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.

Cat to Human Portrait Transformation

Using the provided reference image, turn the cat into a realistic human portrait while keeping the same lounging pose, camera angle, framing, and the same plush turquoise chair background. Preserve the cat’s orange-and-white color pattern as inspiration for the human look: make the person a fair-skinned young woman with warm ginger hair and a soft, calm expression. Dress her in cozy off-white loungewear: 1 loose knit V-neck sweater and 1 pair of cream drawstring pants, seated naturally in the chair with relaxed body language. Make it photorealistic, soft natural indoor lighting, editorial lifestyle photo quality.

Source

Dark Space Emperor Movie Poster

A dark epic sci-fi movie poster for {argument name="film title" default="EMPIRE OF THE VOID"}, in a highly cinematic photorealistic space-opera style. Centered composition with a tall alien emperor standing behind a massive circular command table, viewed from slightly below to make the figure feel imposing and godlike. The emperor wears ornate black and gunmetal biomechanical armor with sharp layered pauldrons, engraved metallic filigree, a long dark cloak, and a regal elongated helmet-crown. The face is mostly obscured in shadow, giving a mysterious, intimidating presence. One large clawed hand rests prominently on the glowing table in the foreground. The command table displays a luminous spiral galaxy hologram in blue-violet tones, with concentric interface rings and subtle cosmic light. The setting is the bridge of a colossal starship or throne room, with two curved panoramic windows framing outer space. Outside the windows are 8 visible spacecraft total: 4 on the left side and 4 on the right side, including several large battleships and smaller escort vessels. A huge planet fills much of the left background, while the right background shows a purple-blue nebula and starfield. Two secondary alien officers are seated in the mid-background, 1 on the left and 1 on the right, each at glowing futuristic control consoles. The color palette is deep black, steel gray, indigo, violet, and cold blue, with dramatic rim lighting, glossy metallic reflections, and dense atmospheric shadows. The layout should feel like a premium blockbuster poster: a centered top tagline in small uppercase text reading {argument name="tagline" default="INTELLIGENCE. DOMINION. DESTINY. THE GALAXY BOWS TO NONE."}, the giant title near the bottom in elegant spaced serif capitals with the words {argument name="main title" default="EMPIRE"} on the first line and {argument name="subtitle" default="OF THE VOID"} below it, with the letter O stylized as a glowing black-hole eclipse. Beneath that, add a glowing imperial trident-like emblem centered above the release line {argument name="release text" default="COMING SOON"}. At the very bottom, include a dense strip of tiny production credits in authentic movie-poster typography. Ultra-detailed, symmetrical, ominous, imperial, galactic conquest mood, vertical one-sheet poster, premium VFX key art, sharp focus, high contrast, dramatic sci-fi lighting.

