LogoGPT Image 2 AI
HjemRessurser
Creation
AI VideoAI ImagePrompts
  1. Hjem
  2. GPT Image 2 Prompts Library
  3. Side 28
Prompt Library

GPT Image 2 Prompts Library · Side 28

Gratis, klar-til-kopi GPT Image 2 prompts for tekst-til-bilde-generering — kuratert fra X, YouMind og det bredere fellesskapet.

Explore 700+ hand-picked GPT Image 2 prompts for posters, infographics, illustrations, and product mockups. Copy, tweak, and run each prompt in one click — reference images included where needed. Side 28 av 64.

Kopier en prompt nedenfor og kjør den direkte i vårt AI-bildeverksted.

1 526 prompts44 kategorier1 241 med referanser6 utvalgte

Viser 649-672 av 1526

Humanity's Last Day X Timeline

A prompt designed to visualize a screenshot of the X (Twitter) timeline on the final day of humanity, showcasing the model's ability to render text and UI elements consistently.

Av もみむ🗯️AIもみむ出版
21.4.2026
GPT Image 2 prompt
{argument name="theme" default="A screenshot of X's timeline on the last day of humanity"}
Referanse

Neon Macau RPG Street Battle HUD

This prompt generates a cinematic AAA-style gameplay screenshot of a sword-wielding protagonist in a neon-soaked 1990s Macau street, ideal for game concept art, UI mockups, or action RPG promotional visuals.

Av steventang
21.4.2026
GPT Image 2 prompt
A cinematic third-person action RPG gameplay screenshot set in {argument name="city setting" default="Macau"} in {argument name="year" default="1995"}, viewed from behind a lone male protagonist walking down the center of a crowded wet street at night. The scene is dense with neon signage in Chinese characters, casino signs, jewelry shops, banks, and old mid-rise buildings packed tightly on both sides, with a glowing landmark tower in the distance shaped like a giant illuminated lotus or stylized casino facade. The main character is centered in the frame, wearing a dark leather jacket and jeans, seen from the back, holding a glowing blue energy sword in his right hand and a second short weapon or baton in the left. The street is lined with 6 clearly visible cars, including vintage sedans and 2 taxi cabs, while many pedestrians stand on sidewalks and spill into the street, creating a busy urban nightlife atmosphere. The pavement is rain-slick and highly reflective, mirroring the red, blue, gold, and green neon lights for a moody cyberpunk-meets-1990s crime-drama look. Add a complete in-game HUD overlay: top left has a circular character portrait, the player name {argument name="player name" default="Mayar"}, a red health bar reading 1230 / 1930, a blue energy bar reading 508 / 1160, and a small green status icon; top right has a circular minimap and the location text {argument name="location label" default="MACAU CHRONICLES: 1995"}; bottom left has a semi-transparent chat box with multiple lines of player chat text; bottom right has 6 numbered ability icons in a hotbar, with blue sword-slash skills, a silhouette movement skill, a hand-power icon, a golden weapon skill, and a glowing ultimate icon. Highly detailed, photorealistic game-engine rendering, centered composition, dramatic atmospheric lighting, realistic crowd scale, immersive AAA open-world RPG interface, nostalgic 1990s East Asian street mood with subtle sci-fi fantasy elements.
Referanse

Rainy Neon Macau RPG Screenshot

This prompt generates a photorealistic open-world action RPG gameplay scene in a rainy neon-lit old Macau street, ideal for game concept art, UI mockups, or cinematic promo visuals.

Av steventang
21.4.2026
GPT Image 2 prompt
A cinematic third-person action RPG gameplay screenshot set in {argument name="city district" default="Macau Peninsula, New Road"}, styled like a gritty realistic open-world crime game from the late 1990s. The camera is positioned behind a lone male protagonist standing in the middle of a rain-soaked street, full body visible from the back, wearing a dark brown leather jacket with a large embroidered dragon on the back, blue jeans, and dark shoes. He faces a dense neon-lit urban street packed with pedestrians, old shopfronts, hanging signs, tangled cables, weathered apartment buildings, and a tram approaching from the center distance. The scene is at dusk under a cloudy sky, with wet asphalt reflecting orange, pink, blue, and white neon signage. The environment should evoke old Macau with a Cantonese-Portuguese visual mix: narrow streets, vintage cars, crowded sidewalks, multilingual storefront signs, one large glowing sign reading GRAND LISBOA, another reading BANCO TAI SIN, and many vertical Chinese neon signs layered into the depth of the street. Include 1 tram in the middle lane, 1 black sedan parked on the left, and around 10 to 14 pedestrians crossing or standing on both sides of the road. Add realistic puddles, crosswalk lines, metal railings in the right foreground, storefront interiors with warm lighting on the left, and a moody noir atmosphere. Overlay a complete game HUD: top left circular character portrait with the name {argument name="character name" default="龍少爺"}, level 36, 3 horizontal status bars; top center compass with N and E markers plus the location text and a clock reading 14:35; top right circular minimap with icons; right side quest tracker panel containing 3 quest groups with Chinese text, with the top highlighted quest header in orange; bottom left semi-transparent chat box with 4 colored lines of Chinese system and world chat plus a text entry bar and a send button; bottom center hotbar with 7 skill slots labeled 1, 2, 3, 4, 5, Q, E, R, including a large ornate circular ultimate icon at the center and one glowing purple skill icon; bottom right 5 menu icons with Chinese labels for character, inventory, quests, friends, and settings, plus a small ESC indicator. Ultra-detailed, photorealistic rendering, dramatic reflections, dense environmental storytelling, realistic crowd scale, high-end PC game UI, muted cinematic color grading with neon highlights.
Referanse

Fantasy Manga Page in Forest

A detailed monochrome fantasy manga page showing a witch and elf child after defeating two monsters, followed by a comedic emotional misunderstanding reaction shot.

