Last updated on 260428 03:41:52 UTC
Languages: English | 简体中文 | 繁體中文 | 日本語 | 한국어 | Français | Deutsch | Español
A curated collection of the best GPT Image 2 prompts and examples. Your go-to resource for mastering prompt engineering and exploring the creative potential of OpenAI's GPT Image 2 model.
This repository focuses on high-fidelity image prompts sourced from X (Twitter) and the community. Whether you are looking for photorealistic game screenshots, stylized aesthetics, or complex creative experiments, you will find the most effective inputs here to unlock the full potential of GPT Image 2.
GPT Image 2 represents a significant upgrade from version 1.5, offering:
- Better world knowledge - More accurate representations of real-world objects and scenes
- Improved style understanding - Can replicate specific art styles and game aesthetics
- Enhanced prompt adherence - Follows complex instructions more precisely
- Photorealistic output - Generates images that look like authentic photographs or screenshots
- 📷 Photography & Photorealism
- 🎮 Game & Entertainment
- 📱 UI/UX & Social Media
- 🎬 Video, Animation & Collage
- 📰 Typography & Poster Design
- 📚 Infographics, Education & Documents
- 🎭 Character & Consistency
- 🖼️ Image Editing & Style Transfer
- 📊 Resources
- 🤝 Contributing
Prompts for creating ultra-realistic scenes, authentic photography characteristics, and complex lighting.
Prompt:
Sam Altman, Donald Trump, and Elon Musk working behind the counter of a busy movie theater
Source: @flowersslop
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Create an ultra-realistic urban street group photo at a convenience store entrance at 10 PM summer night. 3-4 young people briefly chatting at the entrance, someone holding drinks, someone sitting on plastic outdoor chairs, someone standing looking at their phone. Bright white light streaming through the glass doors and windows, warm yellow street lights and distant car headlights outside. Characters wearing everyday clothes: T-shirts, shirts, shorts, jeans, sneakers. No internet celebrity styling. Faces and postures must look like real pedestrians, not overly polished. Environment must include real convenience store elements: freezer stickers, promotional posters, trash cans, entrance mats, glass reflections, shared bikes on roadside, water droplets from drink bottles on ground. The image should look like a very authentic life slice captured by a photographer in the city. Focus on testing natural multi-person interactions, night convenience store lighting, glass reflections, and ordinary people's vibe restoration.
Source: 卡尔的AI沃茨
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Create a completely RAW quality, unprocessed, unedited image with full iPhone camera quality. A subway station in USA, a momentary blur. The subway is in motion. In front of the subway, there is an elderly woman and man.
Source: @WolfRiccardo
Prompt:
Amateur iPhone photo at Apple Park during the iPhone 20 keynote, Tim Cook presenting on stage. Shot from the crowd at a distance
Source: OpenNana | @patrickassale
Prompt:
Amateur photo of an open notebook lying flat, filled with handwritten notes in black ballpoint pen. The handwriting is casual and slightly messy, like personnal notes, natural imperfections, crossed out words, underlined headings. Shot from slightly above, natural daylight from a window, no flash. Casual desk setting, shot on iPhone
Source: OpenNana | @patrickassale
Prompt:
A massive pile of rice, and on one single grain of rice there is tiny text that reads "wOw"
Source: @adonis_singh
Prompt:
360 equirectangular image of [place]
Prompt:
35mm film photography, warm vintage Japanese onsen ryokan aesthetic, soft ambient wooden lantern lighting mixed with gentle natural window light, subtle film grain, gentle color shift, high atmosphere editorial style, intimate medium shot, early 20s beautiful Chinese female idol with ultra-realistic delicate refined Chinese features, seductive almond-shaped fox eyes with natural double eyelids, high nose bridge, small sharp V-shaped jawline, flawless porcelain skin with warm ivory undertone, visible subtle skin texture and micro pores, soft natural makeup with dewy glow, subtle rosy flush on cheeks, natural soft pink lips slightly parted, long dark brown hair tied in a loose low bun with some messy strands falling around face and neck, wearing a loose white yukata (traditional Japanese bathrobe) deliberately slipped off one shoulder and loosely tied at the waist, the fabric slightly open revealing smooth skin and subtle cleavage, barefoot, seductive relaxed sitting pose on the edge of a traditional wooden engawa veranda at a vintage onsen ryokan, body slightly turned toward the camera, one leg bent with foot resting on the wooden floor, the other leg gently dangling, one hand lightly holding the yukata collar, the other hand resting on the wooden floor behind her for support, softly arched back to gently accentuate curves, intensely seductive yet gentle and inviting gaze straight at the viewer with soft doe eyes full of quiet temptation and warmth, warm wooden interior with paper sliding doors and distant steaming hot spring in soft focus, gentle rim lighting highlighting skin and fabric texture, authentic vintage film color grading with warm tones, extremely sharp yet soft skin rendering, natural hair strands, realistic fabric wrinkles and drape on the yukata, no plastic skin, no digital over-sharpening, no airbrushing, no blemishes, no moles, no oily skin, no watermark, no text, authentic 35mm film Japanese onsen ryokan atmosphere
