GPT Image 2 Prompts

Explore a growing library of GPT Image 2 prompts with copyable text, source links, and example images across portraits, posters, characters, UI mockups, and comparison scenes.

GPT Image 2 Prompt Library

Browse copyable GPT Image 2 prompt examples with source references and example images. The feed keeps the visual library flexible while the surrounding copy explains how to use each prompt pattern effectively.

VTuber Stream Thumbnail

An anime-style YouTube stream thumbnail featuring a cheerful female VTuber. She has long {argument name="hair color" default="pink with light blue inner highlights"} hair, blue eyes, and wears black and white cat-ear headphones with a boom mic. She wears a white collared shirt with a black and pink star ribbon and a black choker, smiling with one hand near her chin. The background is a gaming room with {argument name="room lighting" default="purple and blue neon"} lighting, showing a desk equipped with 1 white keyboard, 1 mug, 1 glowing cat figure, 1 game controller, and 1 streaming microphone. The left side features large, bold, pop-art Japanese typography: a bright pink top word "{argument name="main text line 1" default="雑談"}" and a bright blue bottom word "{argument name="main text line 2" default="配信"}". Below is a pink banner reading "{argument name="subtitle text" default="今夜もゆるっとトーク!"}". A red "LIVE" badge sits in the top left. Floating speech bubbles, stars, and hearts decorate the composition.

Source

Japanese Rice Advertisement Banner

A high-quality Japanese food advertisement banner. The background features a warm, softly blurred wooden table with golden rice stalks. On the center-right, {argument name="main dish" default="a bowl of steaming white rice topped with a raw egg yolk and soy sauce being poured over it"} sits in a vertically striped ceramic bowl. To the right, a small white dish contains exactly 3 slices of pickles (one green cucumber, one orange carrot, one white radish), alongside a blurred wooden miso soup bowl. In the top-left corner, large elegant Japanese typography reads {argument name="headline text" default="甘み、つや、粘り。三拍子そろったおいしさ。"} with a smaller subheadline below reading {argument name="subheadline text" default="日本の食卓に、最高の一杯を。"}. In the bottom-left corner, a 5kg plastic bag of rice is displayed with the label {argument name="product name" default="新潟県産 こしひかり"}. In the top-right corner, a circular gold foil award badge displays the text {argument name="badge text" default="米の食味ランキング 特A 獲得"}. Small fine print runs along the bottom right edge. The lighting is warm, appetizing, and professional, highlighting the glossy texture of the rice and egg.

Source

Historical Figure Social Media Mockup

{ "type": "mobile social media app UI mockup", "platform": "Twitter/X dark mode", "header": { "status_bar": "time 19:28, bird icon, signal, wifi, battery", "navigation": "back arrow, 'Tweet' title" }, "post": { "author": { "avatar": "portrait of a Chinese emperor in red robes and black hat", "display_name": "{argument name=\"display name\" default=\"Emperor Zhu Yuanzhang\"} 👑 [verified badge]", "handle": "{argument name=\"handle\" default=\"@Emperor_Ming\"}" }, "content": { "text": "{argument name=\"tweet text\" default=\"I have ascended to the Dragon Throne! Today, I am proclaimed as the Emperor of the Ming Dynasty. The era of Hongwu has begun. Let us rebuild our great nation together!\"}", "hashtags": "#MingDynasty #HongwuEra #NewBeginning", "media_grid": { "count": 3, "images": [ "emperor seated on an ornate golden throne in red and gold robes", "wide shot of a grand Chinese palace courtyard with a large crowd", "emperor on horseback leading an army with a red dragon banner" ] } }, "metadata": { "timestamp": "{argument name=\"timestamp\" default=\"1:36 PM · Jan 23, 1368\"}", "engagement": "5,432 Retweets, 8,765 Quotes, 20.1K Likes, 102.3K Views" }, "actions": "reply, retweet, like (red heart with '1'), share, upload" }, "footer": { "reply_bar": { "avatar": "woman in red", "placeholder": "Reply to Emperor Zhu Yuanzhang..." }, "navigation_bar": "home, search, notifications (red '1' badge), messages" } }