Av ゆきしたミノル@動画
21.4.2026
GPT Image 2 prompt
{"type":"black-and-white fantasy manga page","style":{"medium":"detailed inked manga line art with screentone shading","color":"monochrome grayscale","reading":"Japanese manga page","mood":"comic misunderstanding with dramatic fantasy tension"},"layout":{"page_count":1,"panel_count":3,"panels":[{"position":"top wide panel","description":"A forest road framed by tall trees and rocky banks. Two huge monster corpses lie in the foreground left and right, both charred and still smoking, with blackened cracked hides, horns, claws, exposed teeth, and ash-like texture. Between them in the midground stand 2 adventurers: a young witch and a small elf boy. The witch is a slim young woman with long straight light hair, a wide-brim pointed witch hat, a staff topped with an oval ornament, a short fantasy outfit with corset-like bodice, shorts, thigh-high boots, gloves, and a cape. The boy is short and childlike with messy light hair, pointed elf ears, a tunic fastened at the chest, trousers, boots, and a short cape. Both face forward as if they have just defeated the monsters. There are 3 visible Japanese sound-effect texts: one on the left reading \"ぷす\", and two on the right reading \"ぷす\" and \"ぷす\" to indicate smoke sizzling from the corpses."},{"position":"middle horizontal panel","description":"A dramatic close-up of the elf boy rushing forward toward the viewer with both arms fully outstretched for a hug, hands open and large in perspective. His face is centered but obscured by a rectangular gray censorship block. Motion lines radiate from behind him. On the left is 1 vertical Japanese speech bubble reading \"ママー やったよー\"."},{"position":"bottom horizontal panel","description":"A close-up reaction shot of the witch from the chest up, turned slightly to the side, blushing intensely with wide shocked eyes, parted lips, and one hand raised awkwardly near her face. Her long hair spills from under the hat, and her cloak is clasped with a jeweled brooch. The background is filled with speed lines. On the right is 1 vertical Japanese speech bubble reading \"ばー\"."}],"borders":"clean manga gutters and black panel borders"},"characters":[{"role":"witch","gender":"female","age_appearance":"young adult","hair":{"color":"light","style":"long straight hair with bangs"},"outfit":["wide-brim pointed witch hat","staff","cape with brooch clasp","fitted fantasy top","shorts","thigh-high boots"],"expression_in_page":"confused, embarrassed, flustered"},{"role":"elf boy","gender":"male","age_appearance":"child","race":"elf","hair":{"color":"light","style":"short tousled hair"},"features":["pointed ears"],"outfit":["cape","long-sleeve tunic with chest clasp","pants","boots"],"expression_in_page":"excited and affectionate"}],"creatures":{"count":2,"description":"two giant dragon-like or demonic beast corpses, burned and smoking on both sides of the path"},"composition_notes":"Use strong perspective in the middle panel, dense crosshatching and textured blacks on the monster bodies, airy forest detail in the top panel, and expressive manga speed lines in the lower 2 panels. Keep the page in polished fantasy manga style suitable for a serialized comic."}
Referanse

Fantasy Manga Spell Explosion Page

This prompt generates a high-impact monochrome fantasy manga page with a young mage unleashing a huge spell and a smaller reaction panel, suitable for comic storytelling or manga-style concept art.

Av ゆきしたミノル@動画
21.4.2026
GPT Image 2 prompt
A dramatic black-and-white fantasy manga page in highly detailed ink line art and screentone style, vertical comic layout with 2 panels. The top panel is a full-width action scene showing a young anime-style mage boy, about early teen age, standing on cracked rocky ground and casting a powerful spell with his right hand extended toward the left side of the frame. He has short tousled light hair, large serious eyes, and a calm, focused expression. He wears a high-collared cape flowing behind him, a buttoned vest over a shirt, fitted trousers, boots, a belt with pouches, and a holstered wand or small magical tool at his hip, with a gemstone clasp at the neck. In front of his open palm is a large glowing circular magic sigil with intricate concentric symbols and runic ornamentation. The spell erupts into a massive explosion of white light and energy, blasting outward with dense speed lines, black debris clouds, shattered rock fragments, dust, and impact textures filling the background. Large stylized Japanese sound effect lettering is integrated into the art on the left side and along the lower foreground, creating a sense of repeated roaring impacts. The composition is dynamic and high contrast, with the boy positioned right of center and the blast dominating the left half. The bottom panel is a narrower reaction shot, count 1 character visible, showing a long-haired witch in profile facing left, wearing a wide-brimmed pointed hat with an ornate band and gem, and a cloak fastened with a jewel brooch. Her face is partially obscured by a solid rectangular censorship block over the central face area. Behind her, the same explosion and flying rubble continue in the distance. A single vertical Japanese speech balloon on the right reads "…うわ". Use clean manga panel borders, polished professional shonen fantasy comic style, intense motion, crisp pen hatching, screentone shading, and cinematic impact.
Referanse

Fantasy Manga Battle Page

A detailed monochrome manga page showing a young witch in a desperate fantasy battle, suitable for generating polished comic-book action scenes.

Av ゆきしたミノル@動画
21.4.2026
GPT Image 2 prompt
{"type":"black-and-white fantasy manga page","style":"high-detail seinen manga ink art, monochrome screentone shading, dramatic speed lines, explosive impact effects, dynamic perspective, polished commercial Japanese manga page","setting":"rocky battlefield or cavern-like wasteland during a monster attack","page":{"reading":"right-to-left manga composition","panel_count":4,"gutter":"clean white borders","sections":[{"title":"top action panel","position":"top","count":1,"labels":["witch casting spell at monster"]},{"title":"reaction panel","position":"upper-middle","count":1,"labels":["close-up with speech balloon"]},{"title":"attack panel","position":"lower-middle","count":1,"labels":["dark projectile rushing toward boy"]},{"title":"bottom dramatic panel","position":"bottom","count":1,"labels":["witch reaching forward and shouting"]}]},"characters":[{"role":"main heroine mage","name":"{argument name=\"character name\" default=\"Noel\"}","appearance":"young woman, slim build, long straight light-colored hair, large expressive eyes, no glasses","outfit":{"count":6,"pieces":["large pointed witch hat with band and oval gem ornament","flowing cape","decorated corset-style top with lacing and filigree trim","short shorts","thigh-high boots","round gem pendant necklace"]},"weapon":"wooden staff with an eye-shaped ornament near the top"},{"role":"secondary character","appearance":"young man with short messy light hair, cloak or scarf around shoulders","placement":"appears at right side of the third panel, facing incoming attack"},{"role":"monster","appearance":"large dragon-like beast with open jaws, rough scales, glowing or intense eyes, aggressive posture"}],"panels":[{"index":1,"composition":"rear three-quarter view of the mage dominating the left side, extending one hand toward the right while holding her staff low in the other hand; a circular magic sigil forms in front of her palm and detonates into a huge white blast colliding with the roaring dragon-like monster on the right; shattered rock, debris, ink splashes, and oversized Japanese sound-effect lettering around the explosion","emotion":"urgent combat"},{"index":2,"composition":"narrow horizontal close-up of the mage's face under the brim of her hat, looking startled with a small sweat drop; speed lines radiate behind her; one vertical speech balloon","text":{"count":1,"items":["あ…!"]},"emotion":"sudden realization"},{"index":3,"composition":"wide horizontal panel showing a black, violent slash or wave-shaped attack tearing across the frame from left toward the boy on the right; heavy motion blur, ink streaks, debris, and large jagged Japanese sound effects emphasize the speed","text":{"count":1,"items":["ザオオオッ"]},"emotion":"imminent danger"},{"index":4,"composition":"large foreshortened close-up of the mage lunging toward the viewer, one hand reaching out desperately, mouth open in a shout, hair and cape blown backward, staff visible to the side; dense speed lines fill the background; large vertical speech balloon on the left","text":{"count":1,"items":["ノエル! 早くー"]},"emotion":"panic and urgency"}],"tone":"dramatic, tense, heroic fantasy battle scene","rendering_notes":"keep everything strictly grayscale with crisp inked contours, detailed hatching, intense contrast, manga-style Japanese lettering left as-is, and cinematic page flow suitable for a fantasy action comic"}
Referanse

Catgirl Student Council Reference Sheet

A polished anime character design sheet showing a black-cat-themed schoolgirl with turnaround views, eight expressions, outfit details, palette swatches, and small story vignettes for character development or visual design work.