Source: @BubbleBrain
Prompt:
35mm color film photography with harsh direct on-camera flash, specular highlights on skin and clothing, strong catchlights in eyes, high contrast flash illumination, authentic film grain and color shift, high fashion fresh innocent basketball court editorial style, intimate first-person low-angle POV shot from below, early 20s sexy Chinese female idol with ultra-realistic delicate refined Chinese features, seductive almond-shaped fox eyes with natural double eyelids, high nose bridge, small sharp V-shaped jawline, flawless realistic porcelain skin with cool ivory undertone and visible flash specular highlights, fine delicate skin texture with subtle pores micro details and natural dewy glow under flash, fresh natural sporty makeup with soft dewy glow, subtle natural flush on cheeks, natural pink lips slightly parted, subtle natural freckles across nose and cheeks, long dark brown hair tied in a high playful ponytail with some loose strands framing the face and realistic loose strands, wearing a loose white tank top and white high-waisted basketball shorts, white knee-high sports socks, seductive natural leaning pose against the basketball hoop pole on the outdoor court at dusk, body angled sideways with naturally arched back and hips gently pushed back to accentuate perky round hips and sexy butt curve, one leg naturally extended forward toward the camera and the other leg slightly bent to emphasize long sexy legs, both hands lightly resting on the basketball pole at shoulder height, intensely seductive playful yet pitiable doe-eyed gaze straight at the viewer with soft vulnerable longing eyes and a gentle teasing smile full of quiet temptation and desire, harsh direct on-camera flash creating sharp specular highlights and strong catchlights, background with blurred basketball court and hoop under dusk sky, high contrast film color grading with natural flash look, extremely sharp yet soft skin rendering with authentic 35mm direct flash aesthetic, natural hair strands, realistic fabric texture on tank top and shorts with socks detail, no plastic skin, no digital over-sharpening, no airbrushing, no blemishes, no moles, no oily skin, no watermark, no text, authentic 35mm direct flash film basketball court look --ar 9:16
Source: @BubbleBrain
Prompt:
9:16 vertical - editorial portrait, single subject soft black mist filter, subtle haze, gentle highlight bloom, muted tones minimal indoor space, clean background, slight texture young Korean woman, minimal makeup, natural skin texture outfit: fitted ribbed knit top or soft camisole layered under a loose shirt, paired with high-waisted shorts or skirt; fabric slightly clings to body shape, soft and natural, no revealing elements hair: slightly messy, natural volume pose: sitting on floor with one leg bent and the other relaxed, body slightly leaning, shoulders not aligned, head tilted composition: subject slightly off-center, negative space present expression: calm, slightly distant, natural lips lighting: soft side light, gentle shadow falloff mood: understated, quiet, subtly sensual through natural body lines, relaxed and unposed quality: fine grain, slight softness, realistic look
Source: @BubbleBrain
Prompt:
A beautiful woman looking at her phone on the subway; a candid photo.
Source: @AntCaveClub | @underwoodxie96
| Example 1 | Example 2 | Example 3 | Example 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
90s + point-and-shoot camera quality
Source: @sunyunran
Prompt:
Generate a photo from 2003, shot with a digital camera, of five-year-old me with my mom and dad in the courtyard of our residential compound in China.
English Translation: Original prompt: 生成一张 2003 年，用数码相机拍摄的，5 岁的我和爸爸妈妈在中国的家，小区院子里的照片。
Source: @pangyusio
| Example 1 | Example 2 | Example 3 | Example 4 |
|---|---|---|---|
 |
 |
 |
 |
Comment: GPT Image 2 leaked. When were you when OpenAI solved yellow filter in GPT Image? Here are some cool outputs from the model, coming soon. This looks like it's going to be the best image model yet! Source: @marmaduke091
Prompts that leverage GPT Image 2's ability to replicate specific game aesthetics and generate authentic-looking gameplay screenshots.
Prompt:
A Hitman level where you are in the OpenAI HQ and your mission is to steal GPT-6 without getting caught
Source: @flowersslop
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Generate a Black Myth: Wukong game scene where Wukong is knocked away by Erlang Shen
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
GTA 6 in-game footage, very detailed, very realistic. Close-up shot taken from a stationary 4k monitor. (There's a slight blurriness in the image, as it feels like it was taken handheld). A wide, bright environment. Realistic details. The character is walking on the beach with / :dog.
