GPT Image 2 Prompts

Trouvez des prompts GPT Image 2 pour portraits, affiches, personnages et couvertures de videos courtes. Copiez et creez tout de suite.

Bibliotheque de prompts GPT Image 2

Choisissez un rendu qui vous plait, copiez le prompt, remplacez le sujet, puis lancez la creation.

total : 2 199

7-Day Fashion Lookbook Infographic

{ "type": "7-day fashion lookbook infographic", "header": { "title": "{argument name=\"main title\" default=\"一周穿搭指南\"}", "subtitle": "{argument name=\"style keywords\" default=\"温柔 | 靓丽 | 优雅\"}", "slogan_cn": "优雅不设限,自信每一天", "slogan_en": "{argument name=\"english slogan\" default=\"ELEGANCE HAS NO LIMIT, BE CONFIDENT EVERY DAY\"}" }, "subject": "{argument name=\"subject description\" default=\"young elegant Asian woman\"}", "layout": { "columns": 7, "column_elements": [ "day_header", "main_portrait", "4_detail_thumbnails", "outfit_specs", "keywords_colors", "3_color_swatches", "star_ratings", "fabric_price", "4_season_icons" ], "days": [ { "day": "周一 (MONDAY)", "outfit": "beige blazer suit", "scene": "场景:重要会议 / 正式商务" }, { "day": "周二 (TUESDAY)", "outfit": "pink blazer suit", "scene": "场景:日常通勤" }, { "day": "周三 (WEDNESDAY)", "outfit": "cream knit cardigan set", "scene": "场景:生活休闲" }, { "day": "周四 (THURSDAY)", "outfit": "champagne slip dress", "scene": "场景:外出私会" }, { "day": "周五 (FRIDAY)", "outfit": "blue knit top, white skirt", "scene": "场景:休闲社交" }, { "day": "周六 (SATURDAY)", "outfit": "white sports bra, purple leggings", "scene": "场景:运动休闲" }, { "day": "周日 (SUNDAY)", "outfit": "beige lounge knitwear", "scene": "场景:居家 / 约会" } ] }, "footer": { "tips": "{argument name=\"footer tips\" default=\"Tips: 根据天气与场合灵活调整,配饰是提升整体造型的关键;保持自信与舒适,才是穿搭的最终目的。\"}", "legend": [ "春: 春季适用", "夏: 夏季适用", "秋: 秋季适用", "冬: 冬季适用" ] } }

Source

Pet Brand Identity System Board

{"type":"brand identity system board","brand":{"name":"{argument name=\"brand name\" default=\"狗东西\"}","english_mark":"{argument name=\"logo letters\" default=\"GDX\"}","industry":"{argument name=\"industry\" default=\"宠物行业\"}","date":"{argument name=\"date\" default=\"2024.05\"}","tagline":"爱它・懂它・陪伴它"},"style":{"overall":"minimalist premium corporate presentation, clean Swiss-inspired layout, large white margins, light gray divider lines, muted neutral background","palette":"deep forest green and warm off-white with small accent swatches","mood":"modern, trustworthy, pet-friendly, refined"},"canvas":{"aspect_ratio":"3:4 vertical","background":"soft warm white"},"header":{"left_title_cn":"品牌视觉识别系统","left_title_en":"BRAND IDENTITY SYSTEM","right_tagline":"爱它・懂它・陪伴它","center_logo":"large custom GDX wordmark with a white dog silhouette integrated into the middle of the logo, bold geometric curves and diagonal strokes, dark green","center_subtitle":"狗东西"},"layout":{"sections":[{"title":"基础信息","position":"upper left","count":3,"labels":["品牌名称","行业属性","设计时间"]},{"title":"设计网格","position":"mid upper left","count":1,"labels":["logo construction grid with ratio 1.618:1"]},{"title":"概念草图","position":"mid upper center","count":4,"labels":["dog head and letters sketch","outline refinement","intermediate combined mark","final simplified mark"]},{"title":"灵感来源","position":"mid upper right","count":6,"labels":["minimal white architectural arch","golden retriever profile photo","curved arch block composition","green leaf close-up","dark green material square","light wood texture square"]},{"title":"创意理念","position":"middle left","count":4,"labels":["设计哲学与符号意义","品牌定位服务","色彩心理学与策略","可扩展性与适应性"]},{"title":"品牌应用","position":"middle center to right","count":7,"labels":["名片 正反面","信纸信封","APP图标","网站首页 / 网站图标","产品包装 / 购物袋","店面门头 / 标识牌","mobile app icon variants"]},{"title":"色彩规范","position":"lower middle left","count":5,"labels":["主色","辅助色","暖灰色","浅绿色","强调色"]},{"title":"字体规范","position":"lower middle center","count":2,"labels":["思源黑体 CN","思源柔黑体 CN"]},{"title":"最小使用尺寸","position":"lower middle right","count":2,"labels":["horizontal logo minimum size","stacked logo minimum size"]},{"title":"安全留白区域","position":"bottom left","count":1,"labels":["clear space diagram around logo"]},{"title":"错误使用示例","position":"bottom center to right","count":5,"labels":["do not distort","do not change colors","do not add effects","do not stretch vertically","do not place on busy photo background"]}],"grid":"strict multi-column editorial board with thin dividing rules and consistent spacing"},"logo_design":{"form":"GDX lettermark","integration":"negative-space dog silhouette embedded between G and X, reading as the letter D while showing a standing dog profile facing right","color":"dark green","subtitle":"Chinese brand name centered below with thin horizontal lines on both sides"},"color_swatches":[{"name":"墨绿色","hex":"#1E3D34"},{"name":"米白色","hex":"#F5F3EF"},{"name":"暖灰色","hex":"#E5E2DB"},{"name":"浅绿色","hex":"#A8C5B1"},{"name":"暖橙色","hex":"#E0A86E"}],"applications":{"mockups":7,"details":"business card set, stationery set, smartphone app icon screen, website hero with dog photo, dark green and kraft shopping bags, storefront signage, grouped square app icon variations"},"typography":{"families":["思源黑体 CN","思源柔黑体 CN"],"weights":["Bold","Medium","Regular"]},"rendering":"high-resolution graphic design presentation board, flat front-facing view, crisp vector logo, realistic mockup inserts, subtle shadows only inside product mockups, no perspective distortion"}

