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.

Explore prompts

9,970 prompts

Comic / Storyboard - Warrior Versus Stone Golem — output

Comic / Storyboard - Warrior Versus Stone Golem

カーブミラーGPT Image 2
Create a dramatic fantasy manga-style battle scene in a vertical 4:5 composition. A colossal ancient stone golem towers over a lone armored warrior in ruined temple grounds, viewed from a low cinematic angle that makes the golem feel gigantic. The golem is built from thousands of irregular gray-brown stone blocks and broken masonry, with patches of green moss and hanging vines between the rocks. It has glowing cyan-blue magical eyes and exactly 4 visible glowing blue rune cores: one in the center of its chest, one on its forehead, one on its right shoulder, and one on its left forearm. Its massive left hand reaches forward toward the viewer with oversized stone fingers, while the other arm is pulled back in a fist as if charging an attack. In the foreground, show one human warrior seen from behind, wearing dark medieval armor, boots, and a long tattered navy-black cape with gold trim, holding a sword in the right hand and a round shield on the left arm, braced in a defensive stance facing the giant. Surround them with swirling stylized dust clouds and speed-line smoke ribbons, rendered like inked manga effects: exactly 8 major white dust plumes curling around the scene, with black outlines and dynamic motion arcs. Add flying pebbles, cracked stone debris, broken columns, scattered ruins, and mossy rubble across the ground. The background is a vivid bright blue sky with scattered white clouds and distant ruined stone towers. Visual style: highly detailed anime fantasy illustration, sharp black ink outlines, comic-book crosshatching, dynamic perspective, saturated colors, strong sunlight, high contrast, epic action energy, no text, no watermark. Use {argument name="hero description" default="a lone dark-armored warrior with a tattered navy-black cape and sword"}, {argument name="monster description" default="a colossal moss-covered stone golem with glowing cyan rune cores"}, {argument name="setting" default="ancient ruined temple courtyard under a bright blue sky"}, {argument name="magic glow color" default="cyan blue"}, and {argument name="dust effect" default="large white manga-style swirling dust plumes"}.
Comic / Storyboard - Wide Minimal Character Sheet — output

Comic / Storyboard - Wide Minimal Character Sheet

なおき🥞GPT Image 2
Goal: Create a wide minimalist character sheet illustration for {argument name="character name" default="a young man with curly black hair"}, showing two separate character sheet designs side by side on one horizontal canvas.

Canvas: Very wide panoramic aspect ratio, about 4:1, pale light gray background, clean empty margins, no borders, no shadows, no watermark.

Layout: Split the canvas into two equal halves. Each half is its own character sheet with the same basic layout: a bold title at the top, three full-body turnaround poses arranged left to right, and one larger face/head close-up on the far right. Use exactly 2 character sheet panels total.

Left panel details: Use a soft flat cartoon style with warm skin tone and muted colors. Show exactly 4 character drawings: 1 front full-body view, 1 side full-body view, 1 back full-body view, and 1 large face close-up. The character has rounded curly black hair, a simple friendly face, a long neck in the close-up, a loose dark navy/black sweatshirt, cropped white pants, and black shoes. Place the full-body poses evenly spaced, with the face close-up larger than the body poses.

Right panel details: Use a thinner, more monochrome sketch-like flat style with dark navy outlines and minimal color accents. Show exactly 4 character drawings: 1 front full-body view, 1 side full-body view, 1 back full-body view, and 1 large face close-up. The same character wears a dark oversized sweater, white straight pants, and dark shoes. Add subtle purple accent lines on the sweater sleeves. The face close-up is simplified, with short dark hair, ears visible, and a black collar shape beneath the head.

Text content: Draw only the following text labels, each spelled exactly as written. In the left panel: CHARACTER SHEET, FRONT, SIDE, BACK, FACE. In the right panel: CHARACTER SHEET, FRONT, SIDE, BACK, FACE. Use bold uppercase hand-drawn sans-serif text for the titles and small uppercase labels under each view.

Visual style: Clean vector-like illustration, simple rounded forms, slightly imperfect hand-drawn lines, limited palette of black, dark navy, white, pale peach skin, and light gray background. Keep the composition airy and balanced.

Constraints: Do not add any other letters, symbols, logos, signatures, dates, URLs, watermarks, or decorative text. Keep exactly 8 character drawings total: 6 full-body turnaround views and 2 face close-ups.
Product Marketing - Luxury Grandmaster Chime Watch Ad — output

Product Marketing - Luxury Grandmaster Chime Watch Ad

TWneseGPT Image 2
Create a luxury editorial advertising poster for {argument name="brand name" default="PATEK PHILIPPE"} featuring the {argument name="watch model" default="Grandmaster Chime"}. Canvas: vertical 4:5 magazine ad, photorealistic, dark opulent still-life composition. Layout: place a large ornate rose-gold grand complication wristwatch on the right side, angled slightly toward the viewer, with a brown crocodile leather strap and an extremely detailed engraved case, bezel, lugs, pushers, crowns, guilloché dial, Roman numerals, multiple subdials, moonphase, date ring, and tiny brand markings on the dial. Put the typography on the left side in elegant gold: a Calatrava-style cross emblem at the top, then the large uppercase brand name, then spaced small caps “GENEVE”, a thin ornamental divider, the model name on two lines in refined serif type, and the tagline {argument name="tagline" default="CRAFTED FOR THE FEW."} in small caps with a short horizontal rule below. Scene details: set everything on glossy black marble with white veining and warm reflections. Include exactly 5 main visible objects: 1) the wristwatch as the hero object, 2) a pair of black-and-gold cufflinks with gold cross emblems at lower left, 3) a black-and-gold fountain pen entering from the lower right with an engraved gold nib, 4) a dark whiskey tumbler blurred in the upper left background, and 5) a brown leather case or box blurred in the upper right background with a small gold emblem. Lighting and style: cinematic low-key lighting, warm amber highlights, deep shadows, shallow depth of field, premium Swiss watch advertising aesthetic, polished metal reflections, macro product photography, ultra-detailed engraving, rich brown and gold palette, sophisticated masculine mood. Constraints: keep all text crisp and readable, preserve the exact hierarchy and left/right composition, avoid extra logos, extra watches, extra jewelry, hands, people, watermarks, or modern UI elements.
Social Media Post - Anime Space Bass Guitar Cover Art — output