Av Nobu-Kobayashi : Generative AI Technology
21.4.2026
GPT Image 2 prompt
{
  "type": "anime character reference sheet",
  "style": "clean polished light novel / visual novel design sheet, white background with pale lavender ornamental border, high-detail cel shading, elegant school-themed layout, soft cool lighting, crisp line art",
  "character": {
    "name": "{argument name=\"character name\" default=\"霧島 澪奈\"}",
    "role": "serious female class committee president with black cat traits",
    "age": "high school student, second year",
    "species traits": [
      "2 black cat ears on top of head with soft inner fur",
      "1 long black cat tail"
    ],
    "appearance": {
      "gender presentation": "female",
      "build": "slender, refined, modestly curvy",
      "height note": "tall for a schoolgirl",
      "hair": {
        "color": "{argument name=\"hair color\" default=\"deep blue-black\"}",
        "style": "very long straight hair with slight waves at the ends, center-parted bangs, long sidelocks, half-up back section tied with a ribbon bow"
      },
      "eyes": {
        "color": "{argument name=\"eye color\" default=\"golden amber\"}",
        "shape": "sharp but elegant"
      },
      "expression base": "cool, intelligent, reserved"
    },
    "outfit": {
      "count": 9,
      "pieces": [
        "navy school blazer with white piping",
        "white dress shirt",
        "large striped ribbon tie in navy and burgundy",
        "school crest on blazer chest",
        "red armband with white Japanese text indicating committee leader",
        "blue plaid pleated skirt",
        "black semi-sheer tights",
        "black loafers",
        "thin gold-framed glasses"
      ]
    },
    "props": {
      "count": 1,
      "items": [
        "dark school folder or notebook with crest"
      ]
    }
  },
  "layout": {
    "format": "landscape character design board",
    "sections": [
      {
        "title": "プロフィール",
        "position": "left column",
        "count": 1,
        "labels": ["bio text block"]
      },
      {
        "title": "SIDE / BACK",
        "position": "upper center",
        "count": 3,
        "labels": ["front full body", "side full body", "back full body"]
      },
      {
        "title": "表情バリエーション",
        "position": "upper right",
        "count": 8,
        "labels": ["通常(真面目)", "微笑み", "注意・叱責", "驚き", "照れ(目をそらす)", "困り顔", "優しい笑顔", "猫っぽく甘える"]
      },
      {
        "title": "ディテール・アクセサリー",
        "position": "middle right left panel",
        "count": 5,
        "labels": ["委員長の腕章", "校章", "猫モチーフのヘアピン", "リボンタイ", "細フレーム眼鏡"]
      },
      {
        "title": "制服ディテール",
        "position": "middle right right panel",
        "count": 3,
        "labels": ["chest and tie close-up", "waist and blazer trim close-up", "sleeve cuff and buttons close-up"]
      },
      {
        "title": "カラーパレット",
        "position": "lower left",
        "count": 5,
        "labels": ["髪", "瞳", "制服", "アクセント", "肌"]
      },
      {
        "title": "特徴的なポーズ・仕草",
        "position": "lower center",
        "count": 4,
        "labels": ["指示を出す(委員長モード)", "資料を確認する", "考え事をする", "休み時間の素顔(少し気が抜けた瞬間)"]
      },
      {
        "title": "その他設定・ポイント",
        "position": "lower right",
        "count": 2,
        "labels": ["classroom vignette", "moonlit night vignette"]
      }
    ]
  },
  "text blocks": {
    "headline": "{argument name=\"headline text\" default=\"霧島 澪奈\"}",
    "subheadline": "学級委員長",
    "quote": "{argument name=\"quote\" default=\"みんなのために、きちんとしなきゃ。それが、私の役目だから。\"}",
    "body text": "dense Japanese profile text with school year, birthday, height, blood type, club or committee, personality notes, likes and dislikes"
  },
  "composition notes": [
    "main full-body figure at left, holding glasses near face and carrying a folder",
    "side and back character turnaround in the center",
    "all portraits and close-ups neatly framed in pale lavender boxes with decorative corner motifs",
    "facial close-ups cropped at shoulders",
    "bottom-right vignette images show the character in a classroom and outdoors at night under a crescent moon",
    "overall palette is navy, black, lavender, white, burgundy, muted gold, and soft skin tones"
  ]
}

Skills Introduction PPT Slide

A simple prompt to generate a presentation slide explaining a specific topic, specifically for gpt-image-2's layout capabilities.

Av 世界是棵树
21.4.2026
GPT Image 2 prompt
What is {argument name="topic" default="Skills"}? Introduce it using a single PPT slide.

JAL First Class Cabin Interior

A specific prompt attempting to generate the first-class cabin interior of a Japan Airlines A350-1000 aircraft.

Av studioあぽろん
21.4.2026
GPT Image 2 prompt
{argument name="aircraft" default="JAL A350-1000"} {argument name="class" default="First Class"}
Referanse

Japanese Absurd Lifestyle Ad Grid

A four-panel Japanese mock advertising collage for surreal comfort products, useful for generating humorous editorial-style product posters with realistic models and clean typography.