Source

Vintage Skeuomorphic Icon Set

生成一组复古的拟物化风格的{argument name="设备类型" default="老式电子设备"}的图标,图中配有图标名称

Source

Surreal Gothic Hall with Floating Masks

A highly detailed, cinematic wide shot of a grand, dark gothic hall with a {argument name="atmosphere" default="dark fantasy"} aesthetic. In the center, a single figure wearing a {argument name="clothing" default="long white robe"} kneels on a highly reflective stone floor, facing an ornate golden altar illuminated by a row of lit candles. To the right of the kneeling figure, a single {argument name="floor object" default="wooden violin"} rests on the ground. The cavernous room is framed by massive dark stone pillars detailed with {argument name="accent color" default="glowing blue"} ethereal cracks and veins. Suspended from the high ceiling are dozens of {argument name="floating objects" default="white porcelain theatrical masks"} hanging on thin strings, filling the upper half of the space and creating a haunting, surreal atmosphere. The lighting is dramatic and moody, featuring a rich color palette of deep blacks, tarnished golds, and cool blue accents. Format 16:9.

Source

Twelve Gold Saints Card Grid

生成圣斗士星矢12个黄金圣斗士的12宫格卡牌图片,每张卡牌上写上对应的中文名,每行4个,宽高比{argument name="比例" default="16:9"}。

Source

Philosophical Concept Infographic Manuscript

一幅令人叹为观止、极其复杂的「{argument name="主题" default="儒释道根本区别"}」概念世界构建信息图杰作,整体被设计成一部底蕴深厚的东方古典神话传说手稿。背景:纯白色复古纹理画布,浅米黄做旧牛皮纸底色,轻微毛边与水渍感。核心布局:中心视觉采用宏大的「竖版蛋形分层结构」,从上到下依次为释、道、儒三层。边缘边角料:四个角落点缀精细微观插图,主题为古代观测手记、法器与符文。色彩:低饱和鼠尾草绿、浅金、米白为主色调,整体清淡柔和,无刺眼高饱和色彩。细节:建筑线条、山水皴法、莲花纹路、云气层次清晰可见,纤毫毕现。无缝融合:三层之间通过云雾与水流自然过渡,释的佛光、道的太极云气、儒的书卷气息无缝衔接。风格:{argument name="艺术风格" default="古典墨水线稿+低饱和数字水彩"},清淡国风古籍手稿质感。文字标注:纯正繁体汉字,采用斑驳古风宋体,每个标注含短标题+一行诗意描述,由暗金色发丝引线连接对应细节,无重复指向。画幅比例:3:4竖版独立完整标题区(顶部):`儒釋道·根本區別`中心分层标注:顶层「釋」:`釋`,`人與自己的關係`,`無我·治心·放得下`中层「道」:`道`,`人與萬物的關係`,`無為·治身·想得開`底层「儒」:`儒`,`人與人的關係`,`無己·治世·拿得起`左右辅助标注:左侧:`淨`,`清淨明心,斷除煩惱`;`靜`,`順應自然,回歸本心`;`敬`,`敬畏責任,積極入世`右侧:`60歲+修心`,`看淡得失,回歸內心`;`35-55歲做人`,`張弛有度,順應規律`;`7-35歲做事`,`銳意進取,建功立业`底部总结文字:`入世與出世之間的平衡,是高級的生活智慧`

