vertical tagline" default="光と闇の狭間で、少女は運命を選ぶ―"}.

Main character: Center-right foreground, a teenage Japanese schoolgirl summoned to a fantasy world, long windblown black hair, face intentionally obscured by a soft rectangular blur, wearing a navy sailor school uniform mixed with polished silver knight armor: armored breastplate, right shoulder pauldron, metal gauntlets, brown leather belts, gold buckles, medallion brooch, short pleated skirt, navy neckerchief, and a sheathed or lowered sword angled across the lower left. Her pose is heroic and slightly angled, hair flowing across the logo area.

Story box: On the left under the logo, place a small parchment-like story panel headed “STORY” in English followed by dense Japanese body copy, imitating a game magazine synopsis. Keep the copy small and mostly decorative but visibly Japanese.

Gameplay strip: Directly below the key visual, create exactly 4 bordered screenshot panels with numbered captions. Panel 1 shows a colorful top-down world map with forest, river, and tiny player character. Panel 2 shows an indoor dialogue/event scene with two characters and a text box. Panel 3 shows an action battle in a fiery volcanic battlefield with a bright sword beam. Panel 4 shows a final boss RPG battle against a large purple demon with status boxes and HP/MP interface. Above these panels place a bold yellow-and-red Japanese headline using {argument name="feature headline" default="美麗グラフィック&重厚ストーリー!4つの世界が君を待つ!"}.

Bottom product section: Divide the bottom third into a left product bundle area and a right sales-copy area. Left side must show exactly 3 purchasable items: a tall game box with the same cover art, a black cartridge labeled with the game logo, and a shiny CD soundtrack disc. Also show a small black retro fictional handheld/console device underneath, clearly not a real brand. Add a red stamp-style label meaning “Now on sale,” a price block reading 7,800 yen plus tax, and a note about an original soundtrack CD bonus.

Compatibility row: Along the bottom left, include exactly 5 small hardware/media icons or badges: 1) AI computer, 2) 8-bit system, 3) CD-ROM, 4) 16-BIT, 5) iOS / android series compatible. Use fictional names and avoid any real company logos or real console designs.

Right sales-copy area: Large bold red Japanese promotional text in stacked lines, advertising a grand adventure, emotional story, a girl warrior challenging darkness, and changing the fate of the world. Under it, add smaller black explanatory Japanese copy about a 16-bit-style fantasy RPG with beautiful graphics, moving BGM, and heavy story. At the bottom right, add a fake publisher logo and company block reading “Noel Soft,” with a blue decorative emblem, fictional address, phone number, copyright line, and small credit text.

Visual style: Detailed anime illustration blended with retro Japanese magazine ad design, ornate fantasy logo typography, halftone print texture, gold trim, dramatic lighting, saturated blues, reds, purples, and yellows. The ad should feel slightly overstuffed and playful, like a parody of classic JRPG marketing.

Constraints: Use no real company names, no real console names, no real logos, and no real hardware silhouettes. Keep the main character’s face blurred/obscured. Include exactly 4 gameplay screenshot panels, exactly 3 product bundle items, and exactly 5 compatibility badges. Make the overall page text-heavy and convincingly Japanese, while preserving the fictional branding.
Infographic / Edu Visual - NPB Prospect Ranking Top 30 Graphic — output

Infographic / Edu Visual - NPB Prospect Ranking Top 30 Graphic

GOCHAN(🐯公)GPT Image 2
Goal: Create a dramatic Japanese baseball prospect ranking infographic in the style of a professional MLB prospect list, titled {argument name="main title" default="NPB PROSPECT RANKING TOP 30"}, with a large featured player on the right and a ranked list on the left.

Canvas: Square 1:1 sports graphic, dark navy stadium-at-night background with bright floodlights, blue smoke, subtle sparks, high contrast lighting, glossy broadcast-graphics finish. Use bold condensed athletic typography, metallic white and red accents, thin cyan divider lines, and team-logo style icons.

Header: At the top left, add a small red date ribbon reading 「2026年5月24日時点」. Below it, place the huge title with “NPB PROSPECT RANKING” in distressed white block letters and “TOP 30” underneath, with “30” in bright red. Under the title, add a Japanese subtitle: 「MLB型本卒業基準(野手130打数未満 / 投手50投球回未満)」 and a small explanatory line: 「OVR=総合評価 / FV=将来価値 / ETA=一軍定着見込み / Risk=リスク」.

Featured player: On the right half, show one dynamic male baseball pitcher in a white-and-blue Fighters-style uniform, blue cap, jersey number 11, mid-throw with one arm forward holding a baseball and the other arm extended back. Face should be intentionally blurred or softened like an anonymized sports mockup. Use dramatic rim light from the stadium lights. At the lower right, add the featured name in large Japanese type {argument name="featured player name" default="柴田獅子"}, with romanization below: {argument name="featured romanization and position" default="SHIBATA REO / RHP"}. Beneath that, create four stat tiles labeled OVR, FV, ETA, Risk with values {argument name="featured stat line" default="68, 55, 2027, 3"}.