Source: @WolfRiccardo
| Mission 1 | Mission 2 | Mission 3 | Mission 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a GTA 6 gameplay screenshot of [scene description]
Prompt:
Help me generate a screenshot of Trump versus Khamenei in the mid lane in League of Legends
Source: @underwoodxie96
Prompt:
Create a single image containing a grid of 100 completely unique pixel art items, each with a meaningful label
Comment: Reconstructed from the tweet text. The author notes that the full prompt is in a reply. Source: @ProperPrompter
| Scene 1 | Scene 2 | Scene 3 | Scene 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
In the game Zelda totk link is in a e531 series train made by him
Source: @marmaduke091
| GPT Image 1.5 | GPT Image 2 |
|---|---|
 |
 |
Prompt:
gameplay screenshot of a lion fighting against an npc in gta san andreas
Comment: 1.5 is absolute slop, wrong style, not a screenshot, looking like some fake gta game slop, wrong UI, just everything is bad. 2 just looks like it is supposed to look. Source: @flowersslop
Prompt:
日本のソシャゲのガチャ画面を生成して、
English Translation: Generate a Japanese social game gacha screen. Source: @the_wheel_2024
Prompt:
Speed is livestreaming Grand Theft Auto 6
Source: @marmaduke091
Prompts for creating authentic-looking app interfaces, social media posts, and digital designs.
Prompt:
Generate a screenshot of a TikTok live stream featuring a beautiful woman streaming.
Prompt:
9:16 aspect ratio, generate a screenshot of a Douyin live stream, inside is Liu Yifei live streaming, Liu Yifei is holding a sign in her hand, the sign says Tonight's live stream, welcome to join Yifei for a chat!
Source: OpenNana | @alanblogsooo
| Frame 1 | Frame 2 | Frame 3 |
|---|---|---|
 |
 |
 |
Prompt:
Screenshot of a YouTube video showing someone who time-traveled to the Middle Ages
Source: @flowersslop
Prompt:
"Song Dynasty People's Moments"/"SONG DYNASTY SOCIAL MEDIA FEED", Ancient and modern time-travel humor fusion interface design style, The image simulates a mobile phone social media interface, but the content is entirely Song Dynasty scenes, The avatar is a portrait of a Song Dynasty literati, Username "Su Dongpo SuShi_Official", Post content "Just arrived in Huangzhou, demoted but feeling okay. Made Dongpo pork myself today, tastes amazing, recipe attached:", The attached image is a close-up of Dongpo pork in Gongbi painting style, Likes list "Huang Tingjian, Qin Guan, Fo Yin etc. 126 people", Comments section "Wang Anshi: Hehe" "Sima Guang: Still the same taste", Interface elements such as the like icon are replaced with Song Dynasty patterns, The status bar shows "Great Song Mobile 5G" and "Third Year of Yuanfeng", The color scheme is mobile phone dark mode paired with elegant Song Dynasty tones, A masterpiece of fun collision between history and social media
Source: OpenNana | @Panda20230902
| Douyin Live | Xiaohongshu Post | Weibo Post | Moments Post |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a screenshot of [historical figure name] on [platform name]
Comment: Add more story background to increase the meme quality, such as the historical moment, platform behavior, or the punchline you want the post to land. Source: Article | @MrLarus
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Generate a high-fidelity mobile e-commerce app homepage screenshot inspired by mainstream Chinese e-commerce apps in 2026. The interface should look fully realistic, with complete mobile-app UI logic and strong commercial design polish.
Layout requirements:
- status bar with time 9:41, 5G signal, and battery icon
- search area with a city selector on the left, a rounded search box in the center, and message plus scan icons on the right
- horizontal category tabs below the search area
- top promotional carousel banner
- 10-icon function grid
- flash-sale module with countdown and three product cards
- "guess you like" two-column product waterfall with at least 6 product cards
- bottom tab bar with five tabs
Use these exact Chinese labels where appropriate:
- city: "杭州"
- search placeholder: "搜索耳机、咖啡机、运动鞋"
- category tabs: "推荐、数码、家电、服饰、美妆、食品、运动、家居"
- top banner: "618 预售开启" and "每满300减50"
- function grid: "超市、百亿补贴、秒杀、直播、充值中心、到家、领券、品牌馆、全球购、排行榜"
- flash sale title: "限时秒杀"
- recommendation section: "猜你喜欢"
- bottom tabs: "首页、分类、购物车、消息、我的"
All Chinese text must be clear and readable with realistic fonts. Spacing, icon style, white space, shadows, rounded corners, dividers, and tag treatments should closely resemble a real app screenshot rather than concept art.
Source: 卡尔的AI沃茨
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Create a high-fidelity Chinese music app player interface screenshot in mobile portrait orientation, with refined visuals similar to a modern streaming player. Use dark mode, with the background derived from a blurred and diffused version of the album cover colors. Place a large square album cover in the center with subtle shadow and rounded corners.