Source

Douyin Livestream Screenshot Prompt

9:16 的图片比例,生成一张抖音直播的截图,里面是 {argument name="人物" default="刘亦菲"} 在直播,{argument name="人物" default="刘亦菲"} 手里拿着牌子,牌子里写着 {argument name="牌子内容" default="今晚直播,欢迎来参亦菲畅聊!"}

Source

Korean English Learning App Promo

{"type":"mobile app promo poster mockup","style":"clean modern Korean edtech marketing graphic, soft light gray background, premium iPhone-style device renders, rounded UI cards, subtle shadows, blue and purple gradient accent palette, minimalist sans-serif typography","subject":"English learning mobile app advertisement layout","branding":{"app icon":"rounded square icon with a blue-to-purple gradient background, large white letters EN, small sparkle star in the top right","headline":["매일 10분,","영어가","습관이 되는 시간"],"body":"재미있는 학습과 실전 표현으로 영어 자신감을 키워보세요."},"layout":{"canvas":"wide horizontal poster","sections":[{"title":"left marketing text column","position":"far left","count":3,"labels":["다양한 학습 콘텐츠","실전 회화 연습","학습 현황 관리"]},{"title":"main app home screen","position":"center-left phone","count":6,"labels":["header greeting","today lesson banner","progress card","weekday mini tracker","recommended content card 1","recommended content card 2"]},{"title":"conversation practice screen","position":"center phone","count":6,"labels":["top progress bar","instruction header","speaker A message card","speaker B message card","key phrase card","bottom microphone button"]},{"title":"word learning screen","position":"top-right phone","count":5,"labels":["top progress bar","large vocabulary word","part-of-speech and meaning line","example sentence block","2 bottom choice buttons"]},{"title":"study statistics screen","position":"bottom-right phone","count":5,"labels":["tab selector","date range row","2 summary stat cards","weekly bar chart","chart labels"]}],"phone_count":4},"screens":{"home":{"status_bar_time":"9:41","greeting":"안녕하세요, {argument name=\"character name\" default=\"지민님\"}! 👋","streak":"🔥 7","banner":{"korean_label":"오늘의 학습","title":"Daily Conversation","subtitle":"일상에서 자주 쓰는 표현 배우기","button":"시작하기","illustration":"coffee cup and small EN flashcard on a blue gradient panel"},"progress":{"section_title":"학습 진행 상황","completion":"75%","monthly_goal_text":"이번 주 목표","monthly_goal_value":"3/4일 학습 완료","weekday_count":7,"weekday_labels":["월","화","수","목","금","토","일"]},"recommendations":{"section_title":"추천 콘텐츠","items":[{"title":"Travel English","subtitle":"여행에서 유용한 표현","level":"초급","thumbnail":"airplane window or travel scene"},{"title":"Business English","subtitle":"비즈니스 상황별 표현","level":"중급","thumbnail":"two business people in suits"}]},"bottom_nav":{"count":4,"labels":["홈","학습","복습","마이페이지"]}},"conversation":{"title":"Daily Conversation","progress":"2/10","header":"다음 대화를 읽고 따라 해보세요.","subheader":"Read the conversation and repeat.","messages":[{"speaker":"A","bubble_text":"What are you doing this weekend?","translation":"이번 주말에 뭐 할 거야?"},{"speaker":"B","bubble_text":"I'm planning to go hiking with my friends.","translation":"친구들과 등산 갈 계획이야."}],"phrase_box":{"label":"핵심 표현","phrase":"be planning to ~","meaning":"~할 계획이다","button":"이문 보기"},"bottom_hint":"눌러서 말해보세요","controls":{"microphone":true,"keyboard":true}},"vocabulary":{"title":"단어 학습","progress":"5/12","word":"{argument name=\"vocabulary word\" default=\"adventure\"}","pronunciation":"[ədˈventʃər]","meta":"명사 모험, 탐험","example_en":"It was a great adventure.","example_ko":"그것은 멋진 모험이었다.","illustration":"misty mountain landscape with a lone hiker wearing a backpack","buttons":["모르는 단어","아는 단어"]},"stats":{"title":"학습 통계","tabs":{"count":3,"labels":["주간","월간","전체"]},"date_range":"5월 12일 - 5월 18일","summary_cards":[{"label":"학습 시간","value":"3시간 45분"},{"label":"연속 학습","value":"7일"}],"chart":{"type":"weekly vertical bar chart","bar_count":7,"x_labels":["월","화","수","목","금","토","일"],"y_labels":["0분","30분","60분"]}}},"decor":{"background_shapes":["large soft blue organic wave in the bottom-left corner","3 small sparkle stars scattered around the composition"]},"rendering":"high-resolution product marketing render, straight-on front view phones, balanced spacing, crisp Korean UI text, polished app-store presentation"}