Av た〜や|AI副業ワーママ
21.4.2026
GPT Image 2 prompt
{
  "type": "Japanese satirical lifestyle product ad collage",
  "style": "clean magazine advertisement layout, minimalist studio photography, soft natural indoor light, muted neutral palette, humorous premium branding, Japanese copy-heavy poster design",
  "layout": {
    "format": "2x2 grid",
    "count": 4,
    "sections": [
      {
        "position": "top-left",
        "product": "desk corner gloomy fairy breeding set",
        "headline": "デスクの隅に、こっそり住まわせる。 デスクの隅のモヤモヤ妖精飼育セット",
        "romanized subtitle": "MOYAMOYA-YOSEI",
        "theme color": "warm beige and black",
        "scene": "a young Japanese man with short messy black hair in a dark jacket sits behind a desk, his face partially obscured by a rectangular blur, resting one hand near a transparent acrylic display case containing a small fluffy black creature with round beige eyes",
        "counted items": {
          "feature icons": 3,
          "feature labels": [
            "観察ケース(通気口つき)",
            "モヤモヤの素(スペシャル配合)",
            "飼育マニュアル"
          ],
          "physical items on desk": 3,
          "desk items": [
            "transparent display case",
            "black fluffy gloomy fairy creature inside the case",
            "small standing manual booklet labeled 飼育マニュアル"
          ]
        },
        "sales badge": "シリーズ累計販売数 45,000セット突破!"
      },
      {
        "position": "top-right",
        "product": "take-home awkward air pack",
        "headline": "気まずい空気、持ち帰りませんか?",
        "subheadline": "見えない空気を、持ち帰って客観視。",
        "series mark": "KIMA-KUU SERIES",
        "theme color": "white and blue",
        "scene": "a young Japanese man with short textured black hair in a white shirt holds up a clear resealable plastic pouch filled with fluffy white material, the pouch printed with 気まずい空気 and MAZAI? KUUKI, his face partially obscured by a rectangular blur",
        "counted items": {
          "speech bubbles": 1,
          "speech bubble text": [
            "密閉チャックで鮮度そのまま。"
          ],
          "benefit icons": 3,
          "benefit labels": [
            "におい・温度をしっかり保持",
            "ラベリングで原因を可視化",
            "思い出として保存できる"
          ],
          "product variants": 2,
          "variant labels": [
            "気まずい空気保存パック",
            "気まずい空気保存パック(5枚入り)"
          ]
        },
        "usage ribbon": "会議・飲み会・沈黙のあとに。",
        "sales badge": "シリーズ累計販売数 98,000個突破!"
      },
      {
        "position": "bottom-left",
        "product": "procrastination hourglass",
        "headline": "「いつかやる」を可視化する。 先延ばし砂時計",
        "romanized subtitle": "SAKINOBASHI SUNADOKEI",
        "theme color": "soft beige and charcoal",
        "scene": "a young Japanese man with tousled black hair in a dark blazer leans on a wooden table beside a large clear glass hourglass filled with beige sand, his face partially obscured by a rectangular blur",
        "counted items": {
          "feature icons": 3,
          "feature labels": [
            "15分計測",
            "静音設計",
            "ガラス製"
          ],
          "visible text on hourglass": 2,
          "hourglass labels": [
            "まだやれる",
            "やらなきゃ"
          ],
          "color variants": 4,
          "variant labels": [
            "クリア",
            "スモーク",
            "アンバー",
            "ミント"
          ]
        },
        "sales badge": "シリーズ累計販売数 63,000個突破!"
      },
      {
        "position": "bottom-right",
        "product": "weighted shoulder sloth",
        "headline": "肩にのせて、ちょうどいい圧。",
        "subheadline": "存在は感じる。会話はしない。",
        "product name": "圧のかかるナマケモノ(肩乗りタイプ)",
        "romanized subtitle": "ATSU NO KAKARU NAMAKEMONO",
        "brand": "NAMAKEMONO LIFE",
        "theme color": "sage green and warm gray",
        "scene": "a young Japanese man in a light gray sweatshirt faces slightly away while a realistic plush sloth drapes over his shoulders with its arms hanging forward, his face partially obscured by a rectangular blur; a small circular inset at lower right shows the back view of the sloth sitting across the shoulders",
        "counted items": {
          "feature bullets": 3,
          "feature labels": [
            "ほどよい重さ(約1.2kg)",
            "絶妙な沈黙",
            "撫でれる脱力フェイス"
          ],
          "inset photos": 1,
          "inset text": [
            "背中にも ちょうどいい。"
          ]
        },
        "usage ribbon": "在宅ワーク・勉強・読書のおともに。",
        "sales badge": "シリーズ累計販売数 112,000匹突破!"
      }
    ]
  },
  "subjects": {
    "human model": {
      "count": 4,
      "description": "same handsome young Japanese man across all four panels, slim build, short messy black hair, clean stylish casual look, calm slightly tired expression, face covered by a soft-edged rectangular mosaic blur in each panel"
    }
  },
  "composition notes": "each panel uses editorial product-ad composition with large Japanese typography on the upper left or upper center, product hero shot in the lower half, sales numbers in large serif numerals with laurel wreath graphics near the bottom, generous white or beige negative space, realistic props and believable consumer packaging",
  "image quality": "high-resolution commercial photography, crisp detail, subtle shadows, premium catalog finish"
}
Referanse

Japanese Ikemen Product Ad Grid

A 2x2 collage of polished Japanese lifestyle product ads featuring the same handsome young man, ideal for recreating stylish mock social media campaigns or parody brand posters.

Av た〜や|AI副業ワーママ
21.4.2026
GPT Image 2 prompt
{"type":"Japanese social media ad collage","style":"clean modern lifestyle product advertising, minimalist Japanese graphic design, photorealistic studio portraits, soft natural lighting, muted premium color palette, square 2x2 grid layout with thin white dividers","layout":{"grid":"2 columns by 2 rows","count":4,"sections":[{"position":"top-left","product":"foldable portable floor mat","brand":"TATEMAT","primary color":"off-white and dark teal","headline":"移動を、無駄にしない。","subheadline":"もう、立ったまま寝なくていい。","sales badge":"56,000枚突破!","cta":"折りたたみ式 どこでもマット","feature icons count":4,"feature labels":["超軽量 580g","折りたたみ 3秒","防水仕様","専用収納ケース付き"],"usage photos count":3,"usage labels":["駅で","オフィスで","フェスで"],"footer":"あなたの『ちょっと休みたい』を、いつでもサポート。","subject":"a handsome young Japanese man with droopy gentle eyes and tousled dark hair, wearing a light gray overshirt over a white top, leaning his arms on a folded gray mat held upright"},{"position":"top-right","product":"personal safety alarm with idol photo insert","brand":"OSHIBOU","primary color":"deep blue","headline":"推しが、守ってくれる。","subheadline":"推し防犯ブザー","variant text":"写真は自分で選べる 全8色展開","color dots count":8,"feature badge":"大音量 130dB","feature icons count":3,"feature labels":["LEDライト付き","カバンに簡単装着","電池付きですぐ使える"],"sales badge":"販売累計 78,000個突破!","footer":"安心も、推しも、いつも一緒に。","subject":"the same handsome young Japanese man with droopy gentle eyes and tousled dark hair, wearing a navy shirt, holding a small blue keychain safety buzzer near his face with a tiny portrait visible inside the device, city-light bokeh background"},{"position":"bottom-left","product":"stress relief bite ball","brand":"","primary color":"soft peach and cream","headline":"そのイライラ、歯で解決。","subheadline":"噛むだけで気分リセット。","sales badge":"120,000個突破!","feature pills count":3,"feature labels":["食品グレードシリコン使用","水洗いOKでいつでも清潔","選べるかたさ3種類"],"variants count":3,"variant labels":["やわらかめ","ふつう","かため"],"footer":"噛むことは、わたしのメンテナンス。","subject":"the same handsome young Japanese man with droopy gentle eyes and tousled dark hair, wearing a soft cream sweater, holding a peach-colored squishy bite ball shaped like a simple smiling face near his mouth; three product variants displayed along the bottom"},{"position":"bottom-right","product":"designer garbage bag","brand":"DUSTY","brand subtitle":"GARAGE BAGS","primary color":"light gray and charcoal","headline":"見せたくなる、ごみ袋。","subheadline":"生活感を、デザインに。","sales badge":"95,000枚突破!","color swatches count":5,"feature labels count":2,"feature labels":["破れにくい二重構造","においを漏らさない"],"footer":"ごみの日も、ちょっと好きになる。","bottom strip":"DUSTY GARBAGE BAGS / 20L / 30枚入り","subject":"the same handsome young Japanese man with droopy gentle eyes and tousled dark hair, wearing a white hoodie, holding up a semi-translucent gray garbage bag printed with DUSTY in bold black letters"}]},"character consistency":{"count":1,"description":"same male model appears in all 4 ads, attractive ikemen look, gentle droopy eyes, slim build, youthful Japanese man in his 20s, medium-length messy black hair"},"typography":{"language":"Japanese","look":"bold editorial headlines, clean sans-serif body text, premium e-commerce ad layout with laurel sales badges, icon rows, rounded buttons, and small product callouts"}}
Referanse

Japanese Izakaya Highball Poster

This generates a photorealistic Japanese izakaya promotional poster featuring a woman holding a highball, bold pricing text, and four featured dishes for restaurant advertising or nightlife-themed visuals.