Main ranking layout: On the left, create exactly 30 ranked entries. The first 5 entries are large horizontal cards stacked vertically, each with rank number, team-style logo, Japanese player name, position, and four stat columns OVR, FV, ETA, Risk. Rank 4 must be highlighted in red with a glowing border as the featured player.

Top 5 visible entries, exactly:
1. 前田悠伍, 投手, OVR 72, FV 60, ETA 2026, Risk 2, gold SoftBank-style logo.
2. 齋藤響介, 投手, OVR 70, FV 55+, ETA 2026, Risk 2, black-and-gold Buffaloes-style logo.
3. 立石正広, 内野手, OVR 69, FV 55, ETA 2026, Risk 2, yellow-and-black Tigers-style logo.
4. 柴田獅子, 投手, OVR 68, FV 55, ETA 2027, Risk 3, Fighters-style diamond logo, red highlighted card.
5. 有薗直輝, 三塁手, OVR 67, FV 55, ETA 2026, Risk 3, Fighters-style diamond logo.

Lower ranking table: Beneath the top 5, create a compact two-column table containing exactly ranks 6 through 30, with 25 rows total split into left column ranks 6–18 and right column ranks 19–30. Include columns: #, 選手名, 位置, OVR, FV, ETA, Risk. Use small team logo blocks at the start of each row and alternating dark blue row fills. The exact visible ranked names should be:
6 石垣元気; 7 孫易磊; 8 松下歩叶; 9 中村優斗; 10 石塚裕惺; 11 常廣羽也斗; 12 内藤鵬; 13 草加勝; 14 杉山遙希; 15 佐藤龍月; 16 横山聖哉; 17 石見颯真; 18 小田康一郎; 19 平川蓮; 20 坂井陽翔; 21 今朝丸裕喜; 22 東松快征; 23 堀柊那; 24 井坪陽生; 25 イヒネ・イツア; 26 村田怜音; 27 宮崎一樹; 28 西村瑠伊斗; 29 石代エトニー; 30 齊藤優汰. Keep stats small but believable and consistent with the top list.

Footer: Add a thin dark footer bar with a Japanese note: 「注:OVRは100点満点の独自総合評価。FVはMLB系スカウティングで使われる将来価値の目安。ETAは一軍定着見込み時期。Riskはリスク評価。」

Constraints: Use only Japanese text where specified, keep the composition dense but legible, no real league trademarks copied exactly, no watermark, no extra ranks beyond 30, and preserve the exact count of 1 featured player, 5 large ranking cards, 25 compact table rows, and 4 featured stat tiles.
Comic / Storyboard - Desert Rebel Convoy Portrait — output

Comic / Storyboard - Desert Rebel Convoy Portrait

Dilshad HussainGPT Image 2
Create a vertical cinematic 35mm film still of {argument name="character name" default="a rugged desert rebel leader"} walking directly toward the camera through a violent sandstorm in a desert war zone. The central subject is one adult man, waist-to-knees visible, centered and dominant, with messy dark hair, a stern intense expression, dusty tan skin, stubble, a partially unbuttoned faded denim shirt with two chest pockets, a silver chain necklace, dark cargo pants, black fingerless gloves, and bracelets on one wrist. He carries exactly two main props: a double-barrel shotgun held diagonally across his torso in his right hand, with wooden stock and twin barrels pointed upward to the viewer’s right, and a heavy rusty metal chain dangling from his left hand. Behind him are exactly four vehicles: two motorcycles flanking him in the midground with bright circular headlights and helmeted riders, and two armored trucks farther back with glowing headlights, roof gunner silhouettes, and soldiers standing in turrets. Add exactly five visible background fighters: two motorcycle riders, two turret soldiers, and one small distant figure between the trucks. The atmosphere is packed with blowing sand, dust clouds, sparks, flying debris, heat haze, and dramatic backlight from a low golden sun, creating a rebellious action-movie mood. Use a shallow depth of field: the man and shotgun are sharp, while the convoy behind is slightly blurred by dust. Color grade with warm ochre, amber, sepia, and smoky gray tones, high contrast, gritty texture, realistic skin and fabric detail, anamorphic cinematic lighting, and dynamic motion. Canvas is vertical 9:16, full-body heroic composition, no text, no watermark, no logos.
YouTube Thumbnail - Canyon Train Action Hero — output