Social Media Post - Anime Space Bass Guitar Cover Art

ふぁーねりあん / FahnerianGPT Image 2
Create a wide 16:9 anime-style music cover artwork set in outer space above the curved blue-and-white Earth. The main subject is 1 floating teenage girl with a moody expression, pale skin, short messy {argument name="hair color" default="black hair with pink highlights"}, red eyes, and a pink X hair clip, wearing an oversized {argument name="shirt color" default="black"} T-shirt printed with a simple white cat face, black shorts or a short skirt, and bare legs. She is drifting diagonally across the frame while holding 1 mint-green electric bass guitar with a wooden neck and visible strings, angled from lower center to upper right. Add 1 fluffy white cat floating beside her on the right, eyes closed, tail raised, appearing weightless and serene. Include exactly 4 music-related objects: 1 bass guitar in her hands, 2 small square red-and-black portable speakers with circular red drivers floating at left and upper right, and 1 black computer keyboard with neon pink outlines floating at the bottom right. Use long looping pink audio cables that arc around the composition like orbital paths. The background is a starry black cosmos with a vivid colorful nebula cloud in hot pink, orange, purple, and blue stretching across the upper center, bright starbursts, tiny sparkles, small floating crystal fragments, and thin orbital lines. Lighting should be dramatic and cinematic with magenta rim light, cool blue Earth glow, high contrast shadows, clean cel-shaded anime rendering, detailed line art, glossy highlights on the guitar, and a dreamy melancholic J-pop album-cover atmosphere. No title text, no logo, no watermark.
YouTube Thumbnail - Anime Skywork MV Thumbnail — output

YouTube Thumbnail - Anime Skywork MV Thumbnail

萌奈のおすそわけ♥️AI動画クリエイター🩷AI & Web3 KOL🧡💛💚GPT Image 2
Goal: Create a bright anime YouTube thumbnail / promotional banner for a music-video-making tutorial, with a cute idol mascot on the right and bold Japanese headline typography on the left.

Canvas: Wide 16:9 horizontal banner, approximately 1200×675 px. High-energy composition with no empty space, suitable for social media and YouTube promotion.

Layout: Left half is dominated by large stacked headline text with thick outlines and drop shadows. Right half shows exactly 1 anime idol girl in close-up from chest up, facing the viewer, mouth open in an excited singing expression. Bottom-left contains exactly 1 white rounded rectangle brand badge with the Skywork logo icon and the word “Skywork.”

Text content: Use exactly 3 main Japanese headline blocks: 1) top small headline {argument name="top headline text" default="音楽情報入れたら"}, 2) central oversized headline {argument name="main headline text" default="MVが完成した!"}, 3) bottom subtitle {argument name="subtitle text" default="Skywork Video 1つで完結!"}. Make the central “MV” especially huge, bright red with white and dark navy outlines; make the rest of the central headline yellow-gold with red and dark outlines. Keep the typography bold, rounded, pop-style, slightly tilted, with strong white stroke and dark red/navy shadow for readability.

Subject details: The character is {argument name="character name" default="Mona"}, a cheerful kawaii anime idol girl with long brown twin tails, amber-orange sparkling eyes, flushed cheeks, and glossy highlights. She wears a white and lavender frilly idol outfit with a high collar, bare shoulders, purple trim, gold star ornaments, and large lavender bows in her hair. She holds exactly 1 magical idol wand near the right edge: gold handle, circular gold frame, purple faceted star gem in the center, small angel wings, ribbons, crescent moon, and star charms. Visible accessories should include exactly 2 large lavender hair bows, 1 gold crescent moon hair ornament, multiple small gold star clips, 1 star choker ornament, and the single wand.

Background and effects: Use a vivid pink, purple, orange, and gold concert-light background with diagonal speed lines, bokeh, confetti, sparkles, and glow effects. Add exactly 1 white music note icon floating above the character’s head. Scatter many small star-shaped sparkles around the text and character. Overall mood is celebratory, cute, magical idol, and energetic.

Visual style: Polished modern Japanese anime illustration, glossy eyes, soft cel shading, clean line art, saturated colors, strong rim lighting, high contrast, crisp thumbnail readability, commercial PR banner style.

Constraints: Keep all important text fully legible and inside the safe area. Do not add extra characters, extra logos, watermarks, QR codes, or additional text beyond the specified headline and Skywork badge.
Product Marketing - Japanese Lunch Menu Poster — output

Product Marketing - Japanese Lunch Menu Poster

不自然対数GPT Image 2
Goal: Create a vertical Japanese restaurant lunch menu poster for a casual diner, warm and slightly retro, with appetizing food photography and large brush-style Japanese typography.

Canvas: Portrait poster, approximately 4:5 ratio, parchment beige background with subtle paper texture, thin gold border, red and gold decorative accents, leaf motifs in the corners, and a festive Japanese pattern strip along the far left edge.