Interface requirements:
- top status bar with time 18:26
- navigation bar with a back arrow on the left, a centered title, and a more-options icon on the right
- playback progress bar showing current time 01:42 and total duration 04:18
- playback controls for shuffle, previous, play/pause, next, and repeat
- lyrics section showing 5 to 7 scrolling lines, with the current line highlighted
- action row with like, comment, download, add to playlist, and share
- bottom area with device casting and playback queue entry
Use these exact Chinese labels and names:
- title: "正在播放"
- song name: "海边的晚风"
- artist: "林秋"
- album: "夏夜实验室"
The lyrics layout, button icons, reflections, shadows, and dark-mode layering should feel like a real production interface rather than a Dribbble concept.
Source: 卡尔的AI沃茨
Prompt:
Generate a UI design system for me in xx style, including web pages, mobile, cards, controls, buttons, and others
Source: OpenNana | @stark_nico99
Prompt:
Generate image:
Taobao product detail page of a T-800 robot, showing:
front, side, and back three-view drawings of the robot,
product price,
product details,
functions and usage scenarios
Source: OpenNana | @rionaifantasy
Prompt:
玄武门之变的朋友圈
English Translation: A WeChat Moments-style feed for the Xuanwu Gate Incident. Source: @Tz_2022
Prompt:
Generate an image, aspect ratio 3:4. A real-life smartphone photo, with the phone screen displaying a WeChat chat interface in Chinese dialogue, featuring chat bubbles and a Word document attachment. The interface has green and gray bubbles. The screen shows obvious fingerprints, smudges, and scratches. The glass has strong reflections and a direct light source creating glare. The frame is slightly tilted, handheld shot, natural ambient light, imperfect composition, strong sense of realism, casual everyday snapshot, high detail, 4K. The dialogue is between a boss and an employee: the employee sends a file to the boss, and the boss replies that they will take a look first.
English Translation: Original prompt is in Chinese. Source: @Gorden_Sun
Prompts for generating video-like content, animated sequences, and collages.
Prompt:
48 宫格已经满足不了我了，我需要 100 宫格画面 8k
English Translation: A 48-panel grid is no longer enough for me. I need a 100-panel image, 8K. Source: @ianneo_ai
Prompt:
entire superman movie image collage one shot in one output
Source: @chetaslua
![Where Is [Your Name] Crowd Search Poster](https://pbs.twimg.com/media/HGcgIHWWgAEph1p?format=jpg&name=large)
Prompt:
Create a "Where is [Your Name]" crowd search poster set in the city of your choosing
Comment: The author mentions the full prompt is in the comments, but the core reusable prompt pattern is given in the post text. Source: @icreatelife
Prompt:
Create an eight-panel manga about GPT-Image-2 launching today
Source: @venturetwins
Prompts for creating high-quality typography, posters, and commercial designs with Chinese text. Source: 卡尔的AI沃茨
Prompt:
Generate a 1980s propaganda poster. Use the exact slogan "热烈庆祝GPT-Image-2全量开放". Include Sam Altman, Dario Amodei, and Elon Musk, and give Dario Amodei a red scarf.
Source: @akokoi1
Prompt:
生成鹿鼎记海报，展现韦小宝跟老婆XXX，忠于原著的描述，夸大特点，强调女性的美艳和男性的气质
English Translation: Generate a poster for The Deer and the Cauldron, showing Wei Xiaobao with wife XXX. Stay faithful to the original description, exaggerate the distinguishing traits, and emphasize the woman's beauty and the man's temperament. Source: @caiziboshi
| Poster 1 | Poster 2 | Poster 3 | Poster 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
XXX 代言 XXX 汽车海报
English Translation: XXX endorses XXX car poster. Source: @johnAGI168
Prompt:
Design a 3:4 vertical poster for a new Chinese trendy tea launch. Use a New Chinese visual style that feels light-luxury and restrained. The palette should be dark green, off-white, and gold, with rice-paper texture, elegant negative space, landscape accents, and modern layout design.
Main subject:
a visually appealing cold-brew tea with tea leaves, citrus, ice cubes, and touches of gold foil.
The poster must accurately display the following exact Chinese copy:
"山川茶事"
"山柚观音"
"冷泡系列"
"新品上市"
"一口清醒，半城入夏"
"限定尝鲜价"
"中杯 16 元"
"大杯 19 元"
"门店活动"
"第二杯半价"
"加 3 元升级轻乳版"
"每日前 100 名赠限定杯套"
"推荐风味"
"观音茶底 / 西柚果香 / 轻乳云顶 / 冰感回甘"
"活动时间 4月20日 至 5月10日"
"扫码点单"
"SHANCHUAN TEA"
Fine print:
"图片仅供参考，请以门店实际售卖为准"
Maintain a clear promotional hierarchy while keeping the overall feeling sophisticated rather than cheap or overly e-commerce-like. Pay special attention to small text, numbers, prices, info modules, and Chinese typography aesthetics.