Source

Four Japanese Ojisan Parody Posters

{"type":"four-panel Japanese TV drama and anime parody poster collage","style":"polished cinematic key visuals, high-detail Japanese entertainment poster design, dramatic lighting, glossy typography, realistic-to-stylized mix depending on panel","format":"2x2 grid collage with 4 distinct poster panels separated by thin borders","panels":[{"position":"top-left","title":"おじさん、JKになる。","genre":"body-swap school drama","subtitle":"これは、人生やり直しの物語…のはずで、全然約束してない物語。","tagline":"娘と同じ学校へ転校!? しかもなぜか女子高生の見た目に!?","subject":"a middle-aged man transformed into a Japanese high school girl, standing on a quiet suburban street lined with houses and cherry blossoms","appearance":"shoulder-length dark brown bob haircut with soft bangs, sailor-style school uniform with white blouse, navy collar and cuffs, red necktie, navy pleated skirt, carrying a dark school bag over one shoulder","pose":"three-quarter back view, turning to look toward camera with a calm, slightly uncertain expression","environment":"spring morning, blue sky, utility poles, petals drifting in the air, soft cinematic depth of field","text_bottom":"2025年7月5日(土)より放送開始! TOKYO MX 毎週土曜 22:30〜/BS11 毎週土曜 23:00〜 ほか"},{"position":"top-right","title":"推しはおじさん!?","genre":"idol makeover comedy","subtitle":"〜最年長アイドル、推されてます〜","tagline":"アイドルを夢見るのに、遅すぎることなんて…転生しませんとなんでないよね!?","subject":"a middle-aged man reimagined as a cute pink idol girl in a dressing room","appearance":"long wavy brown hair, pink floral hair accessory, frilly pink dress with lace trim and puff sleeves, soft makeup, holding a makeup brush near her face","pose":"seated at a vanity, seen from behind and slightly to the side, looking into a mirror framed by glowing round bulbs","environment":"warm backstage dressing room, tabletop crowded with cosmetics, perfume bottles, brushes, and small containers, golden mirror light reflecting on wood surfaces","text_bottom":"2025年10月3日(金)より放送開始! AT-X 毎週金曜 21:30〜/TOKYO MX 毎週金曜 22:00〜 ほか"},{"position":"bottom-left","title":"湯けむり女装旅","genre":"hot spring travel drama","subtitle":"しがない凡人、ひとり旅に出しました。","tagline":"誰にも言えないけど、温泉はいつだって、わたしを癒してくれる。","subject":"a middle-aged man presented as an elegant woman in traditional clothing at an outdoor hot spring","appearance":"formal dark blue kimono with floral patterning, pale obi sash, traditional updo hairstyle with decorative flower ornament","pose":"sitting upright on a rock by the water with hands folded neatly in lap, composed and reflective","environment":"misty mountain onsen landscape, river stones, gentle steam, soft natural daylight, serene scenic background","text_bottom":"2025年4月12日(土)より放送開始! BS日テレ 毎週土曜 22:00〜/テレビ愛知 毎週土曜 26:35〜 ほか"},{"position":"bottom-right","title":"魔法少女 おじ☆マギカ","genre":"dark magical girl fantasy parody","subtitle":"〜中年、魔法少女はじめました〜","tagline":"おじさんだって、世界を救ってみせる! 戦ってみせる!","subject":"a middle-aged man transformed into a magical girl heroine in a fantasy battlefield","appearance":"very long blonde twin-tail hair tied with pink ribbons, ornate pink magical girl dress with layered ruffles, white gloves, oversized bow at the chest, heart-shaped wand topped with gold ornament","pose":"full-body heroic stance facing forward, holding the wand upright as if ready for battle","environment":"dark purple fantasy sky, ominous creatures and floating eyes in the background, magical glow, intense dramatic contrast between cute costume and threatening surroundings","text_bottom":"2025年1月10日(金)より放送開始! MBS 毎週金曜 25:53〜/TBS 毎週金曜 26:23〜 ほか"}],"layout":{"count":4,"reading_order":["top-left","top-right","bottom-left","bottom-right"],"shared_design":"each panel uses large vertical or stacked Japanese title typography, a central transformed character, thematic background matching the spoof genre, and bold release-date text along the bottom","color_story":["spring blue and navy","warm pink and gold","natural blue and gray","pink against dark violet"]}}

Source

Surreal Archer With Birds and Animals

A surreal cinematic outdoor portrait of a masked archer standing in a dry field under a deep clear blue sky, framed from about the knees up in a wide horizontal composition. The central figure is a woman-like person in a vintage prairie-style dress and matching headscarf, aiming a simple wooden longbow toward the left side of the frame with an arrow drawn back to the cheek. Her face is fully obscured by a soft rectangular blur or blocky censorship patch, creating an uncanny anonymous effect. The dress is sage green with long puffed sleeves, a gathered waist tied with a sash, and a tiny repeating red-and-white floral dot pattern. A red cloth or scarf is wrapped around the lower face and neck beneath the green head covering. Surrounding her are 12 animals total: 8 white birds flying across the sky at different distances, 2 larger white birds standing on the ground at the right side, 1 deer head entering from the lower left foreground, and 1 sheep or goat head entering from the lower right foreground. Add 1 red-headed crane-like bird at the far right edge, visible in profile. The image should feel dreamlike, strange, and symbolic rather than realistic, with crisp midday sunlight, rich saturated colors, soft film grain, and a slightly vintage editorial-photography look. Use low-angle perspective, dynamic diagonal lines from the bow and arrow, and layered animal placement to create tension and motion. Dry grasses appear at the bottom edge, but the background is mostly open sky. Highly detailed fabric folds, natural bird wing positions, and a balanced surreal composition.