Header: At the top, place a large calligraphic headline reading {argument name="headline text" default="ランチメニュー"}; make the first word black and the second word deep red. Underline it with a rough gold brush stroke. In the upper-right corner, add a red circular seal containing white Japanese text: 「心を込めた おいしい ひととき」, with small pale pink flower decorations nearby.

Main layout: Create exactly five horizontal menu rows separated by thin beige lines. Each row has a circular icon on the left, a bold Japanese dish name in black calligraphy in the middle-left, and a realistic food photo panel filling the right half. Use warm restaurant tabletop lighting and wooden table surfaces in the photos.

Menu rows, exactly 5 items:
1. Red circle icon with a plum blossom, label 「ハンバーグ定食」, photo of a glossy demi-glace hamburger steak set with rice bowl, miso soup, broccoli, carrot, and garnish.
2. Gold circle icon with gyoza dumplings, label 「ジャンボ餃子定食」, photo of large pan-fried gyoza arranged in a row with rice, dipping sauce, soup, and pickles.
3. Green circle icon with a chicken silhouette, label {argument name="karaage item text" default="唐揚げ定食"}, photo of fried chicken karaage set with shredded cabbage, lemon wedge, rice, soup, and cucumber pickles; make the fried chicken look slightly uncanny and overly bulbous as if AI-generated.
4. Purple circle icon with a pig silhouette, label 「生姜焼き定食」, photo of pork ginger stir-fry with onions, shredded cabbage, tomato, rice, miso soup, and pickles.
5. Blue circle icon with a rice bowl, label 「中華丼」, photo of Chinese-style rice bowl topped with shrimp, quail eggs, vegetables, black fungus, thick glossy sauce, plus a small soup bowl.

Notice strip: Beneath the five rows, add a centered bordered beige notice box reading 「※定食ランチにはご飯・汁物付き」, with small gold leaf ornaments on both sides.

Bottom information area: Create a cream-colored panel with two large information blocks. On the left, a brown clock icon and label 「営業時間」 above the time {argument name="business hours" default="9:00〜16:30"}. On the right, a red fork-and-spoon icon and label 「ランチタイム」 above the time {argument name="lunch time" default="12:00〜14:00"}. Separate the two blocks with a thin vertical divider.

Footer: Add a deep red footer band with subtle gold brush decorations and small sparkle accents. Center white Japanese closing text: {argument name="footer message" default="ゆったりくつろげる空間で、皆さまのご来店を心よりお待ちしております。"}

Visual style: Polished Japanese print flyer, izakaya/family restaurant aesthetic, warm reds, browns, cream, and gold, high-resolution food photography mixed with decorative vector elements, clean grid alignment. Keep all Japanese text legible and avoid adding extra menu items or extra text.
Comic / Storyboard - Sleeping Girl in Pitch-Black Room — output

Comic / Storyboard - Sleeping Girl in Pitch-Black Room

Harf_DoneGPT Image 2
Create a vertical anime-style illustration of a young woman sleeping peacefully on a bed in an almost pitch-black room. The scene is viewed from a slightly elevated diagonal angle, showing her full body curled on her side across rumpled white sheets. She has {argument name="hair color" default="dark navy-black"} medium-long hair tied loosely in a low ponytail with loose bangs framing her face, closed eyes, soft eyelashes, and a relaxed sleepy expression with slightly parted lips. She wears {argument name="pajama color" default="pale sky blue"} long-sleeve button-up pajamas with white piping, a small chest pocket, and matching loose pants; her bare feet are visible near the bottom of the frame. A thick white blanket is draped over her waist and legs, catching cool blue shadows. Behind her are exactly 3 visible pink pillows: one large pillow behind her head, one large pillow to the right, and one smaller partially visible pillow farther right. The bed has white wrinkled sheets with pale blue-gray shadows and a few pink areas at the edges. The background beyond the bed is a flat deep black wall or darkness, emphasizing the bed as the only illuminated area. Use dramatic low-key lighting with soft highlights on the bedding and face, strong black negative space, subtle blue-gray shading, and a hand-drawn modern anime illustration style with clean ink lines, painterly crosshatching, muted pastel colors, realistic folds, and a quiet intimate nighttime mood. Add a small handwritten artist signature reading {argument name="signature text" default="HALF_DONE"} in the lower right corner. No extra furniture, no windows, no lamps, no text other than the signature.
Infographic / Edu Visual - AI Image Model Stadium Leaderboard — output

Infographic / Edu Visual - AI Image Model Stadium Leaderboard

Creative DiffusionGPT Image 2
Goal: Create a vibrant 16:9 Pixar-like 3D illustration of a packed robot sports stadium showing a giant electronic scoreboard ranking AI image models, as if it is a summer league leaderboard.

Canvas: Wide landscape image, bright sunny afternoon, blue sky with a strong sun flare in the upper left, colorful flags on the stadium rim, football field at the bottom with a yellow goalpost centered under the scoreboard. The scene is crowded with hundreds of small cute toy-like robots in the stands and foreground, all cheering.

Main subject: A huge black stadium scoreboard dominates the center, supported by metal trusses. It has LED dot-matrix styling, glowing cyan and amber text, thin grid lines, rounded table border, and small yellow star icons on both sides of the main header. At the very top of the scoreboard, show two small amber digital panels reading {argument name="time" default="15:08"} and {argument name="temperature" default="32°C"} with a sun icon.