Av 蓮華@AIおもしろ研究者
21.4.2026
GPT Image 2 prompt
A photorealistic Japanese izakaya advertisement poster in a warm, cinematic night-time mood, designed like a street-front restaurant flyer. The composition is vertical and densely layered. On the right side, a young Japanese woman with long straight dark brown to black hair, wearing a fitted black ribbed turtleneck sweater, stands from the waist up and holds a large frosty glass mug of golden highball with visible ice cubes in her right hand near chest level. Her pose is relaxed and inviting, looking toward the camera, with a soft nightlife ambiance. The background is a narrow traditional izakaya exterior or interior alley scene filled with warm orange lantern light, shallow depth of field, hanging wooden menu plaques with handwritten Japanese calligraphy, and a nostalgic bustling-after-work atmosphere. In the upper left, a large weathered dark wooden signboard dominates the design, with bold white brushstroke kanji reading 蓮華食堂, small hiragana above it reading れんげしょくどう, and a red brush-painted banner beneath reading 大衆居酒屋. Below that on the left-center, place a large cream-colored paper or painted panel featuring the drink special in oversized calligraphic typography: ハイボール above a huge red 198円, with small black(税込) beside the price, the numerals painted in energetic brush style. Along the far right edge, add vertical white Japanese text reading 今日も一日、おかれさまです。 Near the lower right over the woman’s torso, place a red circular stamp-like badge with white handwritten text arranged diagonally and vertically: 旨い!安い!居心地◎. Across the bottom foreground, show exactly 4 plated menu items lined up side by side on rustic ceramic dishes: 1) two glazed grilled chicken skewers with char marks, labeled 焼き鳥 盛り合わせ and priced 580円; 2) a sashimi assortment with slices of tuna, salmon, white fish, shiso leaves and wasabi, labeled お刺身 盛り合わせ and priced 780円; 3) a plate of fried chicken karaage pieces with a lemon wedge, labeled 若鶏の唐揚げ and priced 480円; 4) sliced rolled dashi omelet topped with chopped green onion and served with grated daikon, labeled 出汁巻き玉子 and priced 380円. At the bottom left, include a glowing red paper lantern graphic with black vertical text 居酒屋. Across the bottom center on a dark band, add large bold white brush lettering reading でかい!うまい!安い! and below it smaller white text reading みんなの憩いの場、蓮華食堂へようこそ! Add a decorative red lotus-like emblem near the top sign and another matching emblem in the bottom right corner. Use rich amber highlights, strong contrast, glossy food styling, realistic skin and drink textures, commercial poster polish, and a nostalgic Showa-inspired izakaya promotion aesthetic.
Referanse

Moody Street Portrait at Dusk

This generates a realistic chest-up urban portrait with soft bokeh and an understated evening mood, useful for lifestyle photography and cinematic character images.

Av 蓮華@AIおもしろ研究者
21.4.2026
GPT Image 2 prompt
A realistic candid street portrait of a young East Asian woman standing centered on a quiet urban road at blue hour, photographed from chest-up with a shallow depth of field. She has {argument name="hair color" default="very dark brown"} long straight-to-softly-wavy hair with a center part, smooth natural texture, and a few loose strands framing the sides of her face. She wears a fitted {argument name="top color" default="black"} ribbed turtleneck sweater with long sleeves, minimal accessories, and no visible patterns. Her pose is relaxed and front-facing, shoulders slightly angled, with a calm understated mood. The setting is a narrow Japanese neighborhood street with 2 visible parked or passing cars in the blurred background, low-rise buildings on the left, a line of trees on the right, utility poles and overhead wires receding into the distance, and soft evening ambient light. Use natural skin tones, realistic photography, subtle cinematic color grading, gentle bokeh, crisp fabric detail, and an intimate moody atmosphere, as if shot on an 85mm lens at wide aperture.

Zelda Map in Minecraft Style

A creative prompt merging two popular game aesthetics by building a Zelda: Tears of the Kingdom scene within a Minecraft screenshot.

Av 金尘马
21.4.2026
GPT Image 2 prompt
Generate a game screenshot of {argument name="base style" default="Minecraft"}, where the world is built into a {argument name="map theme" default="The Legend of Zelda: Tears of the Kingdom"} map scene.

Awkward Idol Stream Situation

A narrative prompt for a humorous scene involving an idol streamer caught in an embarrassing moment by their mother.

Av マシモGPT🤖@ChatGPT/Claude/Gemini/GrokなどLLMをこねくり回す人
21.4.2026
GPT Image 2 prompt
Image generation: A {argument name="protagonist" default="girl"} in an {argument name="outfit style" default="embarrassing magical girl-style idol outfit"} is live-streaming an otaku dance, when her mom walks in furious; the girl is in a total panic trying to stop the recording, while the comment section is a storm of confusion and laughter.
Referanse

Cloud AI Agent Architecture Diagram

A polished bilingual enterprise infographic showing a layered cloud AI agent architecture, suitable for presentations, product explainers, and technical marketing.