Prompt:
Create a Chinese infographic poster themed "一杯咖啡 如何来到你手里" ("How a Cup of Coffee Reaches You"). Use a premium information-design style with both educational clarity and commercial appeal. The layout should include directional flow, arrows, data boxes, icons, simple illustrations, and modular cards. Use coffee brown, milk white, ink black, and copper accents.
The infographic must include:
- 01 Planting: elevation 1200-2200m, suitable temperature 18-24C, picking season Nov-March
- 02 Processing: sun-dried, washed, honey process
- 03 Roasting: light roast = brighter, medium roast = balanced, dark roast = richer
- 04 Grinding: pour-over = coarse, espresso = fine, cold brew = medium-coarse
- 05 Extraction: powder-water ratio, water temperature, and time all affect flavor
- flavor keywords: floral / citrus / nutty / caramel / chocolate / smoky
Use this exact fine print:
"适合用于咖啡入门科普与门店展示"
The composition should balance text and visuals while keeping the design elegant. Focus on long infographic handling, numeric information, temperatures, numbered sections, concise descriptions, slash-separated flavor words, and modular layout. It should look like a premium display board, not a classroom slide deck.
Prompt:
Please generate a key character relationship diagram for "XXX".
Source: OpenNana | @yihui_indie
Prompt:
Generate a detailed explanatory diagram of Jingdezhen blue and white porcelain, accompanied by detailed Chinese knowledge analysis.
Prompt:
Please automatically generate a "museum catalog-style Chinese disassembly infographic" based on the [Subject].
The entire image is required to combine a realistic main visual, structural disassembly, Chinese annotations, material descriptions, pattern meanings, color meanings, and core feature summaries. You need to automatically determine the most appropriate main subject, clothing system, artifact structure, era style, key components, material craftsmanship, color scheme, and layout structure based on the [Subject], and the user does not need to provide any other information.
The overall style should be: national museum exhibition boards, historical clothing catalogs, and cultural/museum thematic infographics, rather than ordinary posters, ancient-style portraits, e-commerce detail pages, or anime illustrations. The background uses paper textures such as off-white, silk white, and light tea color, making the overall look premium, restrained, professional, and collectible.
The layout is fixed as:
- Top: Chinese main title + subtitle + introduction
- Left: Structural disassembly area, with Chinese lead lines annotating key components, accompanied by close-up details
- Upper right: Material / craftsmanship / texture area, displaying real texture samples with descriptions
- Middle right: Pattern / color / meaning area, displaying the main color palette, pattern samples, and cultural explanations
- Bottom: Dressing order / composition flowchart + core feature summary
If the subject is suitable for character display, use a full-body standing posture of a real person as the central subject; if it is more suitable for artifacts or single structures, change it to a central subject disassembly diagram, but the overall form remains a complete Chinese infographic. All text must be in Simplified Chinese, clear, neat, and readable, without garbled characters, typos, English, or pinyin. The focus is on highlighting real structures, material differences, cultural explanations, and a catalog atmosphere.
Avoid: poster feel, studio portrait feel, e-commerce feel, anime feel, cosplay feel, random annotations, incorrect structures, blurry text, fake materials, excessive decoration.
Prompt:
Create a high-end skincare e-commerce hero poster for "澄光维稳精华" ("Clarifying Stabilizing Essence"). The style should be clean, light-luxury, and strongly science-skincare oriented. In the center, place a semi-transparent frosted-glass essence bottle filled with golden liquid and subtle water-droplet reflections. Use an off-white to warm gray gradient background with liquid-flow elements and microscopic molecular-structure decoration.
The poster must include the following exact Chinese copy:
"澄光"
"维稳精华"
"修护屏障"
"舒缓泛红"
"细腻透亮"
"第 2 代升级配方"
"核心成分"
"神经酰胺"
"泛醇 B5"
"积雪草提取物"
"微囊脂质体"
"适合人群"
"敏感肌"
"熬夜肌"
"换季不稳定肌"
"限时到手价 229 元"
"买 1 送 3"
"赠洁面 15ml"
"赠精华 5ml"
"赠面霜 10g"
Fine print:
"实际效果因人而异，请坚持使用"
Focus on product selling points, pricing hierarchy, gift-list modules, product naming, and short functional phrases. The result should feel premium and not tacky or overly live-commerce styled.
| Prompt 1 | Prompt 2 | Prompt 3 | Prompt 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a series of gongbi-style bookmark designs
Design a postcard collaboration between The Little Prince and SpaceX
Generate a single-day calendar for April 19, 2026 themed around The Garden of Words
Generate a series of Labor Day hand-held sign designs
Generate a poster for the 2026 Grain Rain solar term
Generate a set of icon fonts for a sports app
Design a film-photography-style poster with the theme "Toward the mountains and sea"
Design a Shanghai postcard in black line-art style
Design a set of seal carvings for "Elon Musk"
Design a set of co-branded promotional materials for Durex and Chayan Yuese
Source: @akokoi1
| Example 1 | Example 2 | Example 3 | Example 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate an image of the authentic manuscript of [classic text title]
Generate an image of the authentic manuscript of [classic text title], and incorporate the emotional core of the work into the calligraphy
Comment: Adding style cues such as in Wang Xizhi style, in Yan Zhenqing style, or with stronger emotional expression improves the result.