Scoreboard text: The main headline must read {argument name="headline text" default="CLASSEMENT DES MODÈLES D’IMAGE IA"}. Under it, create a table with exactly 5 columns labeled RANG, MODÈLE, META-SCORE, PRIX / 1K, EVOLUTION. Include exactly 10 ranked rows, each with a small robot avatar icon next to the model name:
1. GPT-Image 2 — meta-score 1498 — price $211 — evolution =
2. Mai Image 2.6 — meta-score 1427 — price ~$50 — evolution NEW in a small blue badge
3. Reve 2.1 — meta-score 1381 — price $200 — evolution red down arrow 1
4. Grok Imagine Image 2.0 — meta-score 1369 — price $80 — evolution NEW in a small blue badge
5. Reve 2.0 — meta-score 1341 — price $24 — evolution red down arrow 2
6. Muse Image — meta-score 1315 — price (Pas d’API) — evolution red down arrow 2
7. Nano Banana 2 — meta-score 1314 — price $67 — evolution red down arrow 2
8. GPT-Image 1.5 — meta-score 1300 — price $133 — evolution red down arrow 2
9. Seedream 5.0 Pro — meta-score 1287 — price $90 — evolution red down arrow 2
10. Qwen Image 3.0 Pro — meta-score 1265 — price $75 — evolution NEW in a small blue badge

Surroundings: Add a left stadium sign reading {argument name="left banner text" default="AOÛT - SOLEIL & PROMPTS"} with a small sun drawing. Add a right sign reading CRÉER / INNOVER / IMAGINER. Add cheering props including a foam finger reading GO IA! and a purple flag reading IA TEAM. Foreground robots should be larger and out of focus at the edges, including visible jersey numbers 01 and 32, with one small referee robot standing on the field near the scoreboard.

Visual style: Ultra-detailed colorful 3D animation style, glossy plastic robots, playful futuristic sports atmosphere, shallow depth of field, crisp readable scoreboard typography, warm sunlight, high saturation, cinematic composition.

Constraints: Keep all scoreboard numbers and labels legible and exactly as listed, use French text where specified, no logos, no watermark, no extra table rows.
Product Marketing - Two Women in Pastel Pajama Bedroom — output

Product Marketing - Two Women in Pastel Pajama Bedroom

千田結穂(せんだゆいほ)GPT Image 2
Create a realistic, softly lit lifestyle photograph of two young East Asian women sitting close together on a cozy bed in a pastel bedroom, facing the camera with gentle natural smiles. The woman on the left is {argument name="left character" default="a young East Asian woman with long straight black hair parted slightly off-center"}, wearing {argument name="left outfit" default="cream pajamas with a tiny pink floral print"} and hugging a square beige textured pillow across her lap. The woman on the right is {argument name="right character" default="a young East Asian woman with shoulder-length dark brown hair and soft wispy bangs"}, wearing {argument name="right outfit" default="pale blue vertical striped button-up pajamas with white piping"} and holding a fluffy ivory heart-shaped pillow. Place them shoulder-to-shoulder in the center of the frame, waist-up to knees visible, with relaxed hands and distinct faces, hairstyles, and outfits that do not blend together. The bedroom should feel warm and feminine: white metal bedframe, cream and blush bedding, a large pale pink pillow on the right, a furry pink cushion on the left edge, a bedside table with a pleated warm lamp, a small framed picture, and a vase of pale pink roses. Add softly blurred pastel wall prints in the background, sheer white curtains glowing with daylight from the left, warm beige and blush color grading, shallow depth of field, clean natural skin texture, high-resolution photorealism, no text, no watermark, no extra people.
Profile / Avatar - Sleeping Woman in Nightcap — output

Profile / Avatar - Sleeping Woman in Nightcap

Harf_DoneGPT Image 2
Create a photorealistic cinematic close-up of {argument name="character name" default="a young woman"} sleeping peacefully in bed at night, shown from a slightly elevated bedside angle in vertical portrait orientation. She has {argument name="hair color" default="dark brown hair"} with soft bangs and loose strands framing her face, pale delicate skin, closed eyes, relaxed lips, and a serene expression. She wears a white vintage nightgown with lace trim and a white ruffled nightcap with a scalloped frill edge, the cap clearly visible and naturally shaped around her head. Her head rests sideways on a white pillow, one hand gently curled near her face, while a pale blue-gray blanket covers her body and fills the foreground with soft folds. The room is dim and quiet, with a dark blurred bedroom background, faint furniture silhouettes, and cool moonlight entering from the right, creating blue nighttime shadows and subtle highlights on the fabric and skin. Use shallow depth of field, realistic skin texture, soft cinematic lighting, gentle contrast, intimate calm mood, high detail, no text, no watermark, no extra people.
Comic / Storyboard - Cycling Break Forest Scene — output

Comic / Storyboard - Cycling Break Forest Scene

フレアの日常GPT Image 2
Using REFERENCE_0 as the character base, create a cinematic anime story-board scene of {argument name="character name" default="Flare Scarlet"} resting during a cycling trip. Keep her recognizable face, short red hair, red-and-white cycling jersey, black shorts, gloves, shoes, and red helmet from the reference, but change the setting and pose: place her seated casually on a mossy rock in a dense sunlit forest, leaning back with one hand for support, one knee raised, holding a clear water bottle as if taking a break.

Add one new companion cyclist beside her: a tall adult woman with very long black hair, wearing a blue-and-white helmet, white zip cycling jersey, black cycling shorts, black fingerless gloves, white socks, and black cycling shoes. Pose the companion seated relaxed on a separate rock, also holding a clear bottle.

Environment and props: create a realistic wooded mountain trail rest area with tall tree trunks, dense green foliage, dappled sunlight, mossy boulders, and a small weathered Japanese trail sign in the background. Include exactly 2 bicycles: one white touring/road bike leaning against a tree on the left with bags attached, and one dark road bike on the right behind Flare with a purple bottle on the frame.