Source

Coordinate-Based Tourism Magazine Cover

GPT-image-2で緯度と経度が{argument name="座標" default="36.578134061857426, 136.64816743538748"}の位置で地元のお昼の時間雰囲気と快晴の天気に合った二人の女性の写真集を作成してください。指定のキャラクターがその場所で観光してるように馴染ませてください。写真集のように美しくレイアウトしてください。9:16をあますことなく活かし、この雑誌の表紙を作成してください。これで{argument name="場所" default="金沢"}への来客数が影響する重要なページだと思って取り組んでください。

Source

Optimistic AI Portfolio Landing Page

{"type":"editorial portfolio website landing page mockup","style":"clean modern web design, airy premium aesthetic, soft editorial typography, pastel coastal palette, subtle glassmorphism, high-end startup portfolio feel","canvas":{"aspect_ratio":"3:2","background":"soft blue sky outside the browser frame"},"browser":{"type":"desktop browser window mockup","top_bar":{"traffic_lights":3,"address_text":"angaisb._ai","right_icons":3},"corner_radius":"large","shadow":"soft floating shadow"},"brand":{"logo_text":"Angaisb_"},"header":{"nav_count":5,"nav_items":["Home","Projects","Writing","Tools","About"],"active_item":"Home","cta_button":{"text":"{argument name=\"button text\" default=\"Contact\"}","style":"rounded yellow pill"}},"hero":{"layout":"split editorial hero with text on left and image scene on right blending across full width","background_scene":"sunny coastal overlook with blue ocean horizon, fluffy white clouds, pink and orange flowers in the foreground, soft dreamy atmosphere","headline":"{argument name=\"headline text\" default=\"AI, ideas, and optimistic futures\"}","subheadline":"{argument name=\"subheadline text\" default=\"Experiments, tools, and thoughts on artificial intelligence\"}","primary_cta":{"text":"Explore Projects","style":"yellow rounded button with arrow"},"secondary_cta":{"text":"Read the latest","style":"text link with arrow"},"human_figure":{"count":1,"description":"young man seen from behind, short brown hair, wearing a bright yellow sweater, centered-right in the hero, looking toward the ocean"},"overlay_card":{"count":1,"style":"translucent frosted glass rounded card","icon":"small yellow circular icon","text_lines":["Building with curiosity.","Designing for impact.","Sharing openly."]},"network_graphic":{"count":1,"description":"delicate white node-and-line constellation graphic in the upper-right sky"}},"main_sections":[{"title":"Featured","position":"upper content area","count":3,"labels":["BloomBench","ThoughtGarden","Open Prompts"],"cards":[{"title":"BloomBench","tag":"Research","description":"A benchmark for evaluating multimodal LLMs on reasoning about real-world visuals.","thumbnail":"flowers by the sea with light UI overlay"},{"title":"ThoughtGarden","tag":"Tool","description":"A personal knowledge garden where ideas connect and evolve.","thumbnail":"rolling green hills under blue sky with translucent interface panels"},{"title":"Open Prompts","tag":"Resources","description":"A collection of prompts, patterns, and pipelines for practical AI builders.","thumbnail":"flowers and sky with translucent interface panels"}],"top_right_link":"View all projects"},{"title":"Latest Writing","position":"bottom left","count":1,"labels":["The Optimism Loop in AI"],"card":{"title":"The Optimism Loop in AI","description":"Why optimism is a force multiplier for creativity, research, and responsible AI.","date":"May 12, 2024","thumbnail":"rear-view person in yellow sweater standing among flowers"},"top_right_link":"View all"},{"title":"Currently Exploring","position":"bottom center","count":3,"labels":["Multimodal Reasoning","AI for Scientific Discovery","Human-AI Collaboration"],"style":"three soft rounded topic pills with small icons"},{"title":"About","position":"bottom right","count":1,"labels":["More about me"],"text":"I’m an AI creator and researcher who believes technology should amplify human potential and help build a better future.","decoration":"delicate pink flower illustration in the lower-right corner"}],"visual_details":{"cards":"white rounded cards with thin light-gray borders and soft shadows","spacing":"generous whitespace and balanced grid","typography":"large elegant serif display headline paired with clean sans-serif UI text","colors":{"primary_text":"deep navy","accent":"warm yellow","background":"white and pale sky blue","secondary_accents":"soft pink, green, and lavender"}}}