Source: Article | Basic Thread | Emotion Variant
| Wang Xizhi Style | Cursive Script | Running Script | Regular Script |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a calligraphy copybook practice sheet in [script style]
Comment: If you have a reference image for the exact calligraphy style, upload it as an image reference for stronger fidelity. Source: Article | @MrLarus
| Poster 1 | Poster 2 | Poster 3 | Poster 4 |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Automatically generate a high-aesthetic "Silhouette Universe / Collector's Edition Narrative Poster" based on [theme: xxx]. Do not default to common containers such as bottles, hourglasses, glass domes, or pocket watches. Instead, let the AI choose the most symbolic, visually strong, and narratively suitable outer silhouette for the theme. The silhouette can be an artifact, building, gate, tower, archway, dome, stairwell, corridor, statue, profile, eye, hand, skull, wing, mask, mirror, throne, ring, crack, light curtain, shadow, geometric form, spatial cross-section, stage frame, abstract symbol, or any more creative contour that best represents the theme.
The goal is not to simply place a world inside an object, but to make the entire themed universe grow naturally within, around, and through the silhouette itself. The silhouette should be elegant, recognizable, and compositionally dominant. Inside or along its boundary, generate a rich and layered narrative world strongly tied to the theme, including iconic scenes, key architecture or spaces, symbols and metaphors, traces of characters or civilizations, foreground-midground-background depth, and atmosphere with emotional tension. Elements such as doors, staircases, bridges, water, smoke, paths, light sources, ruins, mechanical structures, landscapes, abstract forms, creatures, or props should all feel unified and organically integrated rather than pasted together.
The final composition should feel like a premium collector's poster with strong design value. The large shapes should feel stable, the main silhouette should be unmistakable, and the inner world should have depth, structure, and breathing room. Details should be rich but not crowded. You may add small human silhouettes, distant buildings, beams of light, doorways, bridges, steps, corridors, reflections, skylight, or far-background structures to enhance scale, story, and epic atmosphere. The overall mood should be quiet, grand, refined, and lingering rather than noisy or overloaded.
Blend the feeling of a collector's edition film poster, high-end narrative visual design, dreamy watercolor texture, and fine printed paper. Emphasize paper grain, feathered edges, watercolor brush marks, gentle diffusion, atmospheric perspective, soft haze, selective volumetric light, light passing through mist, generous negative space, and restrained layout. The image should feel premium, poetic, majestic, sacred, nostalgic, quiet, and mythic.
Let the AI choose an elevated color palette based on the theme, but keep it unified, restrained, tasteful, low-saturation, and premium. Avoid chaotic high saturation, cheap neon effects, or plastic digital color. Suitable palette families may include black-gold-gray, cool blue-gray, mist white-gray, brown-red with off-white, dark copper, aged paper tones, deep sea blue, twilight purple, or silver-gray, as long as they serve the theme.
Final requirement: the first glance should give strong theme recognition and a memorable silhouette, the second glance should reveal a complete narrative world, and the third glance should still reward close inspection with depth and aftertaste. Avoid generic backgrounds, hard cut-and-paste compositions, templated fantasy assets, video-game promo art vibes, excessive cartooning, or realism that kills the artistic mood. If appropriate, subtle titles, numbering, signatures, or marks can be added as part of the poster design, but they must never overpower the image.
Source: Article | Thread 1 | Thread 2
| Protein Powder | AI Glasses | Summer Dress | Coffee Machine |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate an e-commerce product detail page for [product name]
| Campus Daily | Restaurant Menu | Textbook Page | Almanac |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate an image of [scene / content]
Comment: This is mainly a stress test for dense Chinese text rendering. For production use, provide the exact content or upload a source image such as a real menu or document photo. Source: Article | @MrLarus
| Traditional Chinese Medicine | Western Medicine |
|---|---|
 |
 |
Prompt:
Generate an image of a handwritten Chinese medicine or Western medicine prescription
Comment: Reference images for the prescription layout, handwriting, or stamp style can produce much better results. Source: Article | @MrLarus
Prompts for structured knowledge cards, educational materials, training plans, travel guides, and other information-dense Chinese layouts.
Prompt:
It's 10 AM in Los Angeles, 11 AM in Denver, 12 PM in Chicago, 1 PM in New York, 6 PM in London, and 2 AM in Tokyo. Render a wall with different analog clocks, each showing the correct time for its city, with the city label below each clock.