Composition: wide horizontal 16:9 frame, both cyclists visible full-body or near full-body, companion on the left and Flare on the right, bicycles framing the scene. Use detailed anime rendering with natural cinematic lighting, realistic depth, crisp character detail, and a calm post-ride atmosphere. Avoid character-sheet panels, labels, stat boxes, or infographic layout.
Infographic / Edu Visual - Steamed Rice Low High Comparison — output

Infographic / Edu Visual - Steamed Rice Low High Comparison

AiRT🎥生成AI動画を創る人GPT Image 2
Goal: Create a square side-by-side quality comparison image of realistic Japanese steamed white rice, divided into two horizontal panels: top labeled {argument name="top label" default="LOW"} and bottom labeled {argument name="bottom label" default="HIGH"}. The subject is a freshly cooked mound of glossy white rice in a ceramic bowl on a warm wooden dining table, with steam rising visibly from the rice in both panels.

Canvas: 1:1 square image, photorealistic food photography, warm indoor breakfast lighting, shallow depth of field, natural brown and cream tones. Use a clean horizontal split exactly halfway across the image, creating 2 stacked panels.

Top panel: Close-up of a beige ribbed ceramic bowl filled high with steaming white rice. The rice should look slightly too uniform and overly shiny, with many similar rounded grains clustered together, subtly artificial but still appetizing. Background is softly blurred and includes exactly 4 supporting food elements: 1 wooden miso soup bowl at back left, 1 small plate with yellow tamagoyaki at lower left, 1 plate with grilled salmon and a white garnish at back right, and 1 small blue patterned dish with pickles at lower right. Add the label {argument name="top label" default="LOW"} in the lower right of this panel, bold red block letters with white outline and dark drop shadow.

Bottom panel: Similar Japanese breakfast scene with another ceramic rice bowl centered slightly lower, filled with steaming white rice. Make the rice grains more irregular in size, shape, angle, and spacing, with a softer, less plastic shine and a more natural cooked texture. Background is softly blurred and includes exactly 5 supporting food elements: 1 grilled salmon plate at back left, 1 small dish of yellow pickles and cucumber slices at lower left, 1 small pale dish of greens at back center, 1 wooden miso soup bowl with green onions at right, and 1 partial tabletop area in warm wood. Add the label {argument name="bottom label" default="HIGH"} in the lower right, bold blue block letters with white outline and dark drop shadow.

Logo/text overlay: Place a black circular logo overlapping the left side of the horizontal split. Inside it, large white text reads {argument name="logo main text" default="AiRT"}, with a small red ladybug-like icon near the letters. Below it add two lines of small white text: {argument name="logo subtitle" default="Ai ART CREATOR"} and {argument name="logo tagline" default="For videos/images"}.

Constraints: Keep the image photographic rather than illustrated. Emphasize the comparison between artificial-looking rice texture in the top panel and more natural irregular rice texture in the bottom panel. Use realistic steam wisps, soft bokeh backgrounds, no extra labels, no additional panels, and no people.
Comic / Storyboard - Gothic Villain Character Sheet — output

Comic / Storyboard - Gothic Villain Character Sheet

Shira WigglesGPT Image 2
Goal: Create a stylized animated-film character design sheet for {argument name="character name" default="a severe aristocratic villain"}, showing consistent costume and facial design from multiple angles.

Canvas: Wide horizontal character reference sheet, approximately 16:9, with an off-white studio background and clean white margins. Split the page into two main panels: a large left panel for full-body turnarounds and a narrower right panel for a dramatic portrait close-up.

Layout: Include exactly 4 character depictions total: 1 front full-body view on the left, 1 side profile full-body view in the center, 1 rear full-body view on the right side of the large panel, and 1 large close-up head-and-shoulders portrait in the separate right panel. Add a thin vertical column of small red decorative lettering along the far left margin, reading like abstract production-sheet text.

Subject details: The character is extremely tall, gaunt, and angular, with a long pale greenish-gray face, sharp hooked nose, heavy red under-eye shadows, drooping eyelids, yellow-green eyes, thin downturned lips, an elongated chin, and exaggerated large red ears. He has messy black hair swept over one eye, with loose stringy strands falling across the face. His expression is stern, tired, suspicious, and aristocratic.

Costume: Dress him in a dramatic {argument name="coat color and pattern" default="deep crimson and black plaid"} long tailcoat with broad squared shoulders, high collar, oversized lapels, black buttons, tapered waist, and long pointed tails. Under it he wears a dark red waistcoat, white shirt collar, and a large burgundy cravat. He has narrow {argument name="trouser color" default="muted olive-green striped"} trousers cropped above polished black formal shoes. The rear view should emphasize the long structured back of the coat and the plaid grid wrapping around the shoulders.

Visual style: High-end 3D animation concept art with painterly digital rendering, exaggerated caricature proportions, sharp graphic shapes, elegant gothic mood, subtle gradients, clean silhouette design, and polished studio lighting. Use saturated reds and blacks against pale skin and muted green clothing accents. The portrait close-up should be more detailed, with visible brush strokes in the hair, skin planes, eye bags, ears, and collar.

Constraints: Keep the character design identical across all 4 depictions. Do not add props, scenery, logos, or extra characters. Keep the background minimal and uncluttered, like a professional animation production turnaround sheet.
Comic / Storyboard - RAKUNE Summer Pajama Storyboard — output

Comic / Storyboard - RAKUNE Summer Pajama Storyboard

タナベ | AI動画 × マーケティングGPT Image 2
Goal: Create a polished advertising storyboard sheet for a 15-second summer pajama commercial for {argument name="brand name" default="RAKUNE"}, themed around fatigue recovery, cool comfort, and energetic mornings.