YouTube Thumbnail - Canyon Train Action Hero

Dilshad HussainGPT Image 2
Create a surreal cinematic action portrait in a vertical 9:16 frame, viewed from an impossible high top-down drone angle above a speeding vintage train racing through a narrow desert canyon at sunset. In the foreground, show {argument name="main character" default="a rugged male action hero"} standing powerfully on the roof of the front locomotive car, body facing forward with a confident stance, wearing {argument name="outfit" default="a dark brown leather jacket blowing in the wind, black shirt, black pants, tactical belt, boots, and a silver chain necklace"}. His face should be intentionally obscured or softly blurred, while his hair is dark and neatly styled. He holds exactly 1 revolver pointed downward in his right hand, with a holster strapped to one thigh. The train should have exactly 1 visible front headlight glowing, a dark vintage engine front with windows and grill, and multiple connected passenger cars curving away into the canyon. Behind him, include exactly 1 military helicopter hovering low above the train with exactly 2 bright searchlight beams aimed down onto the cars. Add exactly 1 fiery explosion on the canyon wall/trackside in the middle distance, plus sparks bursting from the train wheels or rails in the lower left foreground. Environment: red-rock canyon cliffs on both sides, dusty air, motion blur from extreme speed, warm golden-hour sun low on the horizon, dramatic lens flare, volumetric haze, flying dust trails, cinematic blockbuster realism. Use a dynamic wide-angle perspective with the hero large in the lower center, helicopter and explosion behind him, intense depth, high contrast, sharp photoreal detail, epic action-movie color grading, no text, no logos, no watermark.
Comic / Storyboard - Monochrome Industrial Axe Walker — output

Comic / Storyboard - Monochrome Industrial Axe Walker

Dilshad HussainGPT Image 2
Create a high-contrast monochrome 35mm film photograph in a vertical 2:3 frame, medium full-body street-documentary style. Show {argument name="character description" default="a muscular adult man"} in left-facing profile walking from right to left through an unfinished industrial concrete structure. He wears exactly 4 visible clothing/accessory items: a fitted black crew-neck T-shirt, slim black pants, black leather ankle boots, and a metal wristwatch; add subtle round eyeglasses visible at the edge of the face. His face is intentionally anonymized with a soft gray rectangular blur covering the central face area. He holds exactly 1 large axe low in his right hand behind his body, with a pale wooden handle angled diagonally backward and a dark metal axe head pointing downward near his thigh. The setting contains rough concrete columns, exposed scaffolding pipes on the right, dusty ground with small debris, and a smoky blurred background with shallow depth of field. Use natural side lighting, gritty texture, film grain, strong blacks, realistic anatomy, candid walking pose, and no text or logos.
Comic / Storyboard - Cinematic Parking Garage Walk — output

Comic / Storyboard - Cinematic Parking Garage Walk

Dilshad HussainGPT Image 2
Create a hyper-cinematic vertical portrait in a dark underground parking garage: {argument name="character description" default="a mysterious adult man with short slick black hair"} walks directly toward the camera from the center of the frame, wearing {argument name="outfit" default="a fitted long black trench coat over an all-black tactical outfit with black combat boots"}. His right hand is raised toward the lens in a commanding gesture while his left arm hangs naturally, body full-length and centered. Use an ultra-low floor-level camera angle with strong one-point perspective, wet reflective concrete floor, gritty industrial ceiling, pipes, concrete pillars, drifting smoke, mist, and dust particles. Place exactly 2 blurred luxury sports cars flanking him symmetrically, one on the left and one on the right, both facing forward with bright white headlights and glossy dark bodywork. Add exactly 3 visible red warning light sources in the background, two small red light clusters behind the character and red glow reflected on the cars and floor. Add multiple cool white fluorescent tube lights overhead and along both sides, including one bright central ceiling tube above the character, creating blue-white highlights. Lighting should be moody teal, black, and red with strong rim light, volumetric haze, wet reflections, lens flare across the lower torso, shallow depth of field, high contrast, realistic skin and fabric texture, cinematic action-thriller atmosphere, 9:16 vertical composition, no text, no logos, no watermark.
Comic / Storyboard - Overhead Nightclub Staircase Portrait — output

Comic / Storyboard - Overhead Nightclub Staircase Portrait

Dilshad HussainGPT Image 2
Create a vertical cinematic nightclub portrait of {argument name="main subject" default="a confident young man"} standing still in the center of a luxurious spiral staircase, photographed from a dramatic overhead top-down surveillance-balcony angle. The subject wears {argument name="outfit" default="an all-black tailored suit over a black turtleneck"}, has neatly styled dark hair, one hand in his trouser pocket, and the other hand raised in a subtle finger-gun gesture toward the camera. Surround him with exactly 4 blurred clubgoers moving on the stairs: one at lower left in a light shirt, one dark-clad figure at lower center, one woman at right in a black dress, and one partially visible blurred figure at the bottom edge. The staircase is glossy black metal with curved railings, mirrored panels, and sharp geometric blue and violet LED light strips tracing the rails and walls. Add dense white-blue smoke or fog rising behind the subject, dark paneled walls with gold trim, reflections on polished surfaces, and a moody luxury club atmosphere. Use high-contrast cinematic lighting in {argument name="neon color palette" default="electric blue, violet, magenta, and deep black"}, shallow motion blur on the surrounding people, crisp focus on the central subject, wide-angle perspective, vertical 9:16 composition, realistic photography, no text, no watermark.
Product Marketing - Cinematic Dr Pepper Diner Ad — output

Product Marketing - Cinematic Dr Pepper Diner Poster

Hania AiGPT Image 2
Goal: Create a hyper-realistic cinematic luxury soda advertisement poster for {argument name="brand name" default="Dr Pepper"}, with a glamorous nighttime diner atmosphere, glossy red color grading, and dramatic commercial beverage photography.