Av breeze
21.4.2026
GPT Image 2 prompt
{"type":"enterprise cloud AI architecture infographic","style":"clean corporate vector diagram, white background, blue and teal palette, rounded rectangles, thin blue outlines, subtle shadows, Chinese-first bilingual labels with smaller English subtitles, symmetrical layout, presentation-slide quality","canvas":{"aspect_ratio":"4:3","resolution":"1024x768"},"title_block":{"headline":"{argument name=\"headline text\" default=\"云端 AI Agent 架构图\"}","subtitle":"面向云端任务执行、代码协作与大规模调度的 Cloud AI Agent Architecture","position":"top center"},"layout":{"center":{"type":"stacked architecture layers","count":7,"numbered":true,"layers":[{"index":1,"title":"接入与体验层","english":"access and experience layer","count":5,"items":["Web Console","IDE / CLI","API / SDK","消息通知 Slack / Email","任务看板 组织与项目空间"]},{"index":2,"title":"云端控制平面","english":"cloud control plane","count":6,"items":["API Gateway","Auth / Tenant","Session Manager","Project / Workspace Manager","Task Queue","Rate Limit / Billing"]},{"index":3,"title":"Agent 编排层","english":"agent orchestration layer","count":8,"items":["Planner / 任务规划器","Reasoner / 推理引擎","Context Manager / 上下文管理","Memory / 记忆系统","Prompt Builder","Workflow Orchestrator / 工作流编排","Policy Engine / 策略引擎","Multi-Agent Coordinator / 多 Agent 协同"]},{"index":4,"title":"执行运行时层","english":"execution runtime layer","count":8,"items":["Sandbox Pool / 容器沙箱","Browser / HTTP 工具","Code Runner / Terminal","Git Workspace","Tool Adapters / MCP","Scheduler / Worker","Test Runner","Artifact Executor / Patch & Diff"]},{"index":5,"title":"模型与智能服务层","english":"model and intelligence services layer","count":6,"items":["Model Router","Frontier LLM API","Embedding / Reranker","OCR / STT / TTS(可选)","Evaluator / Reflection","第三方 Agent / SaaS"]},{"index":6,"title":"数据与状态层","english":"data and state layer","count":8,"items":["Relational DB","Vector DB","Redis Cache","Session State","Artifact Store","对象存储 Object Storage","配置中心 Config","日志与审计 Logs"]},{"index":7,"title":"云基础设施层","english":"cloud infrastructure layer","count":8,"items":["Kubernetes / 容器编排","Service Mesh","Event Bus","Secrets / KMS","VPC / Network","Autoscaling","CI/CD","多区域容灾 DR"]}]},"left_sidebar":{"title":"可观测性 / SRE","position":"left","count":6,"items":["监控 Monitoring","Tracing","Metrics","日志聚合 Logs","成本与配额 Cost / Quota","告警 Alerting"]},"right_sidebar":{"title":"安全治理","position":"right","count":6,"items":["SSO / RBAC","密钥与 Secrets","策略护栏 Guardrails","人工审批 Human Approval","审计合规 Audit","数据隔离 Isolation"]},"connectors":"bidirectional blue arrows linking both sidebars to every central layer, plus vertical arrows between each architecture layer"},"visual_details":{"icons":"simple blue line icons in every box, one icon per module","boxes":"rounded white cards with pale blue or pale green tinted section backgrounds","section_colors":{"layers_1_2_5_7":"light blue","layer_3":"light lavender","layer_4":"light cyan","layer_6":"light green","sidebars":"deep royal blue headers with white text"},"typography":"bold dark navy Chinese headline, smaller gray-blue subtitle, medium-weight Chinese section labels, compact bilingual item labels"},"footer":{"position":"bottom center","text":"{argument name=\"footer text\" default=\"特性:云原生 • 多租户 • 弹性扩缩容 • 安全合规 • 人机协同 • 任务可追踪 • 支持多模型与多 Agent 协同\"}"},"quality":"crisp infographic, balanced spacing, high legibility, polished enterprise architecture slide suitable for keynote, documentation, and product marketing"}
Referanse

Desktop AI Agent Architecture Diagram

A polished bilingual Chinese-English enterprise architecture infographic for presentations, product planning, or explaining the layered design of a desktop AI agent system.

Av breeze
21.4.2026
GPT Image 2 prompt
{"type":"bilingual enterprise architecture infographic","topic":"{argument name=\"headline text\" default=\"桌面端 AI Agent 架构图\"}","subtitle":"面向代码生成、项目协作与本地执行的 Desktop AI Agent Architecture","style":{"look":"clean modern SaaS diagram, corporate presentation slide, vector infographic","background":"very light gray-white","palette":{"primary":"deep navy blue","secondary":"medium blue","accent":"teal green","support":"soft gray","security":"bright royal blue"},"typography":"bold Chinese headline with smaller bilingual labels, crisp sans-serif, high legibility","icons":"minimal line icons inside rounded white cards with subtle shadows and blue outlines","geometry":"symmetrical grid, rounded rectangles, thin connector arrows, evenly spaced modules"},"layout":{"orientation":"landscape 4:3 presentation slide","header":{"title_centered":true,"title_size":"large","subtitle_below":true},"sections":[{"title":"可观测性","position":"far left vertical sidebar","count":4,"labels":["监控","Tracing","Metrics","错误告警"],"container":"tall rounded blue panel with four stacked cells and icons"},{"title":"安全治理","position":"far right vertical sidebar","count":4,"labels":["权限校验","敏感信息保护","命令确认","审计追踪"],"container":"tall rounded blue panel with four stacked cells and icons"},{"title":"1 用户交互层","position":"top horizontal layer","count":5,"labels":["桌面客户端 UI","会话面板 / Chat","项目工作区","任务与历史","设置与权限"],"container":"wide rounded white panel with blue border"},{"title":"2 Agent 编排层","position":"second horizontal layer","count":7,"labels":["Planner / 任务规划器","Reasoner / 推理引擎","Context Manager / 上下文管理","Memory / 记忆系统","Prompt Builder","权限与安全策略","Workflow Orchestrator / 工作流编排"],"container":"wide rounded white panel with faint purple-blue tint"},{"title":"3 能力执行层","position":"third horizontal layer","count":8,"labels":["文件系统 File System","代码编辑器桥接 IDE Bridge","Terminal / Shell","Git 与版本控制","搜索与检索 RAG","Browser / HTTP 工具","测试执行器 Test Runner","Patch / Diff 执行器"],"container":"wide rounded white panel with faint teal tint"},{"title":"4 模型与服务层","position":"fourth horizontal layer","count":5,"labels":["云端大模型 API","本地模型 Runtime","Embedding / Reranker","语音 / OCR(可选)","第三方 Agent 服务"],"container":"wide rounded white panel with blue border"},{"title":"5 数据与状态层","position":"fifth horizontal layer","count":6,"labels":["项目索引","Vector Store","本地缓存 Cache","会话状态 State","配置中心 Config","日志与审计 Logs"],"container":"wide rounded white panel with faint green tint"},{"title":"6 系统集成层","position":"bottom horizontal layer","count":6,"labels":["操作系统集成","权限沙箱 Sandbox","文件监听 Watcher","通知服务","后台守护进程 Daemon","自动更新 Updater"],"container":"wide rounded white panel with blue border"}],"centerpiece":"six stacked architecture layers in the middle with side governance columns","connectors":"dashed double-headed arrows connect each central layer to both sidebars; vertical blue arrows flow downward between central layers; small upward gray arrows rise from the bottom integration layer into the data layer","footer":{"count":1,"text":"特性: 本地优先 · 可插拔工具 · 人机协同 · 安全可控 · 支持离线/云端混合部署","position":"bottom center"}},"rendering":{"flat_vector":true,"high_resolution":true,"sharp_lines":true,"balanced_whitespace":true,"professional_pitch_deck_quality":true}}
Referanse

Chinese Co-Branded Summer Drink Poster

A high-impact vertical Chinese advertising poster for a limited-time ice cream and coffee collaboration, suitable for brand campaigns, social ads, and product launch visuals.