Canvas: Wide 16:9 presentation board, white background with thin gray grid lines, black header bar at the top and black footer bar at the bottom. Use a clean Japanese TV-commercial planning style, combining dynamic generated key visuals with production notes beneath each frame.

Header: Centered white Japanese headline on the black bar: “{argument name="product headline" default="RAKUNE 疲労回復パジャマ(夏向け) CM ストーリーボード(15秒・10カット)"}”.

Layout: Arrange exactly 10 storyboard panels in a 2-row by 5-column grid. Each panel has a bold panel number and timestamp at the top, a colorful image frame, then three short production-note lines below labeled in Japanese: “画:”, “テロップ:”, and “NA:”. Keep typography crisp and legible, with energetic blue, cyan, white, and yellow accents.

Panels: 
1. Label “1 (0:00-0:01.5)”. Dynamic morning scene of a smiling young Japanese woman in pale blue summer pajamas running or stepping forward energetically against blue-and-yellow speed lines. Large Japanese text in the image: “朝から 全力で いきたい!”. Notes describe the protagonist bursting from bed, energetic impact, and narration about wanting to go full power from morning.
2. Label “2 (0:01.5-0:03)”. Peaceful close-up of the same woman sleeping comfortably on a pillow, surrounded by a glowing blue circular swirl. Large Japanese text: “そのカギは、夜の回復力。” Notes mention entering sleep, recovery power during the night, and a soft expanding blue light.
3. Label “3 (0:03-0:04.5)”. Futuristic blue pajama top and bottom floating in space, outlined with luminous fiber patterns and motion rays. Large Japanese text: “特殊繊維が 血行を促進”. Notes mention the fiber image enlarging and wrapping the body, with an explanation of special fibers promoting circulation.
4. Label “4 (0:04.5-0:06)”. Abstract blue burst background with translucent fatigue particles or bubbles breaking apart and disappearing. Large Japanese text: “疲労物質を 軽減!”. Notes mention fatigue-substance particles scattering and being reduced.
5. Label “5 (0:06-0:07.5)”. The woman stands in pale blue short-sleeve summer pajamas, smiling in a bright airy room, with a curved arrow indicating airflow and breathability. Large Japanese text: “夏でも さらっと 快適!”. Notes mention coolness, moisture absorption, quick drying, and summer comfort.
6. Label “6 (0:07.5-0:09)”. Energetic wake-up pose with the woman smiling, hair moving, bright blue and yellow background rays. Large Japanese text: “ぐっすり眠って スッキリ 回復!”. Notes mention waking refreshed after deep sleep and energy charging.
7. Label “7 (0:09-0:10.5)”. Woman stretching both arms upward joyfully under a bright blue sky with yellow accents and confetti-like energy shapes. Large Japanese text: “さあ、今日も 全力で!”. Notes mention moving forward with a strong smile and a powerful day beginning.
8. Label “8 (0:10.5-0:12)”. Clean product logo card, white background with blue diagonal accents. Center text: “疲労回復パジャマ” above large “RAKUNE”, with a small blue capsule label “夏向け”. Include exactly 4 circular feature icons beneath, labeled “血行促進”, “疲労軽減”, “快適設計”, and “吸汗速乾”. Notes mention product logo and functional icons appearing rhythmically.
9. Label “9 (0:12-0:13.5)”. Collage-style panel split into multiple diagonal lifestyle images of the same woman active in daily life: working at a desk, running or exercising, laughing with a friend, and enjoying hobbies. Large Japanese text: “毎日を、もっとアクティブに.” Notes mention work, exercise, hobbies, and time with friends shown actively.
10. Label “10 (0:13.5-0:15)”. Final clean brand close-up: the woman faces camera and smiles softly on a bright white-blue background. Large Japanese text: “明日の自分に、エナジーを。” plus logo “RAKUNE” and small “夏向け” label. Notes mention ending strongly with the brand logo and message.

Visual style: High-energy Japanese commercial storyboard, realistic attractive young Japanese female model, bright summer lighting, cool blue pajama fabric, glossy advertising typography, comic-like motion streaks, sharp diagonal compositions, clean production-document layout. Use blue and cyan as primary colors, yellow for emphasis, and white for freshness.

Footer: On the black footer bar, centered white text: “TOTAL 15秒 | テンポの速いビートで、エネルギッシュに駆け抜ける構成”.

Constraints: Use exactly 10 numbered cuts, exactly 4 feature icons in panel 8, maintain the 2-by-5 grid, avoid watermarks, avoid extra panels, keep all Japanese visible text readable, and make the storyboard feel ready for a client presentation.
Product Marketing - Deep Blue Desire Fashion Poster — output