Source: @Angaisb_
| Panda | Peacock | Historical Figure | Science Topic |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a high-quality vertical encyclopedia-style infographic for [topic].
This should not be a normal poster or a simple illustration. It should feel like a modular educational infographic that combines the clarity of a field guide, the structure of an encyclopedia page, the polish of a lifestyle knowledge card, and the shareability of a strong social-media explainer.
The image should include:
- a clear and appealing main visual of the topic
- several enlarged detail callouts
- multiple rounded modular information sections
- strong title hierarchy and highlighted key labels
- concise but information-rich educational content
- visual scoring, quick takeaways, or a Top 5 module
Adapt the content sections automatically based on the topic. Useful categories include:
basic profile, classification, appearance, habits or ecology, formation mechanism or structure, growth or usage conditions, care or maintenance advice, risks and cautions, suitable users or use cases, pros and cons, and a quick scorecard.
Visual requirements:
use a clean light background, soft colors, subtle shadows, refined small icons, rounded information cards, and neat layout. The information density should be high but not crowded, and the final image should feel publishable, collectible, and repeatable as a knowledge-card format rather than an advertisement.
Do not make it look like a commercial promo poster. Emphasize knowledge organization, modular information, and a field-guide presentation.
Prompt:
Create an image with a 10x10 grid of objects whose names all start with the letter A
Source: @umesh_ai
Prompt:
Create a scene with 100 elements, and list the 100 elements within the image.
Comment: Reconstructed from the tweet text: "I asked ChatGPT Image v2 for 100 elements in a scene. Not only did it create a scene with 100 items, but it also listed them within the image!" Source: @umesh_ai
Prompt:
Create a 10 × 10 grid of 100 different topics representing recent technological progress. Use a realistic, polished editorial illustration style. Each topic should appear in its own square with a short clear label underneath. Keep the grid neat on a white background. Make every topic visually different and every label correctly spelled. Use these row themes: Row 1: AI models and agents Row 2: robotics Row 3: semiconductors and compute Row 4: networks and smart devices Row 5: biotech and health technology Row 6: energy and power systems Row 7: transport and autonomy Row 8: space and aerospace Row 9: manufacturing and materials Row 10: climate and environmental technology. Show each tile as a realistic mini-scene, product-class object, lab instrument, robot, chip, vehicle, or device that clearly conveys the topic. Keep the overall style consistent, modern, realistic, and visually impressive.
Source: @chetaslua
| Chest | Abs | Pull-Up | Dips |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a Chinese fitness infographic for [topic].
The graphic should be both professional and practical for a normal adult to use as a training reference. Unless otherwise specified, assume the audience is a healthy adult with no major injuries, the goal is muscle gain plus basic strength improvement, the level is beginner to intermediate, the setting is a normal gym, and the total session length should stay within 40 to 60 minutes.
Choose the output format automatically based on the training topic:
1. If the topic is a muscle group or body part, such as chest, lats, biceps, abs, shoulders, or legs, generate a training-plan infographic for that body part.
2. If the topic is an exercise or skill goal, such as pull-up, push-up, dips, or squat, generate an exercise-unlock or progression-plan infographic.
Use a clear, modern, professional, easy-to-read Chinese infographic style in vertical format. The visual design should be clean and suitable for social sharing or personal training reference. Do not write long paragraphs. Each module should use short concise phrases, and numerical information should be prominent.
The infographic must include:
A. Title area with main title and subtitle
B. Training goal area
C. Warm-up area
D. Main training area with 4 to 6 core exercises
E. Progression or unlock logic area
F. Alternative exercises area
G. Execution reminders area
H. Recovery suggestions area
I. Visual design requirements: modern, clean, professional, modular card layout
The final result should be a complete infographic rather than plain text paragraphs.
| Chinese | Math | English | History |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate an image of a 2026 Gaokao exam paper for [subject name]
| City 1 | City 2 | City 3 |
|---|---|---|
 |
 |
 |
Prompt:
Generate a three-day travel guide image for [city]
| Color Analysis 1 | Color Analysis 2 |
|---|---|
 |
 |
Prompt:
Using this portrait, create a diagram-first personal color analysis. Show which clothing colors suit the subject through visual comparison. Keep text minimal and avoid paragraphs.
Source: @AngryTomtweets | @LinusEkenstam
Prompt:
Create a personal color analysis using this picture. Use polo clothing color comparisons to highlight which colors suit the subject best. Make it visual-first, with explanations and tips.
Comment: The original post says the prompt is in a reply; this prompt is reconstructed from the visible output and tweet context. Source: @garenaguinaldox
Prompt:
Create a hairstyle analysis graphic using this picture. Show side-by-side hairstyle comparisons to highlight which hairstyles suit the subject best. Make it visual-first, with short labels only and details about the haircut.