Av breeze
21.4.2026
GPT Image 2 prompt
Create a glossy Chinese co-branded beverage poster advertisement in a bold commercial style, vertical 4:5 composition, with a dramatic split-color background: bright saturated red on the left and vivid royal blue on the right, blended at the center with radiant streaks, motion lines, glow effects, and high-energy product-launch lighting. At the top, place 2 brand logo areas separated by a white multiplication sign: on the upper left, a snowman mascot logo with Chinese text {argument name="left brand name" default="蜜雪冰城"} and smaller subtitle text "始于1997冰淇淋与茶"; on the upper right, the deer logo and wordmark {argument name="right brand name" default="luckin coffee"} with smaller Chinese subtitle "专业咖啡新鲜式". Centered beneath the logos, add a huge Chinese headline in thick 3D extruded display typography with white faces, colored edging, heavy shadows, and a glossy embossed look, reading {argument name="headline text" default="蜜雪冰城 × 瑞幸咖啡"}, with the left half accented red and the right half accented blue. Below it, add a second slogan line in bold Chinese text, split by brand color, reading "冰爽甜蜜,咖啡加倍", flanked by short horizontal white dashes. Under that, place a rounded capsule banner in red-to-blue gradient containing the Chinese text "双品牌联名主打・人气招牌同框". In the center foreground, show 2 hero products: on the left, 1 vanilla soft-serve ice cream in a waffle cone, tall swirl, creamy white, slightly glossy, wrapped in a bright red cone sleeve printed with the snowman mascot and Chinese branding; on the right, 1 large clear plastic cup of iced latte or coffee with visible ice cubes, caramel-brown liquid marbling into milk, condensation droplets on the cup, a transparent lid, the blue deer logo, the wordmark "luckin coffee", and small Chinese subtitle text. Surround the cup with dynamic coffee splash arcs and exactly 3 flying coffee beans. Around the lower foreground, add scattered translucent ice cubes and white milk splashes to emphasize cold freshness. Include exactly 2 comic-style promotional burst badges: a red circular badge on the left reading "联名新品" and a blue circular badge on the right reading "限时热卖". In the bottom left corner, add 1 cute 3D snowman mascot character wearing a red cape and holding a tiny ice cream cone, smiling and winking. At the bottom center, place a large red-and-blue beveled ribbon banner with bold white Chinese text reading "今夏限定 立即尝鲜", ending with a yellow lightning bolt icon on the right. Along the very bottom edge, place 3 small selling-point items separated by vertical dividers: a snowflake icon with "冰爽甜蜜", a coffee bean icon with "咖啡加倍", and a heart icon with "双倍满足". Use premium e-commerce poster rendering, ultra-clean product retouching, glossy highlights, rich contrast, crisp Chinese typography, depth, reflections, splash effects, and an energetic summer limited-edition campaign aesthetic.

Fitness SaaS Landing Page UI

A massive, detailed prompt for generating a complete web and mobile UI/UX design for a premium fitness brand, including dashboards and various interface sections.

Av Stephen Day | Google AI, Agents & Vibe Coding
21.4.2026
GPT Image 2 prompt
Design a modern, high-end {argument name="app type" default="fitness app"} landing page and product UI for a brand called {argument name="brand name" default="PulseFit"}.

Create a clean web design mockup that looks like a real SaaS fitness platform and mobile companion app. The design should feel premium, motivating, and easy to use, with a polished UI/UX style similar to top-tier health and wellness apps.

Include:
- A hero section with bold headline, short subheadline, strong call-to-action button, and app preview
- Dashboard UI showing daily activity, calories burned, workout streak, heart rate, step count, and progress rings
- Workout program cards for strength, cardio, mobility, and recovery
- Nutrition tracking section with macro breakdown and water intake
- Progress charts and analytics widgets
- Trainer/community section with profile cards and testimonials
- Pricing section and download/app store callout
- Mobile app screens shown alongside the web design for a cohesive product ecosystem

Style direction:
- Sleek, minimal, modern interface
- Premium fitness brand aesthetic
- Smooth spacing, strong hierarchy, rounded cards, soft shadows
- Dark mode UI with energetic accent colors like electric blue, neon green, or vibrant orange
- Crisp typography, realistic charts, polished buttons, beautiful onboarding feel
- Highly detailed, dribbble-worthy, startup-quality product design
- Realistic UX case study presentation, not a cartoon or illustration

Composition:
- Full website homepage mockup
- Multiple UI sections visible in one polished presentation
- Slight perspective or straight-on product showcase
- Clean background with subtle gradients and professional lighting
- Ultra-detailed, sharp text rendering, visually balanced layout

Output:
A stunning, realistic UI/UX design concept for a fitness app website and dashboard that looks ready for development.

Make it look like a real Figma product design presentation for a premium startup.
Referanse

Satellite Map of Big Ben Area

This prompt generates a realistic labeled satellite-style aerial map centered on Big Ben and the Palace of Westminster, useful for map comparisons, location visuals, and geographic context images.

Av fofr
21.4.2026
GPT Image 2 prompt
A high-resolution top-down satellite map view of central {argument name="city" default="London"}, focused tightly on the area around {argument name="landmark" default="Big Ben"} and the Palace of Westminster beside the River Thames. Show a realistic aerial orthographic perspective like a modern digital map or satellite layer, with detailed rooftops, streets, trees, shadows, and brownish river water. Place the long Palace of Westminster complex vertically near the center-right, directly along the west bank of the river, with Big Ben at the northeast corner of the palace near the river edge. A bridge labeled {argument name="bridge name" default="Westminster Bridge"} crosses horizontally near the bottom of the frame from left to right over the river. Include 6 visible map labels total: {argument name="label 1" default="Westminster Abbey"}, {argument name="label 2" default="Parliament Square"}, Big Ben, Palace of Westminster, Westminster Bridge, and Horse Guards Parade. Also show Whitehall as a labeled road running vertically on the left side of the palace area. Surrounding details should include Westminster Abbey in the upper left quadrant, Parliament Square above the palace, Horse Guards Parade as a large sandy rectangular ground in the lower left quadrant, dense city blocks and roads on the left half, the River Thames occupying the full right side, embankments on both sides, a few boats on the river, and tree-lined streets and parks. Use natural satellite-image colors, crisp cartographic label overlays, and the look of a zoomed-in online maps screenshot.
Referanse

Woman at Elegant Bar Cafe Table

This prompt generates a warm, photorealistic lifestyle portrait of a woman seated at a refined bar-cafe table, ideal for editorial, hospitality, or social media imagery.

Av Ilia Stepin
21.4.2026
GPT Image 2 prompt
A candid, photorealistic vertical photo of a woman seated at a rustic wooden table in an elegant upscale bar-cafe interior. She is centered in the foreground, leaning forward with one elbow on the table and her hand supporting her chin, giving a relaxed, intimate, editorial lifestyle feel. She has long, dark brown wavy hair parted near the middle, falling over both shoulders, and wears a fitted white ribbed long-sleeve scoop-neck top. On the table in front of her are exactly 3 visible objects: 1 peach-colored cocktail in a wide coupe glass on the left, 1 blue-and-white floral ceramic mug in the center-right, and 1 small clear glass bottle vase holding sparse dried stems on the far right. Behind her is a warmly lit, sophisticated bar with marble or stone finishes, ornate classical ceiling details, shelves of liquor bottles and glassware, glowing backbar lighting, and a bartender in a white shirt and dark apron working in soft focus. Include dark bar stools along the counter and a rich amber, cream, and brown color palette. Use shallow depth of field, soft natural indoor light mixed with warm bar lighting, realistic skin and fabric texture, subtle filmic tones, high-detail luxury hospitality photography, shot at eye level with a smartphone or portrait lens, vertical composition, cozy Parisian cafe meets hotel bar atmosphere.
Referanse

Pastel Anime Promo Poster

A colorful Japanese anime announcement poster featuring a white-haired mascot girl with multiple comedic inset scenes, ideal for recreating TV-series key visuals or parody promotional art.