Source

Japanese Fuji Film Style Portrait

{argument name="aspect ratio" default="9:16"} vertical — {argument name="subject" default="Japanese Fuji film style portrait, single subject"} Fujifilm analog aesthetic (Pro 400H / Superia feel), soft pastel tones, slight green-magenta shift, low contrast, gentle highlight roll-off, fine film grain, subtle halation, slight vignette bright natural daylight, diffused sunlight through window, soft shadows, airy atmosphere young Japanese female idol, natural minimal makeup, fresh glowing skin, realistic texture, slight imperfections outfit: Japanese school uniform (sailor-style or blazer uniform), neatly styled, non-revealing, youthful and clean hair: natural dark hair, straight or softly flowing, a few loose strands pose: front-facing or slight angle toward camera, relaxed posture; one hand gently holding a strawberry near lips, mid-action as if about to take a bite; shoulders relaxed, subtle natural body curve expression: soft playful gaze, light smile or neutral lips, gentle eye contact with camera setting: minimal indoor near window or simple outdoor corner, clean background, everyday atmosphere composition: slightly off-center framing, intimate distance, candid feel mood: fresh, youthful, sweet everyday moment, understated charm quality: ultra-realistic, analog film look, natural imperfections, soft dreamy finish

Source

Zombie Survival RPG Anime Ad Poster

Create a dramatic Japanese mobile game advertisement poster for a zombie survival RPG in anime-meets-cinematic style. Vertical smartphone poster layout, dense promotional design, gritty urban apocalypse background at dusk with ruined city streets, smoke, sparks, debris, damaged buildings, neon signs, and a crowd of zombies advancing from behind. In the center, a young anime woman sits on the ground in a slightly slouched pose, legs bent, holding a black handgun upright in one hand. She has straight dark brown to black shoulder-length hair, pale skin, and an oversized dirty white T-shirt splattered with blood. Large bold Japanese calligraphy is printed vertically on the shirt: "働きたくね〜". Her expression area is simple and unobtrusive so the composition emphasizes the shirt, weapon, and mood. Surround her with multiple zombies in torn clothing, reaching forward from the left and right background. Use an aggressive game key-art design with 8 clearly separated promotional sections. Section 1: top-left game logo block, count 3 text elements, with large distressed English title "Dead or work", smaller Japanese subtitle "ゾンビサバイバルRPG", and heavy black paint-splatter background with white and red text. Section 2: right-side vertical slogan area, count 2 text elements, with white Japanese text "動かない、ただ" above giant red vertical text "生き残れ。" on a rough black brushstroke strip. Section 3: left-middle feature list panel, count 4 feature rows, each row with a white icon in a black square and yellow/white Japanese copy; labels are "ゾンビを倒して生き延びろ", "拠点を強化・カスタマイズ", "仲間と協力プレイ", and "デイリーミッションで報酬GET". Section 4: right-middle bonus campaign panel, count 1 headline plus 3 reward items; headline in a bright jagged yellow burst says "今すぐ始めると 豪華特典プレゼント!" and beneath it are three item boxes labeled "ゴールド ×5000", "レア武器 (7日間)", and "サバイバルキット ×10". Section 5: lower-middle gameplay screenshot strip, count 4 thumbnails with captions; captions are "リアルタイムバトル!", "基地を自由にカスタム!", "仲間とレイドボスに挑め!", and "広大なオープンワールド!". Section 6: bottom call-to-action banner, count 2 text elements, with a huge yellow-and-red Japanese phrase "今すぐダウンロード!" and smaller white/red subtext "あなたの選択が、生死を分ける。". Section 7: bottom-right store badges, count 2 badges, showing App Store and Google Play style download buttons. Section 8: bottom-right QR area, count 1 QR code in a white square. Overall aesthetic: hyper-detailed, high-contrast, grimy, intense, commercially polished gacha/mobile game ad, combining anime character art with realistic zombie horror background. Use bold distressed typography, black brush textures, red warning colors, yellow promotional highlights, metallic UI framing, and cinematic orange-gray lighting.

Source

3D Pop-out Profile Page Illustration

创作一幅超逼真的 3D 插画,描绘一个略微倾斜的 {argument name="平台" default="Twitter/X"} 个人资料页面,背景为简洁的灰色。保留原有的卡通头像。界面必须与真实的 X 截图相似,包含真实的布局、认证徽章、粉丝统计、个人资料横幅和推文部分。 个人资料详情: 一位时尚的{argument name="人物" default="年轻男子"},有着蓬松的亮黑色短发和白皙的皮肤,从个人资料页面的右侧撕开的纸片中跃然而出。他保留了原有的面部特征,只是将表情改为{argument name="表情" default="自然自信的微笑"}。他握着撕开的纸片边缘,纸屑四处飞溅,营造出强烈的 3D 突破效果。 柔和的影棚灯光、电影级的阴影、景深、超高细节、清晰的焦点、逼真的皮肤、逼真的 UI 反射、优质的构图、4K 分辨率、逼真与微妙的皮克斯风格融合。 重要提示: - 请勿更改头像 - 保持 X UI 界面准确 - 保留原有的面部特征 - 角色为男性 - 仅增强笑容 - 确保所有中文文字清晰易读