Canvas: Vertical poster, 2:3 aspect ratio, high-resolution, dark black-and-crimson palette, rain-slick street reflections, glowing neon, shallow depth of field, intense rim lighting, premium magazine-ad finish.

Main subject: A fashionable young woman stands in the left and center foreground, posing confidently beside a shiny red classic car. Her face is intentionally covered by a plain flat medium-brown rectangular censor block with no facial details visible. She has long dark wavy hair around the rectangle, realistic glowing skin on the visible neck and hands, hoop earrings, and a confident fashion-ad stance. Outfit: {argument name="woman outfit" default="burgundy fitted blazer, black crop top, black fitted jeans, and black heeled boots"}. One arm rests near the red car while the other hand extends toward the viewer holding the soda can.

Product focus: A large red chilled soda can dominates the foreground in her outstretched hand, tilted slightly, covered with condensation droplets and glossy highlights. The can label reads "Dr Pepper" in large white script, includes "Est. 1885", a circular "23" flavor emblem, and small decorative branding details. The can should be the sharpest object and feel oversized by forced perspective.

Liquid action: From the open top of the can, a dramatic thick ribbon of dark red carbonated soda erupts upward and sweeps across the entire top half of the poster in an elegant swirling wave. The soda splash contains sparkling bubbles, droplets, and exactly 5 visible ice cubes: one large cube near the upper right curve, one cube just below it on the right, one cube near the center of the splash above the can, one cube slightly lower near the can opening, and one smaller cube to the left of the can in the splash stream.

Background: A nostalgic American diner street at night, glowing with red neon and warm window lights. Include a prominent diner building in the upper background with a bright neon sign reading "DINER" on the roof, silhouettes of patrons inside, two vintage cars parked in front, and urban buildings behind it. On the far left, show a blurred close-up red neon diner sign cropped so only part of the word is visible. On the right background, include a large glowing roadside sign with the brand logo and a circular "23" emblem. Wet pavement reflects red lights and headlights.

Text content: In the lower right, add a large distressed bold condensed white headline reading {argument name="headline text" default="BOLD FLAVOR"} on two lines. Beneath it, add a smaller uppercase tagline reading {argument name="tagline text" default="NOT ORDINARY. NEVER BORING."}, with "NEVER BORING." emphasized in red. Keep the text clean, legible, and integrated into the poster like a professional soda campaign.

Visual style: Hyper-realistic, cinematic, high contrast, glossy commercial retouching, luxury beverage photography, dramatic perspective, volumetric neon glow, crisp condensation, tiny soda bubbles, red-and-black monochromatic atmosphere, premium advertising poster aesthetic.

Constraints: Use exactly one woman, one main soda can, one red car in the foreground, one diner building, one roadside brand sign, and exactly 5 ice cubes. Do not add extra characters in the foreground, do not add additional headline copy, avoid cartoon or illustration style, and keep all visible text spelled clearly.
Product Marketing - Fresh Burst Sprite Ad Poster — output

Product Marketing - Fresh Burst Sprite Ad Poster

Hania AiGPT Image 2
Goal: Create a hyper-realistic cinematic vertical soda advertisement poster for {argument name="brand name" default="Sprite"}, full of freshness, motion, splashing water, citrus, and tropical energy.

Canvas: Vertical poster, 2:3 aspect ratio, high-resolution luxury commercial photography style, bright daylight, saturated greens and aquas, dramatic wide-angle perspective.

Layout: Place huge bold white distressed uppercase headline text in the upper left reading {argument name="headline text" default="FRESH BURST"}, stacked on two lines. Beneath it add a small all-caps tagline reading {argument name="tagline text" default="MORE THAN REFRESHING. IT'S A VIBE."}. Under the tagline place the white italic Sprite logo with a small yellow-green sparkle icon. The center-right foreground is dominated by one oversized plastic soda bottle angled diagonally toward the viewer, held in the woman's right hand, with a green cap and green label. The center-left foreground shows one young woman jumping or floating midair, wearing a lime-green fitted sporty crop top, white denim shorts, and white sneakers; her hair is tied in a messy bun with loose strands. Cover her face with one plain tan rectangular censor block matching the generated image. Use transparent eyeglasses partially visible around the blocked face.