Product Marketing - Deep Blue Desire Fashion Poster

sereinGPT Image 2
Create a square editorial fashion magazine poster with a deep cobalt blue grunge paper background, subtle scratches, dust, fold marks, and worn texture. The main subject is a glamorous young East Asian woman with {argument name="hair color" default="long black wavy hair"}, fair skin, soft smoky makeup, glossy lips, and a red rose tucked behind her left ear. She wears small gold earrings, a delicate necklace, and a romantic white lace tie-front blouse with puff sleeves and a deep neckline. Place her as a large half-body portrait centered slightly right, looking confidently at the camera with a dreamy, cool expression. Overlay enormous tall condensed white typography across the whole poster spelling {argument name="main title" default="DESIRE"}, with the letters stretched vertically and partially transparent so they pass in front of and behind her face and body like a layered fashion layout. Add exactly 4 smaller text elements: a white rectangular label near the upper left reading {argument name="top label" default="DEEP"}, a slim three-line quote at left reading {argument name="side quote" default="A MOMENT / FROZEN, / A DESIRE / ETERNAL"}, a white rectangular label at mid-right reading {argument name="right label" default="BLUE"}, and a small spaced-out label near the bottom center reading "DESIRE". Add exactly 2 white star icons: one near the top above the tall typography and one near the lower right corner. Include exactly 1 secondary cutout figure of the same woman at the lower left, seated casually in the same outfit with blue jeans, long ribbon ties trailing, red rose in her hair, looking downward; scale it much smaller than the main portrait. Use high-fashion photorealism blended with graphic design, crisp studio lighting, cool blue color grading, layered transparent text, clean magazine composition, and no extra logos or watermarks.
Product Marketing - Blue Rose Beauty Poster — output

Product Marketing - Blue Rose Beauty Poster

sereinGPT Image 2
Create a vertical high-fashion cinematic beauty poster featuring {argument name="model description" default="a young East Asian woman with porcelain skin, long wavy black hair, soft glossy lips, and expressive brown eyes"}. The composition is a close-up portrait cropped so only the left half of her face and upper torso dominate the left two-thirds of the canvas, with the right side reserved for typography on a deep electric-blue studio background. She wears a white satin off-shoulder blouse with pale pink lace trim, a delicate silver pendant necklace, and one small gold rounded earring. Place exactly 1 large deep red rose tucked into her hair above her right ear. Add exactly 1 horizontal rectangular inset on the right-center showing a close crop of both eyes, aligned over the blue background. Use elegant editorial lighting, luminous skin retouching, cool blue color grading, subtle film grain, and faint horizontal light streak/glitch blur effects crossing the portrait and typography. Text layout: at the upper right, stack exactly 4 tall condensed uppercase English words, each spaced out vertically: {argument name="top English words" default="LONGING / PURE / DESIRE / EYES"}. Pair the first three with small Chinese captions and place one Chinese phrase below the eye inset: {argument name="Chinese captions" default="渴望 / 纯粹 / 心动 / 会说话的眼睛"}. At the lower right, set the main quote in very tall thin white serif lettering with slight blur and glow: {argument name="main quote" default="Could you still live in my eyes"}. Add exactly 1 tiny uppercase footer near the bottom right reading {argument name="footer text" default="BE SEEN. BE REMEMBERED. ONLY IN MY EYES."}. Keep the mood romantic, mysterious, luxurious, and melancholic, with the model looking directly at the viewer; no extra people, no logos, no watermark.
Product Marketing - Golden Hour Editorial Portrait — output

Product Marketing - Golden Hour Editorial Portrait

AvelyrahGPT Image 2
A hyper-realistic, high-fashion editorial portrait of a young {argument name="ethnicity" default="Asian"} woman in her early 20s with soft, natural brown hair styled in a casual low messy bun with loose strands framing her face. She is holding a smartphone in her hand and a small luxury designer handbag over her shoulder, smiling gently at the camera. She is wearing a {argument name="clothing" default="fitted light-blue ribbed tank top"} and a delicate minimalist silver pendant necklace. Set on a sunlit {argument name="location" default="European city street"} with classic stone architecture in the background. Cinematic golden hour lighting, soft subsurface scattering on skin, ultra-detailed skin texture, shot on 85mm lens, f/1.8 aperture, shallow depth of field, 8k resolution, photorealistic.
Social Media Post - Candid cafe lifestyle photo — output

Social Media Post - Candid cafe lifestyle photo

AqsaGPT Image 2
A candid vertical lifestyle photo of a young woman sitting at a cozy café or casual restaurant, wearing a cute {argument name="dress pattern" default="light-blue gingham"} pinafore dress over a plain white short-sleeve top. She has straight dark shoulder-length hair with soft bangs and a playful thoughtful expression, looking slightly to the side while gently touching her chin with one finger. She holds a slice of {argument name="food" default="pizza"} in her other hand. A small yellow crossbody bag hangs across her outfit, with delicate bracelets on her wrist. Behind her is a colorful café storefront with large glass windows, cute hand-drawn illustrations, cartoon characters, and warm urban surroundings. A pizza box is visible in the foreground. Natural daylight, soft warm tones, candid smartphone photography, youthful casual aesthetic, realistic skin texture, subtle background blur, high detail, vertical composition, 4:5 aspect ratio.
Comic / Storyboard - Vintage European Park Illustration — output

Comic / Storyboard - Vintage European Park Illustration

TaarukGPT Image 2
Create a vertical editorial travel illustration inspired by vintage European travel posters, featuring a peaceful summer afternoon in a grand city park with a recognizable Manhattan-style skyline in the background. Use delicate hand-drawn ink outlines combined with soft, slightly imperfect watercolor washes on warm textured cream paper. Show a wide green lawn filled with people relaxing, reading, walking, jogging, cycling, and having picnics. In the foreground, a casually dressed young couple sits together on a picnic blanket beside a woven basket. Include elegant black vintage park lamps, winding pathways, dense leafy trees framing the composition, and detailed historic and modern skyscrapers rising behind the park. Add a small picturesque stone arch bridge over a calm pond near the bottom of the artwork. Use muted sage green, olive, warm beige, soft blue, pale gray, and subtle golden sunlight, with natural watercolor bleeding, paper grain, fine pen hatching, and an airy sophisticated travel-journal aesthetic. No text, no letters, no logos, no typography, no captions, no signs. Vertical 4:5 composition, highly detailed, elegant, nostalgic, handcrafted watercolor-and-ink illustration.
Social Media Post - Paris street portrait photography — output