Source

Spring School Romance Manga Page

{"type":"full-color shoujo manga page","style":"romantic anime illustration, glossy polished coloring, soft spring light, sparkling highlights, delicate linework, cinematic school-romance mood","format":"single page comic with 5 numbered panels","setting":{"location":"Japanese high school campus walkway lined with cherry blossom trees","season":"spring sakura season","time":"daytime","atmosphere":"dreamy, youthful, sentimental, petals drifting through the air"},"characters":[{"role":"schoolgirl","appearance":"long dark brown hair with soft shine, large expressive eyes, slim build, navy school blazer, white shirt, red plaid ribbon bow, carrying a smartphone"},{"role":"schoolboy","appearance":"messy dark brown hair, slim build, navy school blazer, white shirt, loosened red striped tie, school bag over shoulder"}],"layout":{"sections":[{"title":"1","position":"top-right","count":1,"labels":["girl notices a photo opportunity under the cherry blossoms and holds up her phone to take a picture"]},{"title":"2","position":"top-left","count":1,"labels":["close two-shot selfie setup, the girl holds a pink smartphone while the boy praises how beautiful she looks and suggests posting it"]},{"title":"3","position":"middle-right","count":1,"labels":["the girl reacts in surprise and embarrassment, holding her phone close to her face while asking if he saw it"]},{"title":"4","position":"middle-left","count":1,"labels":["the boy confidently shows the phone screen with the cherry blossom photo and says it will definitely go viral because it has a strong sense of style"]},{"title":"5","position":"bottom full-width","count":1,"labels":["large romantic payoff shot of the two students posing together for a selfie beneath blooming sakura trees"]}],"centerpiece":"panel 5 is the largest panel spanning the full bottom width"},"props":{"count":4,"items":["pink smartphone in the girl's hand","smartphone screen displaying a cherry blossom photo","school bags","falling sakura petals"]},"text":{"speech_bubbles":["それ、すごく綺麗。投稿しないの?","今日こそ、いい写真、撮れたかも…!","えっ…見てたの!?","絶対バズるって。その、センスあるし。"],"caption_boxes":["その一枚が、私の春を変えた。"],"side_text":["#青春ってエモいってこういうこと♡"],"panel_numbers":["1","2","3","4","5"]},"composition":"Use black comic panel borders, 2 small panels across the top, 2 small panels across the middle, and 1 large wide panel at the bottom. Emphasize blush, shy romantic tension, fluttering petals, and sparkling shoujo effects. The final panel shows the boy extending his arm for a selfie while making a peace sign, and the girl standing close beside him also posing with a peace sign. Keep the background filled with soft-focus cherry blossoms and a school building in the distance.","color_palette":"pastel pink blossoms, navy uniforms, warm skin tones, bright spring blue sky, soft white sparkles","quality":"highly detailed, social-media-ready manga cover quality"}

Source

Cute Bunny Hiragana Poster

Create a cute Japanese preschool hiragana poster in a soft watercolor storybook style on warm textured parchment paper. The composition is a vertical educational chart with a whimsical spring border of grass, flowers, mushrooms, butterflies, and tiny chicks, designed like a children's classroom learning sheet. At the top, place a wooden signboard header with colorful rounded Japanese text reading {argument name="headline text" default="うさぎのひらがな表"}, decorated with 2 white bunny mascots, one sitting on the left and one peeking from the right, plus small flowers and leaves around the sign. Below the header, arrange a clean 5-row by 8-column grid for a total of 40 square cells, each cell outlined in thin brown lines on a cream background. In every cell, place one large brightly colored hiragana character near the top and one small kawaii rabbit illustration below it, each rabbit acting out or holding an object associated with the word written underneath. Use a different cheerful color for many of the hiragana characters, including reds, oranges, yellows, greens, blues, and purples, with a hand-painted rounded font. Keep the text and illustrations exactly as follows, left to right by row: row 1 has 8 cells labeled あ, い, う, う, え, え, お, お with the words あめ, いい, いい, うさぎ, えんぴつ, えんぴつ, おやすり, おにぎり; row 2 has 8 cells labeled か, き, く, け, せ, せ, そ, そ with the words きくら, たま, サベツン, ケーキ, せんせい, そう, そう, しょうこう; row 3 has 8 cells labeled た, ち, ち, つ, て, て, と, と with the words たまご, たまご, ちびこ, つと, てんとうむし, てしなを, とうこ, ほしい; row 4 has 8 cells labeled な, に, ぬ, ふ, め, ね, の, ほ with the words まな, ひこうき, みんきう, ふたう, てつ, へびす, ようこ, ほし; row 5 has 8 cells labeled ま, み, む, る, れ, れ, ろ, ん with the words わた, まくら, みかん, むぎぎ, めがみ, もも, ろうた, ん. The bunny illustrations should be varied and adorable: examples include holding an umbrella, cuddling grapes, drawing with a pencil, eating a rice ball, sitting with cherry blossoms, standing under an umbrella, holding fruit, eating cake, dressed as a teacher, holding flowers, sleeping under a blue blanket, carrying eggs, hugging a tomato, holding strawberries, blowing bubbles, wearing a blue cap, holding an orange, sitting with cotton, bathing in a wooden tub, and holding candy. Maintain a gentle pastel palette, soft shading, round faces, rosy cheeks, and subtle ink outlines. The overall feeling should be nostalgic, charming, educational, highly detailed, and perfect for a nursery wall chart or a cute slide cover for a Japanese learning presentation.