Subject details: The woman has a cheerful energetic pose, knees bent as if leaping, torso leaning back slightly, one arm extended toward the camera holding the bottle. The bottle label prominently reads {argument name="bottle label text" default="Sprite"} with smaller visible text “LEMON-LIME”; show clear carbonated liquid, bubbles, condensation, and floating lime slices inside and around the bottle. Make the bottle extremely crisp and glossy, larger than life, occupying much of the right half of the poster.

Environment: A lush fantasy tropical waterfall canyon with turquoise water below, mossy cliffs, dense green foliage, sunbeams, mist, sparkling droplets, and a dreamy resort-like terrace in the lower right with small round tables and umbrellas. Surround the woman and bottle with a circular burst of splashing water, airborne leaves, bubbles, and lime wedges. Include bright backlighting from the upper right and a sparkling sky-blue background.

Counted visible elements to preserve: 1 woman, 1 oversized soda bottle, 1 face-covering rectangular block, 4 main text elements: “FRESH BURST”, “MORE THAN REFRESHING. IT'S A VIBE.”, the left Sprite logo, and the bottle Sprite logo. Include many small bubbles, droplets, leaves, and lime slices as atmospheric particles, but do not add extra brand slogans.

Visual style: Hyper-realistic commercial beverage photography blended with cinematic fantasy advertising, ultra-detailed condensation, high-speed splash photography, glossy product lighting, vibrant green-and-cyan palette, shallow depth of field, dynamic motion, polished premium ad finish.

Constraints: Keep all text legible and in English, avoid extra logos beyond the two Sprite marks, do not add watermark or borders, preserve the censored face block, and make the product the sharpest focal point.
Product Marketing - Cinematic Orange Soda Ad Poster — output

Product Marketing - Cinematic Orange Soda Ad Poster

Hania AiGPT Image 2
Create a vertical hyper-realistic cinematic soda advertisement poster for {argument name="soda brand" default="Fanta Orange"} in a luxury beverage commercial style. A cheerful stylish young woman sits casually in the foreground, wearing an oversized vivid orange hoodie, white shorts, and white sneakers; her dark messy hair is tied up in a loose bun, and her face is intentionally covered by a centered plain tan square censor block. She holds a large chilled orange soda bottle toward the camera with strong foreshortening, making the bottle the dominant foreground subject; the bottle is covered in condensation droplets and sparkling bubbles, with an orange label showing the Fanta-style logo, orange slice graphics, and small flavor text. The background is an extravagant fantasy orange-soda world at golden hour: huge arcing waves of glowing orange liquid splash across the sky, floating orange slices, citrus trees heavy with fruit, sparkling droplets, a small orange hot-air balloon with a soda logo, tiny carnival-like platforms, a Ferris wheel, and warm sunset clouds. Add bold white distressed uppercase advertising text on the right reading {argument name="headline text" default="TASTE THE FUN"}, with smaller white subtext beneath reading {argument name="tagline" default="MORE THAN A SODA. IT'S HAPPINESS."}. Use saturated orange and gold color grading, dramatic rim light, shallow depth of field, ultra-detailed realistic skin and fabric texture, glossy bottle plastic, crisp commercial photography, energetic splash motion, premium poster composition, no extra logos beyond the soda branding, no watermark.
Product Marketing - Cinematic 7 Up Cool Escape Poster — output

Product Marketing - Cinematic 7 Up Cool Escape Poster

Hania AiGPT Image 2
Create a hyper-realistic vertical cinematic soda advertisement poster for {argument name="brand name" default="7 Up"}. A cheerful young woman is lunging forward toward the camera in an energetic selfie-like perspective, holding an opened green soda can extremely close to the viewer in her left hand. She wears a mint green lightweight jacket over a white crop top, white high-waisted shorts, and white sneakers; her brown hair is tied in a messy high bun with loose strands. Her face is covered by a centered opaque rectangular blur block, while the rest of her body and outfit remain sharply detailed. The can is wet with condensation, has a silver pull tab, visible lemon-lime graphics, and the large white-and-red 7 Up logo. A huge curling wave of sparkling water, bubbles, ice-like mist, floating lime slices, green leaves, and splashes wraps around the can and the model, forming a dynamic spiral that leads into the background. Set the scene in a surreal refreshing alpine fantasy landscape with towering snow-covered mountains, waterfalls, turquoise rivers, mossy rocks, lush greenery, and bright tropical accents in the lower distance. Use cool teal, mint, white, and icy blue tones with high contrast sunlight from the upper right, glossy commercial lighting, crisp droplets, and premium beverage-ad photography realism. Add bold distressed white poster typography in the upper left: large stacked headline text “{argument name="headline text" default="COOL ESCAPE"}” and smaller all-caps tagline beneath it, “{argument name="tagline text" default="REFRESH YOUR WORLD."}”. Composition should be portrait 2:3, ultra-detailed, dramatic wide-angle lens, shallow depth of field on the foreground can and hand, motion frozen mid-splash, no extra products, no extra people, no watermark.
Product Marketing - Purple Fantasy Grape Soda Ad — output