Social Media Post - Paris street portrait photography

EeshaGPT Image 2
Photorealistic Paris street portrait, young woman looking back over her shoulder, long wavy dark-brown hair with {argument name="hair accessory" default="tortoiseshell clip"}, ivory embroidered blouse, brown leather shoulder bag, holding {argument name="flowers" default="delicate white flowers"}. Parisian Haussmann buildings, Café de Flore in background, warm natural daylight, soft bokeh, cinematic candid photography, realistic skin and hair, vertical 4:5.
Comic / Storyboard - Samurai fight scene ukiyo-e triptych — output

Comic / Storyboard - Samurai fight scene ukiyo-e triptych

MayaGPT Image 2
Create a vertical triptych (six tall panels side by side) in a classic Chinese ukiyo-e and ink-wash style. Subject: A {argument name="warrior 1" default="young Chinese samurai girl"} and an {argument name="warrior 2" default="older samurai man"} fighting with katanas in front of old-style Chinese architecture (wooden temples, tiled roofs, shoji screens, stone lanterns). Color and mood: Overall black brownish, sepia-toned palette with an aged-paper texture. The background should be darker than the foreground, using deep browns and near-black shadows to create depth and drama. Composition: - Left panel: The two warriors facing off, swords drawn, tense stance. - Middle panel: Mid-clash, blades crossing, sparks and motion lines, dynamic poses. - Right panel: A decisive strike moment, one warrior lunging forward, the other parrying. Keep the characters’ faces, armor, and clothing consistent across all three panels. Use strong ink outlines, subtle watercolor washes, visible brush strokes, and muted earth tones (ochre, umber, sepia, dark brown). The overall feel should be historical, cinematic, and dramatic.
Product Marketing - Indian woman editorial fashion portrait — output

Product Marketing - Indian woman editorial fashion portrait

LaviniaGPT Image 2
Ultra-photorealistic 8K cinematic editorial portrait of a graceful young {argument name="ethnicity" default="Indian"} woman with long, cascading wavy espresso-brown hair, captured mid-motion as she turns gracefully from front to back while glancing back over her shoulder toward the camera with an alluring, subtle smile. She wears a sculpted halter top shaped like iridescent {argument name="wing color" default="deep forest-emerald and brushed-copper"} butterfly wings,lavishly adorned with cascading champagne teardrop crystals and micro-bead fringe, paired with a matching high-waisted shimmering antique-gold sequined skirt. Styled with ornate platinum filigree drop earrings and a minimalist raw-cut emerald choker.Set on an airy luxury penthouse terrace lined with minimalist glass balustrades and sun-dappled potted olive trees, enveloped in warm golden-hour side lighting that casts delicate rim highlights along her silhouette and fine fabric textures. Captured on a full-frame camera with an 85mm f/1.4 lens, creamy bokeh, shallow depth of field, authentic micro skin texture, cinematic color grade, text watermark in bottom center, --ar 9:16
Profile / Avatar - Playful park portrait photography — output

Profile / Avatar - Playful park portrait photography

𝗦𝗮𝗻𝗶𝗮GPT Image 2
Photorealistic candid outdoor portrait of a young {argument name="ethnicity" default="East Asian"} woman enjoying a sunny afternoon in a quiet green park. She has smooth shoulder-length dark brown hair with full straight bangs, naturally framing her face, soft fair skin, subtle rosy cheeks, and large expressive eyes. She makes a cute playful {argument name="expression" default="pout"} while looking slightly to the side, with one arm raised above her head and her palm held like a small sunshade over her forehead. She wears an oversized black graphic T-shirt with a simple floral design and loose sleeves, along with a delicate beaded bracelet on her wrist. The background features a wide freshly cut green lawn, leafy trees, tall grass, a white wooden fence, and a peaceful rural-style landscape. Warm late-afternoon sunlight shines from behind her, creating a bright glowing sky, gentle rim light around her hair, soft natural highlights, and a dreamy summer atmosphere. Casual smartphone photography, slightly soft focus, natural exposure, subtle lens flare, realistic skin and hair texture, authentic candid pose, shallow depth of field, natural green tones, high-resolution photorealistic image, vertical composition, youthful carefree mood.
Social Media Post - Alpine Valley First-Person POV Portrait — output

Social Media Post - Alpine Valley First-Person POV Portrait

MeemGPT Image 2
Wide-angle first-person POV outdoor photograph: The {argument name="subject" default="same young woman from the reference image"}, preserving her facial identity and overall appearance, is relaxing on her back in a {argument name="setting" default="lush alpine valley"} surrounded by vivid green grass, scattered rocks, and towering rugged mountain formations. She wears a {argument name="outfit detail" default="white paisley bandana-style scarf"} over her head, a relaxed light-gray short-sleeve T-shirt, and blue athletic shorts. A metallic stainless-steel water bottle and a hiking pole rest naturally beside her. The camera is positioned low near ground level, creating an immersive wide-angle perspective as she looks toward the surrounding scenery. Dramatic rocky peaks dominate the distance beneath a rich blue summer sky filled with large soft white clouds, with warm sunlight glowing through gaps in the clouds. Natural daylight, authentic outdoor textures, realistic skin and fabric detail, cinematic depth, vibrant yet natural summer colors, ultra-photorealistic, highly detailed, 8K.
Browse by model, style or topic

Find your next direction

ASTORIE CANVAS

Take the idea further.

Bring your prompts, references and results together in one creative space.

Open your canvas ↗