Source

Two-Era Naval Comparison Poster

Create a cinematic Japanese naval comparison poster in a highly detailed, photorealistic concept-art style with a warm amber sunset palette, dramatic clouds, glowing backlight, and a premium movie-poster plus technical-reference-sheet layout. The composition is vertically stacked in 2 main zones. In the top hero zone, show 2 large semi-transparent officer portraits facing inward, one on the left in an older imperial-era naval uniform and one on the right in a modern maritime self-defense force style dress uniform, both wearing peaked caps and formal dark coats, with a blazing sunset sky behind them. Centered between them, place a large serif English title reading {argument name="headline text" default="Fog of Two Eras"}, with a smaller Japanese subtitle beneath it reading ふたつの時代の霧, and 2 lines of Japanese tagline text above the title. Below the title, on the ocean at golden hour, show 2 warships moving forward on slightly converging paths: on the left, a massive World War II-era battleship with many large gun turrets, tower superstructure, weathered dark hull, and white bow spray; on the right, a modern gray guided-missile destroyer with hull number 115 visible on the bow, sleek radar-equipped superstructure, and foamy wake. Add exactly 4 helicopters in the sky, 2 on the left side and 2 on the right side, plus a suspension bridge and distant modern city skyline on the horizon. In the lower reference-sheet zone, split the page into 2 vertical panels, left for the historical ship and right for the modern ship, each panel styled like a parchment-toned military specification board with dark blue and dark red header bars, fine dividing lines, inset image boxes, and Japanese labels. The left panel should present exactly 1 large perspective render of the battleship, 3 orthographic ship views beneath it, 5 small labeled equipment/detail thumbnails in a row, 4 small head-and-shoulders uniform portraits, 4 full-body character turnaround views of a male naval officer, and 1 bottom scale silhouette of the ship beside a human figure with a 20m marker. The right panel should present exactly 1 large perspective render of the modern destroyer, 3 orthographic ship views beneath it, 6 small labeled equipment/detail thumbnails arranged around the ship references, 4 small head-and-shoulders uniform portraits, 4 full-body character lineup views including a central female officer in white cap and dark uniform, and 1 bottom scale silhouette of the ship beside a human figure with a 20m marker. Use visible Japanese headers and labels throughout, including a left historical title area and a right title area naming the ship as 護衛艦 あさつ (DDG-115). Make the whole image feel like a museum exhibition poster, military design sheet, and prestige historical sci-fi drama key art combined, with ultra-sharp detail, realistic metal textures, ocean reflections, aged paper panels, symmetrical balance, and polished editorial typography.

Source

Chengdu Hand-Drawn Food Map UI

A screenshot-style composition of an AI image generation chat interface on a light gray app background, centered inside a rounded white panel with a soft drop shadow. Near the top is a large pale mint-green chat bubble containing a long Chinese prompt in black sans-serif text, describing a hand-drawn food map of Chengdu. Beneath it, aligned slightly left of center, is a square illustrated map image with rounded corners. The map is a whimsical hand-painted tourist food map titled in large Chinese calligraphy at the upper left: {argument name="headline text" default="成都! 吃货暴走地图"}. The subject is {argument name="city name" default="Chengdu"}, presented as a vintage hand-drawn local cuisine map with warm beige paper texture, watercolor ink outlines, and a playful artisanal travel-poster feel. The map shows simplified streets, landmarks, and district labels in Chinese, not geographically exact, but cute and decorative. Place exactly 12 featured food locations, each marked with a numbered red dot and a small Chinese label, each paired with a detailed illustrated dish. Include 12 dishes total: 1) a bowl of skewers in red chili broth, 2) a bowl of sweet water noodles, 3) rabbit head specialty, 4) guokui or stuffed baked flatbread, 5) hot pot, 6) maocai, 7) dan dan noodles, 8) hongyou chaoshou or chili oil wontons, 9) bobo chicken, 10) fuqi feipian, 11) a spicy stir-fried dish, 12) another classic Chengdu snack platter. Add a few Chengdu landmarks sketched in miniature around the roads, including a pagoda-like tower, traditional gate architecture, modern city buildings, and a central emblematic structure. Frame the entire map with a decorative border made of red chili peppers, green vines, and small leaves. Use a warm color palette dominated by chili red, mustard yellow, olive green, tea brown, and muted jade. In the lower right corner of the map, include a small legend box labeled {argument name="legend title" default="图例说明"} with 4 legend items, each with a tiny icon and Chinese text, plus a compass rose with N, E, S, W. Add many handwritten-style Chinese road names and annotations throughout. Outside the image, the interface includes small gray action icons beneath the green chat bubble on the right, a page indicator reading 2/2, an Edit pill button overlapping the lower left of the map image, a circular download/share button at the lower right of the image, and a sparse bottom toolbar with 4 small reaction/action icons. Overall look: clean modern chat UI presenting a generated square illustration, with the map as the visual focus.