Source

Oriental Fantasy City Poster

平面插画,东方幻想风格高端城市海报设计,竖版9:16构图,整体采用对角线+S型流动构图,从左下向右上延展,画面以深邃黑色为背景,自上而下渐变至浓烈暗红色,形成强烈冷暖对比与空间纵深,背景带微弱星尘与颗粒质感。画面中央一条金色流动能量线条如火焰般蜿蜒贯穿,自底部向上延伸,具有流体质感、粒子光效与渐变高光,局部带细微能量碎屑与体积光。 金色流光中逐层浮现{argument name="城市" default="广州"}城市地标建筑群:广州塔为视觉核心,比例突出,周围融合珠江新城高楼群、猎德大桥及现代与岭南建筑元素,建筑采用“精细线描 + 金色发光体块”表现,轮廓清晰、细节丰富,在金色光晕映衬下仿佛悬浮于虚空,形成超现实空间层次,远景轻微雾化增强纵深感。 画面底部为一位东方白发女性形象,长发飘逸,如烟似雾,与金色流光自然衔接并逐渐融合,发丝半透明带渐变光感,姿态柔美,双目微闭,神情宁静,怀抱一束多彩鲜花,花间点缀微光粒子与星点效果,象征人与城市能量的精神连接,人物细节适度简化以突出整体设计感。 光影集中于金色流线、建筑与人物轮廓,形成强烈明暗对比与视觉聚焦,整体氛围宏大、神秘、具有东方神话意境且略带治愈感。色彩以黑与暗红为基底,高亮鎏金为主视觉强调,金色具备丰富明暗层次,辅以小面积高饱和花束色彩点缀,整体高级克制。 页面文字与画面融合排版:顶部居中宋体大字“{argument name="城市" default="广州"}·中国”,下方小字“{argument name="日期" default="2026/04/20"}”,再下方小字“{argument name="作者" default="LIYUE"}”,文字采用淡金色或柔和暖白色,与整体光影统一。高品质细节,电影级光影表现,体积光与粒子细节丰富,画面干净无噪点,超高清8K分辨率,商业级海报质感。

Source

High-Quality Brand Poster Generator

添付の画像2枚は1人の女性です。この画像を使って。 プロのデザイナーが作成した日本向けの{argument name="ブランド名" default="Tiffany & Co."} ポスター広告のビジュアルを4パターン作って。コピーに「{argument name="キャッチコピー" default="With Love, Since 1837\n内なる強さは見つけるのではなく、築き上げるもの。"}」の文字を入れて。服装は、{argument name="ブランド名" default="Tiffany & Co."} のビジュアルに合わせて着換えて、背景も調整して。 とにかく広告のような画像の品質、カメラのしぼり、ライトを意識して。

Source

Anime Character Reference Sheet

{ "type": "anime character reference sheet", "character": { "name": "{argument name=\"character name\" default=\"真田大助\"}", "appearance": "young warrior with long brown hair, wearing samurai-inspired armor, {argument name=\"main color\" default=\"red\"} chest plate and guards, {argument name=\"secondary color\" default=\"navy blue\"} pleated hakama, white short cape" }, "layout": { "header": { "title": "{argument name=\"character name\" default=\"真田大助\"}", "subtitle": "{argument name=\"character concept\" default=\"戦国時代を舞台にした物語の主人公。日英クォーターの若き武将。\"}", "badge": "設定資料" }, "right_side": { "main_portrait": "large full-body standing pose, confident smile", "background_elements": { "emblem": "six-coin crest", "quote": "{argument name=\"catchphrase\" default=\"この国を守る。その誇りと共に。\"}", "scenery": "monochrome Japanese castle with army banners at the bottom right" } }, "sections": [ { "title": "プロフィール", "position": "top-left", "content": "table with 6 rows and descriptive text" }, { "title": "三面図", "position": "mid-left", "count": 3, "labels": ["正面", "側面", "背面"] }, { "title": "表情差分", "position": "top-center", "count": 6, "labels": ["通常", "微笑み", "真剣", "怒り", "驚き", "考え中"] }, { "title": "衣装・装備詳細", "position": "bottom-left", "count": 9, "labels": ["胸当て", "肩当て", "腕甲(籠手)", "脚甲(脛当て・膝当て)", "革靴", "白マント(短)", "帯", "袴/着物部分", "打刀"] }, { "title": "カラーパレット", "position": "bottom-center", "count": 8, "labels": ["真田赤", "濃紺", "金", "白", "茶", "茶褐色", "肌色", "銀"] }, { "title": "世界観", "position": "bottom-center-right", "content": "paragraph of text describing the setting" } ] } }