Product Marketing - Purple Fantasy Grape Soda Ad

Hania AiGPT Image 2
Create a vertical hyper-realistic cinematic fantasy soda advertisement poster for {argument name="beverage name" default="PURPLE SPLASH"}, dominated by vivid purple and lavender tones. The scene shows a joyful young woman kneeling in the foreground, wearing exactly 3 visible outfit pieces: a lavender oversized hoodie, fitted blue denim shorts, and white sneakers. Her body leans forward as she extends one hand toward the viewer, holding one large cold aluminum can labeled {argument name="can label" default="GRAPE SODA"}; the can is covered in condensation droplets and reflects a miniature glowing purple fantasy village inside its surface. Her face is mostly hidden by one opaque square censor block in muted mauve-pink, while dark wavy hair remains visible around it. Above the can, create a dramatic magical grape-soda eruption: one swirling translucent purple liquid splash, many bubbles, sparkling droplets, grape clusters, and exactly 3 clearly visible ice cubes flying through the air. The background is an enchanted nighttime grape kingdom with a full moon, starry sky, glowing clouds, purple vineyards, tiny illuminated houses, castle-like buildings, garden paths, and sparkling firefly-like lights. Include exactly 3 main text elements: a huge bold top headline reading {argument name="headline text" default="PURPLE SPLASH"}, a smaller spaced all-caps tagline underneath reading {argument name="tagline" default="MORE THAN SWEET. IT’S MAGIC"}, and the can text reading {argument name="can label" default="GRAPE SODA"}. Use premium beverage photography lighting, glossy commercial retouching, shallow depth of field, ultra-detailed skin and fabric texture, magical rim lighting, high contrast, sparkling highlights, and a polished fantasy advertising composition. Keep the poster vertical, centered on the woman’s extended hand and soda can, with the headline occupying the top third and no watermark or extra logos.
Comic / Storyboard - Fantasy Huntress and Ancient Beast — output

Comic / Storyboard - Fantasy Huntress and Ancient Beast

Ai BellaGPT Image 2, Nano Banana Pro
Ultra-realistic epic fantasy film still, 9:16 vertical composition, low-angle full-body shot of a lone {argument name="character" default="female barbarian"} with spear and layered leather straps, fur shoulder cape, wrapped boots and bracers, wind-tousled braided hair, gritty skin texture and subtle scars, walking beside an {argument name="creature" default="enormous ancient horned beast"} with weathered stone skin on the face, thick mane, glowing amber eyes, rune-carved forehead, cinematic {argument name="environment" default="foggy mountains"} and distant ruins behind them, rocky ground, volumetric mist, dramatic contrast lighting, muted gray-brown palette, high detail, sharp cinematic focus.
Social Media Post - Vogue Style Fashion Illustration Sketch — output

Social Media Post - Vogue Style Fashion Illustration Sketch

Melis✨GPT Image 2
Turn {argument name="subject" default="this photo"} into a {argument name="style" default="Vogue-style fashion illustration"}, preserving identity and facial features. Minimalist hand-drawn fashion sketch with elegant elongated proportions, bold black ink contours, loose confident linework, flat marker colors, abstract blush accents, graphic lips, stylized eyes, expressive ink hair, and simplified geometric clothing. Clean white background, scanned paper texture, modern luxury editorial fashion illustration.
Profile / Avatar - Emoji-Based Creative Image Generation — output

Profile / Avatar - Emoji-Based Creative Image Generation

AI PulseGPT Image 2
{argument name="emoji set one" default="👳🍀✨🧟"}
{argument name="emoji set two" default="🖤🦇💀🤖"}
{argument name="emoji set three" default="🚧⌚💊🌆"}
{argument name="emoji set four" default="🐲🦆🎯🌈"}
Profile / Avatar - Dreamy Garden Swing Portrait — output

Profile / Avatar - Dreamy Garden Swing Portrait

ShinningGPT Image 2
Use the uploaded portrait only as the identity reference. Strictly preserve the subject’s face shape, facial features, skin tone, hairstyle, hair length, and natural hair color. Do not copy the original portrait’s clothes, background, lighting, pose, or image quality.