Source

Iterative Cinematic Poster Prompts

第一轮:{argument name="产品链接" default="https://t.co/KCaX1FvRqz"} 从这个网站以及网络获取这个产品的信息,并生成能打动用户的宣传海报 第二轮:太美了,参考电影 {argument name="电影1" default="her"} 的配色吧 第三轮:wow 太酷了,再用 {argument name="电影2" default="银翼杀手2049 joi"} 的风格来一版本吧 第四轮:再以{argument name="电影3" default="2001太空漫游"}的风格来一个版本吧

Source

Historical Invention Timeline Grid

Create a square infographic-style illustration showing a {argument name="grid size" default="15×15"} invention timeline grid rendered as a collection of miniature isometric dioramas on aged parchment tiles. The image should look like a curated historical poster that spans from {argument name="starting year" default="1000 BC"} to {argument name="ending year" default="2026"}, with one invention per tile and no repeats. Show the visible portion as a 5 by 5 section of the larger grid, for a total of 25 labeled tiles, arranged in neat rows and columns with thin dark borders. Every tile contains a unique world-changing invention depicted as a small handcrafted diorama with era-appropriate materials, colors, and background details. Use a warm antiquarian palette for ancient and medieval inventions, sepia technical-document tones for early modern inventions, and cooler neon-blue futuristic lighting for modern digital and biotech inventions. Surround the grid with an old-map border and subtle historical cartography textures, blending into faint modern network graphics near the bottom right. Label each visible tile with a small beige caption tag in bold black text. Include exactly these 25 visible inventions in reading order from top left to bottom right: 1) {argument name="first tile label" default="1000 BC"} tile showing an iron plow and simple hand tools in a farm field, captioned "Iron Plow & Tools"; 2) "Concept of Zero" shown as a parchment-like tile with a carved zero symbol and a blue hourglass; 3) "Water Clock" shown as a wooden water-timing device; 4) "Catapult" shown as a wooden siege engine; 5) "Antikythera Mechanism" shown as an intricate bronze gear device in a stone niche; 6) "Concrete" shown as a wheelbarrow filled with wet cement or mortar; 7) "Wheelbarrow" shown as a simple wooden one-wheel cart; 8) "Windmill" shown as a traditional mill on grassy ground; 9) "Gunpowder" shown as a barrel and a mound of black powder with smoke; 10) "Printing Press" shown as a wooden press framed by classical columns; 11) "Mechanical Clock" shown as an ornate standing clock; 12) "Telescope" shown as an early brass telescope on a tripod with sketch-like scientific notes; 13) "Steam Engine" shown on a blue blueprint-style tile with an early locomotive engine; 14) "Electric Battery" shown on a blueprint-like tile with two cylindrical battery cells and wiring; 15) "Telegraph" shown as a Morse telegraph key on technical paper; 16) "Light Bulb" shown glowing warmly on a neutral pedestal; 17) "Automobile" shown as an early red motorcar on a workshop-plan background; 18) "Airplane" shown as a commercial jet aircraft in miniature; 19) "Transistor" shown as a black microchip on a luminous circuit-board tile; 20) "Personal Computer" shown as a beige desktop computer with CRT monitor and keyboard; 21) "World Wide Web" shown as a glowing blue globe over digital circuitry; 22) "Smartphone" shown as a sleek modern phone on a neon-tech background; 23) "CRISPR-Cas9" shown as a DNA double helix in a biotech-blue display tile; 24) "Reusable Rocket" shown as several upright rockets launching with exhaust clouds; 25) "AI (LLMs)" shown as glowing large letters AI in a holographic tech cube; 26) "Brain-Computer Interface" shown as a glowing brain hologram on a blue platform; 27) "Advanced Neural Interface (Implant)" shown as a transparent human head profile with a glowing brain implant, with a small top label reading "2026". Keep the composition highly organized, poster-like, and visually dense, with each tile readable at a glance. Use slight isometric perspective, crisp linework, painterly shading, and polished editorial illustration quality. The overall feeling should combine museum poster, educational timeline, and imaginative historical-tech diorama art.

Source

Smartphone Screen Mockup

Create a {argument name="aspect ratio" default="9:16"} image of an {argument name="subject" default="iPhone playing the world's worst TikTok brainrot video"}.