Source

Grain Rain Festival Poster

设计一张{argument name="节气" default="谷雨"}的海报,{argument name="风格" default="丰子恺绘画风格"}

Source

3D Cinematic Multi-Panel Promo Poster

{ "type": "cinematic promotional poster", "style": "3D CGI animation style, highly detailed, dramatic lighting, caricature characters", "characters": [ { "id": "char1", "description": "large shirtless man with long black hair, beard, and glasses" }, { "id": "char2", "description": "elderly woman in a kimono with white hair tied up" }, { "id": "char3", "description": "small man with a topknot, glasses, mustache, wearing a bright green sweater" } ], "layout": { "panels": [ { "position": "top", "scene": "wide shot of a town street with buildings and a city skyline in the background", "characters_present": ["char1", "char2", "char3"], "text_overlays": [ "{argument name=\"intro text\" default=\"ある日ーー\"}", "いつもの日常がーー", "こんばんは" ] }, { "position": "middle left", "scene": "close-up of char1 looking shocked against a dark fiery background", "text_overlays": [ "{argument name=\"shocked text\" default=\"私が出禁?\"}" ] }, { "position": "middle right", "scene": "close-up of char2 looking angry and pointing against a stormy sea background", "text_overlays": [ "{argument name=\"angry text\" default=\"海を荒らすな!\"}" ] }, { "position": "bottom", "scene": "char3 pointing, char1 screaming, a second instance of char3 falling backwards, and char2 sitting angrily against a fiery chaotic background", "text_overlays": [ "全然出ない!" ], "bottom_titles": [ "{argument name=\"main title\" default=\"パチンコ軍団親のイメチェン\"}", "{argument name=\"subtitle\" default=\"LINEスタンプ販売中\"}" ] } ] } }

Source

Otome Game Choice Ad Mockup

{ "type": "mobile otome game advertisement mockup", "characters": { "male_target": "{argument name=\"male character description\" default=\"handsome young man with messy black hair, golden eyes, smoking a cigarette, holding a smartphone\"}", "female_player": "{argument name=\"female character description\" default=\"cute anime girl with pink twin-tails, purple eyes, blushing, holding a smartphone\"}" }, "layout": { "top_section": { "main_headline": "危険な彼を攻略せよ!", "sub_headline": "どっちを選ぶ?", "game_logo": "{argument name=\"game title\" default=\"Midnight Love Panic\"}" }, "status_ui": { "love_meter": "LOVE 99% with a nearly full pink progress bar", "male_level": "Lv.99 with 6 hearts (5 filled red, 1 empty)", "female_level": "Lv.1 ねおち with 5 hearts (1 filled purple, 4 empty)" }, "interaction_elements": { "male_message_box": "{argument name=\"male chat message\" default=\"今夜、ヒマ?俺の部屋来れそう?😈\"}", "female_thought_bubble": "え、呼び出し…?行ったらどうなるの…?", "warning_badge": "Red jagged starburst reading 99%が失敗 with sweat drops" }, "decision_section": { "prompt_text": "彼の気持ちをつかむのは、あなたの選択次第——", "buttons": [ { "color": "green", "text": "{argument name=\"choice 1 text\" default=\"会いに行く\"}", "indicator": "downward green arrow" }, { "color": "red", "text": "既読無視", "indicator": "downward red arrow" } ] }, "bottom_section": { "call_to_action": "今すぐプレイ with angel wings", "sub_text": "タップで運命が変わる…", "currency_display": "9,999 + with a purple diamond icon" } } }