Create a realistic dreamy outdoor portrait on a {argument name="object" default="rope swing"} in a {argument name="location" default="lush flower garden"}, vertical 3:4 composition. The subject is turning back slightly while sitting on the swing, hair gently flowing in the wind, wearing a {argument name="outfit" default="soft white flowing summer dress"}. Strong golden backlight from the sun, glowing rim light around the hair and shoulders, soft haze, subtle lens bloom, natural motion blur, cinematic romantic atmosphere. Real camera photography, natural skin texture, soft focus background, delicate flowers and deep green foliage, high-end editorial photo style.
Social Media Post - Dark Fantasy Portrait with Shadow Demon — output

Social Media Post - Dark Fantasy Portrait with Shadow Demon

HarborisGPT Image 2
Ultra-realistic dark fantasy cinematic portrait of a stylish {argument name="subject" default="young fair handsome boy"} standing alone on broken railway tracks at sunset twilight, surrounded by glowing purple magical smoke and neon violet energy cracks on the ground, giant shadow demon made of dark cosmic smoke with glowing purple eyes towering behind him in the sky, mystical glowing purple butterfies flying all around, dramatic anime-inspired supernatural atmosphere, boy wearing {argument name="clothing" default="maroon oversized sweatshirt"} with white text print {argument name="text" default="HarBoris"}, loose light blue baggy jeans, red cap, red-white sneakers, red tote bag in hand, holding a glowing purple katana sword pointing downward, calm fearless pose, looking at camera, urban abandoned railway crossing background, trees and electric poles around, cinematic volumetric lighting, ultra detailed, hyper realistic, powerful aura, fantasy cyberpunk vibe, highly detailed, masterpiece, sharp focus, epic composition.
Profile / Avatar - Korean-Style Expressive Character Illustration — output

Profile / Avatar - Korean-Style Expressive Character Illustration

HeisenLegacyGPT Image 2
Create an illustration of what you imagine “me” to be like: A {argument name="style" default="Korean-style"} illustration featuring a distinctive and unforgettable character design, with natural and expressive facial emotions, a {argument name="composition" default="half-body"} composition, and dynamic posing. The clothing should have intricate and refined details. Use a {argument name="art style" default="hand-drawn doodle style"} with ink splashes, loose and spontaneous lines, and a blend of soft pastel tones with ink shading. The goal is to achieve a comic sketch-like texture.
Social Media Post - Stylish Man in Grunge Collage Background — output

Social Media Post - Stylish Man in Grunge Collage Background

HeisenLegacyGPT Image 2
Use reference photo to create A stylish young man posing with one hand on his face and the other in his pocket, wearing a {argument name="outfit" default="loose black shirt and light beige pants"}. He has voluminous curly hair and is standing against an {argument name="background" default="artistic black-and-gray collage background"}. The collage includes elements like a {argument name="collage elements" default="police car, flying horse (Pegasus), perfume bottle, newspaper text, record label, and abstract paint textures"}.Aesthetic dark grunge mood with soft white outline glow around the person, realistic lighting
Profile / Avatar - Moody Night Window Reflection Selfie — output

Profile / Avatar - Moody Night Window Reflection Selfie

ShinningGPT Image 2
Use the uploaded portrait as the identity reference only. Preserve the subject’s real face shape, facial features, skin tone, and hairstyle. Do not copy the original portrait’s clothes, background, pose, lighting, or image quality.

Create a realistic vertical 9:16 night street reflection selfie. The subject is seen through a glass window reflection, holding a smartphone in front of part of the face, wearing a {argument name="outfit" default="simple dark outfit"}. Soft messy hair moved slightly by wind. Behind the glass, a bright {argument name="store name" default="24H MART"} or subway-station shop appears heavily blurred with horizontal motion streaks, creating a cinematic long-exposure feeling. Moody cool-toned lighting, soft haze, glass reflections, subtle smudges, realistic shadows, natural skin texture, authentic phone-shot atmosphere, slightly melancholic urban mood. Any visible signs should be simple English text such as "{argument name="sign text" default="ROSE"}", no Chinese characters.
Social Media Post - Classroom Window Melancholy Portrait — output

Social Media Post - Classroom Window Melancholy Portrait

ToroJushiAiGPT Image 2
Ultra-realistic portrait of a beautiful {argument name="subject" default="Korean woman"} standing at her desk, in the quiet act of packing up — one book held loosely against her chest with both arms wrapped around it, head turned to look out the classroom window one final time before leaving. She is not looking at anything specific. It is the look of someone memorizing a room. Camera positioned at medium distance, capturing her from mid-thigh up, centered between the window light and the classroom.
Wearing a {argument name="shirt" default="soft sage green linen button-down shirt"} tucked loosely into {argument name="trousers" default="high-waisted dark rust-brown wide-leg trousers"}, a simple brown woven belt. Tan suede ankle boots. A large worn leather tote bag half-packed on the desk beside her, a scarf tied to the handle. Thin layered gold chains at the collarbone. Long dark brown hair down, loose natural waves, a simple center part.
The golden afternoon light has taken on the deep amber quality of late-hour sun — now warm and almost melancholy, the longest shadows of the day cutting across the rows of empty desks. The window she looks toward is a rectangle of burning amber against the quieting classroom. Her sage linen shirt catches the warm light differently from every direction — cool shadow on one side, warm amber on the other.
35mm analog film grain, Kodak Portra 400 — sage green and rust brown in deeply harmonized Portra earth tones, the melancholy of late golden hour fully expressed in color. Hyper-realistic linen texture, leather bag wear detail. Shallow depth of field. Mood: the end of something unnamed, time passing without drama, the most quietly devastating composition in the series.
Profile / Avatar - Vintage Film Camera Portrait — output