Source

Football Teammates at Golden Gate Bridge

A realistic outdoor sports portrait of two American football players standing side by side on a waterfront path with the Golden Gate Bridge filling the background on a bright clear day. The player on the left wears a muddy white jersey with the large black number 10, white football pants, black undersleeves, and holds a white helmet at his side. The player on the right wears a muddy dark navy jersey with the large white number 23, white football pants, a white long-sleeve undershirt, and holds a white helmet at his side. Both athletes are full torso to upper thigh in frame, facing the camera, with their inside arms around each other’s shoulders in a friendly post-game pose. Their uniforms are visibly grass-stained and dirty from play. Behind them is blue water, a rocky shoreline, a heavy rust-colored chain barrier, distant hills, and the red-orange towers and suspension cables of the Golden Gate Bridge. Natural daylight, documentary-style sports photography, shallow depth of field, sharp fabric texture, realistic anatomy, candid team-bonding mood. Obscure both faces with smooth rectangular blur blocks for privacy.

Source

Disco Ball Chameleon

A highly polished studio photograph of a {argument name="animal" default="chameleon"} entirely covered in tiny mirrored disco-ball tiles, perched on top of a large mirrored disco ball sphere. Side profile facing right, with a curled tail, four visible legs gripping the sphere, a prominent casque-like head crest, round glossy black eye, and a delicate row of small scales along the jawline. The mosaic tiles should reflect silver, pale blue, and cool white highlights, creating sparkling specular reflections across both the animal and the sphere. Use a soft glamorous disco atmosphere with a smooth teal-blue gradient background, shallow depth of field, and scattered circular bokeh lights in the background. Soft diffused lighting with bright glints on the mirrored surfaces, ultra-detailed texture, photorealistic materials, centered composition, elegant and whimsical luxury aesthetic.

Source

Vintage Confidential Case File

A top-down cinematic still life of a vintage classified investigation dossier on a dark wooden desk, centered and filling most of the frame. Show a stack of 3 worn manila folders with aged beige paper, frayed corners, subtle stains, and realistic paper texture. The top folder is the main focus and is clipped at the upper left with 1 oversized silver paper clip. Across the top center, typewritten in bold black monospaced letters, the headline reads {argument name="headline text" default="CONFIDENTIAL – CASE FILES"}, with a thin horizontal rule beneath it. Add 2 red rubber-stamp marks: one near the top right that says "CONFIDENTIAL" at a slight angle, and one boxed stamp at the bottom right that says "HANDLING INSTRUCTIONS DO NOT COPY – DO NOT DISTRIBUTE". On the right edge of the folder stack, show 1 tabbed side label reading "CASE FILE NO." with a thick black redaction bar below it. The page is formatted like an old typewritten government document with crisp black ink, slightly uneven typewriter alignment, and many thick black redaction bars obscuring key details. Include 5 numbered sections exactly: "1. SUBJECT INFORMATION", "2. CASE SUMMARY", "3. DETAILS", "4. ASSOCIATES / CONTACTS", and "5. REMARKS". In the header area, include the fields "CASE FILE NO.:" with a black redaction bar, "DATE OPENED:" with {argument name="date opened" default="March 14, 20___"}, "PREPARED BY:" with a black redaction bar, "DEPARTMENT:" with {argument name="department" default="Internal Investigations Unit"}, and "CLASSIFICATION:" with {argument name="classification" default="CONFIDENTIAL"}. In section 1, include 6 fields: "NAME:", "ALIASES:", "DATE OF BIRTH:", "PLACE OF BIRTH:", "LAST KNOWN ADDRESS:", and "OCCUPATION:", each followed by black redaction bars of varying lengths. In section 2, add a short paragraph of typewritten case-summary text with 5 separate black redaction bars embedded in the sentence flow. In section 3, add a second paragraph of details text with 4 separate black redaction bars embedded in the paragraph. In section 4, include a 3-column table labeled "NAME", "RELATIONSHIP", and "NOTES", with 3 filled rows, each cell content covered by black redaction bars. In section 5, add a short warning sentence about sensitive information and unauthorized disclosure. Use warm sepia lighting, soft shadows around the folder edges, subtle vignetting, and a moody noir archival atmosphere. The image should feel realistic, tactile, secretive, and slightly dusty, like a confidential detective case file photographed for a crime thriller or worldbuilding prop.

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?