Source

GPT-Image-2 Creative Prompt List

1、生成{argument name="角色" default="不知火舞和貂蝉"}的游戏对战海报图 2、生成一张{argument name="团体类型" default="K-pop团体"}时尚专辑封面 3、请你生成 《{argument name="作品名" default="斗破苍穹"}》 的关键人物关系图 4、帮我截一张上传图片的抖音首页的女网红图

Source

Side-by-Side AI Portrait Comparison

A side-by-side desktop screenshot comparing two nearly identical AI-generated street portraits in separate macOS-style image viewer windows against a light cyan background. The layout has exactly 2 vertical image panels, left and right, each inside a rounded gray window with a toolbar at the top. The left window title begins with "NanoBanana2" and the right window title begins with "ChatGPTImage2". In both panels, show the same scene concept: a fashionable young woman standing on a city street corner during golden hour, framed from about mid-thigh to head, with her face obscured by a soft rectangular blur block. She has long wavy brown hair, a slim figure, and wears an off-shoulder fitted cream long-sleeve top, high-waisted blue jeans, a delicate small pendant necklace, and a large black leather shoulder bag with a gold-tone chain strap over one shoulder. She holds a black smartphone in one hand near her waist. The background in both panels is an upscale tree-lined urban street with warm sunlight filtering through leaves, shallow depth of field, parked and passing cars, old stone buildings, and a visible green traffic light near the upper left area of each image. Keep the two portraits very similar but not identical: the left image is slightly cooler, softer, and more muted with a centered street perspective; the right image is slightly warmer, brighter, and crisper with stronger sunlight on the hair and more dramatic bokeh. Include exactly 2 viewer windows separated by a narrow vertical gap, with subtle drop shadows and a clean tech-comparison presentation, as if capturing a visual comparison between two image generation models.

Source

Long Stretching Tabby Cat

A whimsical minimalist illustration of a {argument name="cat breed" default="brown tabby kitten"} standing upright in the exact center of a very tall vertical canvas, with an absurdly elongated, noodle-like body stretched from near the top to near the bottom of the image. The cat has a small round face with large glossy eyes, a tiny open smiling mouth, upright ears, and both front paws raised straight overhead in a cheering pose, showing 2 visible pink paw pads. Its fur is soft and detailed, with a white muzzle, chest, belly, and inner legs, and brown-and-black tabby stripes along the sides, head, arms, and tail. The body is extremely thin and vertically extended, creating a comical exaggerated proportion, while the hind paws remain tiny and close together at the bottom. Add 1 curved striped tail extending to the lower right side. Above the head, place the playful hand-drawn text "{argument name="caption text" default="nyoooon~"}" in a rounded dark brown font, arched like a semicircle, with 6 short motion lines around the raised paws and text to emphasize stretching. Use a clean off-white or warm light beige background with no scenery, only a faint soft oval shadow beneath the feet. Cute, polished digital illustration style, centered composition, high detail in fur, humorous Japanese sticker or social-media mascot vibe, lots of negative space, extra-tall aspect ratio.