Profile / Avatar - Vintage Film Camera Portrait

ToroJushiAiGPT Image 2
Ultra-realistic portrait of a beautiful {argument name="subject" default="Korean woman"} standing near the back of the classroom, a vintage 35mm film camera raised to her eye, looking through the viewfinder — photographing something outside the frame. One eye closed, one focused into the lens. Expression concentrated, quiet, self-contained. Camera (our camera) positioned to her side and slightly behind at a three-quarter angle, capturing her in profile.
Wearing a {argument name="outfit" default="camel-colored vintage corduroy blazer"} over a soft white crew-neck tee, tucked into high-waisted dark brown straight-leg trousers. Dark brown leather belt, matching brown leather loafers. Thin wire-frame round glasses. Long dark brown hair pulled into a {argument name="hairstyle" default="relaxed low bun with a tortoiseshell clip"}, a few strands loose at the temples. A well-worn canvas camera bag on the desk behind her.
The golden afternoon light is the subject and the setup simultaneously — she is photographing it. The light streams in from the classroom windows directly into her lens, creating a warm backlit flare around the camera body. Her camel blazer glows like aged amber in the golden hour. Dust motes float beautifully in the shaft of light she is aiming at.
35mm analog film grain, Kodak Portra 400 — camel corduroy and dark brown trousers rendered in deep, lush Portra earth tones. Hyper-realistic corduroy fabric texture, vintage camera metal and glass detail, lens flare. Shallow depth of field. Mood: the photographer inside the photograph, self-referential and quietly poetic, editorial with intellectual texture.
Product Marketing - Typography City Skyline Silhouette — output

Product Marketing - Typography City Skyline Silhouette

GadgetifyGPT Image 2
2x2 grid, 16:9, do this for 4 famous cities: Anchor: The word "{argument name="city names" default="TOKYO / PARIS / CAIRO"}" :: morphed into the exact skyline silhouette of that city ::4 Morphology: Black and white negative space typography art. Letters stack and stretch vertically to become skyscrapers, towers, and spires. The tallest letters become the tallest landmarks. Skyline is readable as both a word and a city profile simultaneously ::3 Material Physics: Crisp white ink on black cardstock. Hard edges. No curves softened. Each building-letter has distinct window details etched into it ::3 Illumination: Flat graphic, dusk silhouette quality, high contrast ::2 Render Stack: City tourism poster, travel print, architectural illustration ::1 Negative: [color, gradients, realistic windows, people, vehicles, clouds, ground reflection] :: -1
Profile / Avatar - Cinematic Night City Street Portrait — output

Profile / Avatar - Cinematic Night City Street Portrait

MaercihGPT Image 2
{
  "prompt": {
    "subject": "uploaded image",
    "appearance": {
      "hair": "{argument name="hair style" default="messy dark brunette hair tied in a loose updo with soft strands framing the face"}",
      "expression": "quiet, introspective, emotionally distant",
      "features": "natural skin texture, delicate facial structure, soft cinematic realism"
    },
    "wardrobe": {
      "top": "earth-toned textured wool coat or knit outerwear",
      "style": "minimal cinematic street fashion"
    },
    "pose": "side profile portrait in a moving crowd, looking slightly forward",
    "scene": {
      "location": "{argument name="location" default="crowded nighttime city street"}",
      "environment": "busy urban crowd with blurred moving people",
      "background": "heavy warm bokeh lights and out-of-focus silhouettes",
      "atmosphere": "moody, cinematic, emotionally isolated"
    },
    "composition": {
      "framing": "tight medium close-up",
      "angle": "eye-level side profile",
      "depth_of_field": "extremely shallow depth of field",
      "focus": "sharp focus on the subject's eyes and face",
      "foreground": "blurred silhouettes partially obscuring the frame edges",
      "aspect_ratio": "4:5"
    },
    "lighting": {
      "type": "practical street lighting",
      "style": "soft cinematic tungsten lighting",
      "color": "warm amber highlights with gentle shadow falloff"
    },
    "camera": {
      "style": "cinematic street photography",
      "lens": "85mm prime lens",
      "aperture": "f/1.4",
      "settings": {
        "iso": 800,
        "shutter_speed": "1/125"
      }
    },
    "color_grading": {
      "tones": "warm browns, muted blacks, amber highlights",
      "contrast": "medium contrast",
      "style": "cinematic film grading"
    },
    "quality": {
      "render_style": "ultra realistic cinematic photography",
      "details": "natural skin pores, realistic hair strands, organic lighting, authentic crowd motion blur"
    }
  }
}
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 ↗