Source

Epic Concept/Movie Poster Generator

GPT-Image-2 提示词请围绕{argument name="主题" default="中国历史上的皇帝排名"}自动生成一张顶级概念海报 / 信息图式电影海报。 要求 AI 根据这个主题,自动推导并统一设计以下全部视觉系统,不需要我额外指定: - 核心主体(可以自动判断更适合人物、产品、建筑、器物、符号、场景或抽象意象) - 底部支撑结构 - 上方悬浮符号或精神象征 - 场景包裹元素 - 隐喻系统 - 色彩层级 - 材质对比 - 光影逻辑 - 标题、副标题、辅助文案 - 品牌感与高级感表达方式 最终画面必须是: 一张震撼、精密、统一、电影级、超高细节、可用于高端印刷的概念主视觉海报。 【总风格】 超写实 3D 商业 CGI 渲染,融合电影级布光、奢侈品视觉语言、未来感概念设计与史诗级构图。画面必须具有“唯一主视觉核心”,不能杂乱,不能像拼贴,不能像普通电商海报。 【自动推导规则】 AI 必须依据【主题】自动决定最合适的: 1. 核心视觉隐喻 2. 主体类型与姿态 3. 支撑结构形式 4. 悬浮元素形式 5. 场景外壳与空间氛围 6. 主色、辅色、强调色 7. 材质组合 8. 文字气质与版式风格 【构图规则】 - 绝对高级感 - 强烈中心秩序,整体统一 - 允许中轴对称或接近中轴的史诗级构图 - 视觉重力明确,从上到下形成清晰的层级落点 - 边缘负空间干净、克制、有呼吸感 【视觉质量】 - 超高细节 - 体积光清晰 - 材质真实 - 反射、折射、阴影、雾气、景深自然 - 整体达到高端品牌 campaign key visual / luxury invitation poster / conceptual editorial poster 水准 【排版系统】 - 整体为 90% 视觉,10% 文字 - AI 根据【主题】自动生成最匹配的主标题和副标题 - 标题必须简洁、锋利、有气势 - 文字必须尽量少而准,不要堆字 【署名要求】 在画面底部角落自然加入作者署名: @a9quant

Source

Retro Kids TV Magazine Cover Prompt

2000年代のこども向けテレビ特撮専門月刊誌「{argument name="雑誌名" default="テレビくんマガジン"}」{argument name="号数" default="4月号"}の表紙画像を作って。新番組仮面ライダー系特撮ヒーローのバストアップ実写画像を中央に1人配置。周囲に平成ウルトラマン系、戦隊系ヒーローの小さなカットを配置。実在しない架空のテレビ番組名をいくつかと豪華付録の画像を並べて。ライダーやウルトラマンの名前は必ず実在しないものを想像力豊かに創作してください。赤緑青原色系ボールド縁取りのグラデーションフォントなどを駆使し、とにかく雑多で全部盛りの印象に。Format {argument name="比率" default="1:1.414"} 画像生成モデルは duct-tape-1か duct-tape-2、3 を指定したいです。

Source

Oriental Aesthetics Ink Wash Poster

新中式水墨山水海报,竖版9:16构图,东方极简美学风格, 大面积留白,整体色调为春日清晨氛围(青绿色、雾蓝、淡灰、浅墨),低饱和、清透柔和,高级质感。 画面主体为奇峻巍峨的群山,从中间平静湖面的两侧拔地而起,占据左右两侧画面, 山体以水墨晕染表现,浓淡干湿变化丰富,局部融入淡青绿色渲染,体现春意生机。 山峰被湿润轻柔的晨雾包裹,雾气层层递进,与浅青蓝天空自然融合,形成空气透视与空间纵深。 湖面如镜面般平静,呈现微青绿色调,倒映山体与天空,反射略带柔焦与雾化扩散效果,增强春日湿润与梦幻氛围。 中景一艘带弧形篷顶的小木舟缓慢漂浮,船桨轻触水面形成细腻涟漪,水纹自然扩散,整体保持极静状态。 船上为一位红衣渔女,体量较小(远景比例),人物简化处理为水墨剪影 + 轻微设色, 身着低饱和朱砂红传统服饰(非鲜艳红),颜色略被雾气柔化, 人物面部不刻画细节,仅保留轮廓与姿态(如轻扶船篷或执桨), 红色在水面形成淡淡倒影,作为画面唯一暖色视觉焦点。 岸边点缀疏林与春季新生植被,采用淡墨 + 淡青绿点染,虚实结合,增强节奏与生命气息。 少量飞鸟在远空掠过,轻盈疏散分布,增强空间层次与灵动感。 画面顶部居中竖排书法:“{argument name="标题" default="东方美学"}”,采用传统手写行书或行草风格(王羲之笔意), 笔触自然起伏、提按分明,带飞白与墨韵扩散效果,避免字体感。 书法颜色为深墨青或柔和墨黑,与整体画面统一。 整体风格:水墨 + 现代极简设计融合,春日禅意、空灵湿润、宁静氛围, 冷暖对比克制,电影感光影,高级艺术海报质感,8K超清细节。

Source

GPT Image 2 Prompts FAQ

How should I use these GPT Image 2 prompts?

Start by copying a prompt that matches your target style, then replace the subject, color palette, camera language, or composition details with your own requirements. The most reliable results usually come from editing a strong base prompt instead of writing every prompt from scratch.

Are these prompts good for beginners?

Why does this page include example images and source links?

Will this GPT Image 2 prompt library keep growing?