Av ビームマンP ver40
21.4.2026
GPT Image 2 prompt
Create a bright pastel anime promotional poster for a fictional slice-of-life comedy series. The composition is a single-page Japanese key visual with one central heroine surrounded by 5 inset moments and blocks of Japanese promotional text, like an anime announcement poster. The main character is {argument name="character name" default="もち子さん"}, a cute anime girl with short white hair fading slightly to silver, large vivid blue eyes, soft blush, and twin low ponytails tied near the shoulders. She wears an oversized black hood shaped like floppy dog ears, round silver goggles resting on top of her head, a black choker with a metallic round ornament, a futuristic white-and-black cropped sleeveless top with dark trim and a small emblem on the chest, detached black arm sleeves, very short white shorts with black edging, a thigh strap on one leg, and a small side pouch on her hip. Her body proportions are youthful anime stylization with a slim, petite build and slightly exaggerated chest, rendered in polished modern moe character design. In the center she stands full body, smiling gently and holding a pen in one hand as if she is about to make a mistake while writing, facing the viewer in a confident but sweet pose. Around her place 5 supporting vignette images of the same character: 1) upper left, a chibi-style scene of her outside with both hands thrown forward, crying dramatically with a huge open mouth; 2) lower left, a comic scene of her collapsed face-down on a desk or floor with scattered papers, exhausted and flustered; 3) upper right, a bust shot where she holds up a paper sign with Japanese writing, looking apologetic; 4) lower right, a cheerful half-body portrait with closed smiling eyes and one fist raised in encouragement; 5) a small background portrait in the top-right area behind text, used like a character callout. Include exactly 2 small white mascot creatures near the bottom center, both round and mochi-like with tiny dot eyes and tiny mouths, one larger and one smaller, floating or standing beside the heroine. Use a soft spring palette with pink, sky blue, mint green, peach, cream, and white, with scattered cherry blossom petals and rounded graphic shapes in the background. The background should feel airy and celebratory, with subtle outdoor scenery blur and decorative swooshes, stars, and bubbles. Fill the poster with Japanese typography in a playful commercial-anime style: a large multicolored title at top left, small header text above it, a red seasonal broadcast announcement, a pink "あらすじ" synopsis box on the left with several lines of body text, a blue explanatory box at bottom left, vertical and angled exclamatory phrases around the character, and a pink production-credit block at bottom right. Add a prominent character name banner in the upper right. Keep all text looking like authentic printed Japanese promotional copy, using rounded colorful fonts for the title and clean magazine-style fonts for the smaller text. Overall mood: adorable, comedic, slightly chaotic, professionally designed anime key art for a new TV series announcement in 16:9 landscape format.
Referanse

Polaroid of Japanese Class Graduation Photo

This prompt generates a nostalgic realistic instant-film photograph of a Japanese high school class group portrait, ideal for school memory, graduation, or retro snapshot visuals.

Av やきいも|週1クリエイトで飯を食う
21.4.2026
GPT Image 2 prompt
A realistic instant film Polaroid photo lying flat on a light wooden tabletop, photographed from directly above. The instant print has a thick white border with a wider bottom margin and slight natural wear, with the photo area showing an indoor Japanese high school classroom group portrait. Inside the photo, about 35 graduating students in dark school uniforms are gathered in 4 loose rows in front of a green chalkboard, posing casually and tightly packed together. Most are teenage boys in black gakuran-style jackets or dark blazers with white shirts and ties, and several teenage girls wear dark blazers with blue plaid skirts. One girl near the front left stands out in a white cardigan and blue plaid skirt. Many students hold up peace signs, some crouch in front, and a few sit or recline on the wooden floor in the front row. Faces are intentionally obscured with soft square mosaic blurs, but the overall feeling is candid, nostalgic, and celebratory. On the chalkboard, hand-drawn Japanese graduation messages in chalk include large central text reading "3年B組" and celebratory doodles and writing around it such as graduation-themed congratulations. The classroom has white walls, a round wall clock above the chalkboard, a fluorescent ceiling light fixture, and subdued indoor lighting. The image should feel like a slightly dim early-2000s or 2010s snapshot with mild grain, soft focus, authentic flash photography, and natural color fading, emphasizing a sentimental end-of-school memory preserved as a physical instant print.
Referanse

Japanese AI Creator Studio Banner

A polished Japanese promotional hero banner for an AI creator membership community, suitable for website headers, social posts, and announcement thumbnails.

Av やきいも|週1クリエイトで飯を食う
21.4.2026
GPT Image 2 prompt
Create a polished Japanese promotional website hero banner for an AI creator community, in a clean modern tech-advertising style with a soft pastel blue, lavender, pink, and white palette. Wide 16:9 composition. On the left, place a gradient triangular abstract logo and the Japanese brand name {argument name="brand name" default="AIクリエイターズ・スタジオ"} at the top, followed by a thin horizontal divider line with a small sparkle in the middle. Below that, a very large bold Japanese headline in 2 lines: first line in dark navy, 「AIで、もっと自由に」, second line in large gradient letters from cyan to purple to pink, 「創る・学ぶ・つながる」. Under the headline, add a short Japanese subheadline with sparkle marks, 「AIを使ったクリエイティブの可能性を広げるメンバーシップコミュニティ」. On the right half, create a collage of 6 rounded-rectangle visual cards showing AI-generated creative examples: 1 large dreamy portrait of a woman in profile surrounded by flowers and butterflies in ethereal purple light, 1 small cyberpunk city nightscape with a lone figure, 1 small close-up portrait of a fluffy cat, 1 small luxury cosmetic bottle labeled “AI STUDIO”, 1 small bright seascape or flower-field landscape with a woman in white, and 1 large tablet drawing device on a desk with a stylus, plus a coffee mug and plant nearby. Add a circular gradient badge overlapping the right side with bold white Japanese text {argument name="badge text" default="一緒に創造の未来をつくろう!"}. Across the lower middle, place a translucent white rounded feature panel containing exactly 5 equal columns with thin separators and pastel line icons. Each column has an icon, a bold Japanese title, and a smaller 2-line description. The 5 titles are: 「限定コンテンツ」, 「実践プロンプト&テンプレート」, 「メンバー限定コミュニティ」, 「最新情報をいち早くお届け」, 「メンバー限定特典」. Use fitting icons for book, monitor, chat bubble, light bulb, and gift. Add concise Japanese descriptions beneath each title. At the very bottom, add a dark navy footer strip with a large Japanese call-to-action sentence in white and pink-purple gradient accent text, {argument name="footer text" default="AIを味方に、あなたの『創りたい』をカタチに。さあ、一緒に創ろう!"}. Overall look: premium startup landing page, glossy soft lighting, subtle sparkles, glassmorphism panels, clean typography, Japanese text rendered accurately, highly legible, visually balanced, professional social media thumbnail quality.
  • Forrige
  • 1
  • More pages
  • 27
  • 28
  • 29
  • More pages
  • 64
  • Neste