Source

Biological Illustration Page

{argument name="対象" default="デメニギス"}の体の構造を解説するカラー図鑑のページ

Source

Cute Chinese LLM Training Infographic

{"type":"cute educational infographic poster","topic":"{argument name=\"headline text\" default=\"大语言模型的训练过程\"}","subtitle":"{argument name=\"subtitle text\" default=\"从海量数据中学习,变成\"超会聊天\"的小助手!\"}","style":{"overall":"adorable Chinese explainer poster, pastel classroom infographic, rounded panels, soft cream background, hand-drawn cartoon aesthetic, clean vector-like illustration, warm friendly educational tone","palette":["cream","lavender","sky blue","mint green","yellow-orange","pink","soft brown"],"rendering":"high-quality flat illustration with subtle shading, crisp Chinese typography, sticker-like accents, tiny stars, hearts, arrows, and speech bubbles"},"layout":{"format":"horizontal poster","grid":{"rows":2,"columns":4,"count":8},"sections":[{"title":"1. 数据收集","position":"top-left","count":1,"labels":["网页","新闻","对话"]},{"title":"2. 数据预处理","position":"top row, second column","count":1,"labels":["今天 天气 真不错!!","今天|天气|真|不错!!"]},{"title":"3. 预训练","position":"top row, third column","count":1,"labels":["今天气","很好","不错","?","......"]},{"title":"4. 监督微调(SFT)","position":"top-right","count":1,"labels":["问:太阳为什么会发光?","答:因为……","好回答!"]},{"title":"5. 奖励模型训练(RM)","position":"bottom-left","count":1,"labels":["回答A 更好!","回答B 一般般"]},{"title":"6. 强化学习(RLHF)","position":"bottom row, second column","count":1,"labels":["奖励 +1","惩罚 -1"]},{"title":"7. 评估与测试","position":"bottom row, third column","count":1,"labels":["知识能力","推理能力","安全性","稳定性"]},{"title":"8. 部署与应用","position":"bottom-right","count":1,"labels":["聊天","写代码","写文章"]}],"topDecorations":{"count":4,"items":["small sparkle at upper left","pink confetti strokes around title","cute white mascot at upper right holding star wand","speech bubble saying 冲鸭!"]},"bottomDecorations":{"count":4,"items":["small white mascot with bow tie at lower left","summary strip with 7 rounded mini boxes and arrows","centered closing sentence at bottom","sticky note at lower right with encouraging text and heart"]}},"characters":{"main mascots":{"count":4,"types":["white chubby bear-like mascot with pink cheeks","small round robot with dark face screen and green antenna","human teacher girl with ponytail","fluffy white character with magnifying glass"]},"recurring_robot_design":"short cute robot, rounded body, pale green and cream shell, navy face display with glowing eyes, tiny limbs"},"sectionDetails":[{"title":"1. 数据收集","panelColor":"lavender","scene":"a white mascot wearing a purple cap uses a laptop while sitting beside a stack of colorful books; floating rounded labels represent internet sources","caption":"数据越多,知识越丰富!"},{"title":"2. 数据预处理","panelColor":"sky blue","scene":"the mascot in a blue cap sweeps scattered paper scraps into a bucket; a messy sentence is transformed into segmented clean text with an arrow","caption":"把脏乱差,变得整整齐齐~"},{"title":"3. 预训练","panelColor":"mint green","scene":"the robot reads an open green book while speech bubbles around it show simple tokens and responses, suggesting language learning","caption":"像小朋友学习说话一样!"},{"title":"4. 监督微调(SFT)","panelColor":"golden yellow","scene":"a teacher points to a labeled QA card while a cute mascot listens; emphasis on human-labeled high-quality question-answer pairs","caption":"老师带着学,回答更靠谱!"},{"title":"5. 奖励模型训练(RM)","panelColor":"pink","scene":"the robot stands between a green check mark and a red X, comparing two answer options to learn which is better","caption":"学会\"挑好答案\"!"},{"title":"6. 强化学习(RLHF)","panelColor":"lavender-blue","scene":"the robot holds tools while a reward board shows positive and negative score feedback with arrows","caption":"鼓励好的,改掉不好的!"},{"title":"7. 评估与测试","panelColor":"blue","scene":"a fluffy white examiner character holds a magnifying glass beside a clipboard checklist with four green check marks","caption":"全方位体检,确保质量过关!"},{"title":"8. 部署与应用","panelColor":"soft red-pink","scene":"the robot appears surrounded by application icons for chat, writing, coding, and documents, showing real-world deployment","caption":"正式上岗,陪你聊天写作啦!"}],"bottomSummary":{"title":"总结一下:","count":7,"steps":["数据收集 打好基础","预处理 整理数据","预训练 学知识","监督微调 学会回答","奖励模型 学会评判","强化学习 对齐人类偏好","评估测试 质量把关"],"closingText":"{argument name=\"closing sentence\" default=\"这就是大语言模型从\"啥也不懂\"到\"超级学霸\"的成长之路!٩(๑^o^๑)۶\"}"},"language":"Simplified Chinese","quality":"polished social-media-ready infographic, balanced spacing, legible Chinese text, charming and highly shareable"}

Source

FAQ GPT Image 2 Prompts

Comment utiliser ces GPT Image 2 Prompts ?

Commencez par copier un prompt proche du style vise, puis remplacez le sujet, la palette, le langage camera ou les details de composition par vos propres besoins. Modifier un bon prompt de base donne en general des resultats plus stables que tout reecrire depuis zero.

Ces prompts conviennent-ils aux debutants ?

Pourquoi cette page inclut-elle des images d'exemple et des liens source ?

Cette bibliotheque GPT Image 2 va-t-elle continuer a grandir ?

Attribution / License

Certains exemples peuvent inclure des prompts open source tiers. Mentions tierces