Comment: The original post says the prompt is in a reply; this prompt is reconstructed from the visible output and tweet context. Source: @garenaguinaldox
Prompt:
A picture says a thousand words. GPT Image 2 creates them.
Comment: The tweet text is the only visible prompt-like instruction; the result is a chalkboard-style long-form document visualization. Source: @gabrielchua
| Demon Slayer | Naruto | Bleach | Dragon Ball |
|---|---|---|---|
 |
 |
 |
 |
Prompt:
Generate a high-design character relationship poster for [theme]. This should not be a normal illustration, but a relationship map that combines information visualization, narrative structure, poster-level design, and strong fidelity to the style of the original work.
Automatically do the following:
- identify the source work and its core setting
- select the 6 to 12 most representative key characters, with an upper limit of 15 if needed
- identify and show the most important relationships, such as family ties, romance, friendship, alliances, hostility, mentor-student links, master-servant links, manipulation, betrayal, and hidden relationships
- choose the most suitable composition automatically, such as protagonist-centered, dual-core confrontation, faction layout, family tree, or time-evolution structure
- extract the work's style DNA, including color palette, worldbuilding symbols, texture language, emotional tone, layout rhythm, and iconic motifs
- convert those style elements into the overall visual design of the relationship map rather than simply copying an official poster
- use different colors, line styles, and arrows to distinguish relationship types while keeping the layout clear and readable
- emphasize the core characters most strongly, secondary characters next, and minor characters less prominently to create a clear visual hierarchy
- keep every character name easy to read, optionally with role or faction tags if useful
The final piece should:
- make the main relationships understandable at a glance
- strongly evoke the original work's identity and setting
- combine infographic clarity with premium poster design
- feel unified, polished, complete, and suitable for social sharing or poster display
- avoid cheap flowchart vibes, clutter, or information overload
Prompts testing character consistency across multiple images and style replication.
| Nano Banana 2 | GPT-Image |
|---|---|
 |
 |
Prompt:
Create a 16-panel expression grid of a silver-haired, blue-eyed anime girl. Her face shape, hairstyle, and clothing must remain highly consistent across all panels. The 16 expressions should include: happy, sad, angry, surprised, shy, speechless, evil grin, contemplative, curious, proud, wronged, disdainful, confused, scared, crying, and a heart expression.
Source: 卡尔的AI沃茨
Prompt:
Based on this character and background, please create a character reference sheet similar to official setting materials.
- Includes three-view drawings: front view, side view, and back view
- Add variations of the character's facial expressions
- Break down and display detailed parts of the clothing and equipment
- Add a color palette
- Include a brief explanation of the worldview setting
- Overall, use an organized layout (white background, illustration style)
Source: OpenNana | @MANISH1027512
Prompts for image editing, style transfer, and reference-based generation. Source: 卡尔的AI沃茨
| Before | After |
|---|---|
 |
 |
Prompt:
Enhance this iPhone photo with ChatGPT so it looks like a professional photographer and designer worked on it.
Comment: Reconstructed from the post text, which compares an iPhone photo with a ChatGPT-enhanced result. Source: @dezainaz_ceo
| Reference | GPT-Image | Nano Banana 2 |
|---|---|---|
 |
 |
 |
Prompt:
Theme: \"77 (cat's name) X KFC\" collaboration poster. Generate a collaboration poster featuring the same pet (absolutely consistent in appearance and coloring) with KFC brand identity (red-and-white color scheme, classic logo, restaurant scenes, etc.). Have the cat wear a KFC employee uniform, KFC employee hat, and name badge while standing at the counter, selling fried chicken, burgers, and combo meals, and interacting with chicken buckets, fries, soda, and other brand elements. The style should be lively and fun with a commercial co-branding feel, suitable for online promotion and event posters. Add appropriate Chinese copy freely for the poster.
| Original | Nano Banana 2 | GPT-Image |
|---|---|---|
 |
 |
 |
Prompt:
Colorize this comic page and translate it into Chinese, placing the text in the original positions while maintaining composition and image details consistently
- Share your prompts and results on X with #GPTImage2
- Join discussions about prompt engineering techniques
Contributions are welcome! Please feel free to submit a Pull Request.
To contribute:
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Contribution guidelines:
- Include the full prompt text
- Provide the X/Twitter source link
- Add example images when possible
- Categorize appropriately
Special thanks to the Chinese AI community for contributing these comprehensive test cases. The new prompts added in the Typography, Photorealism, UI/UX, Consistency, and Image Editing sections are from:
卡尔的AI沃茨 (Carl's AI Watts) - 五大真实场景横测GPT-image-2和Nano Banana2
These real-world test scenarios provide valuable insights into GPT Image 2's capabilities across typography, realistic scene generation, UI recreation, character consistency, and image editing tasks.
All the content are from the internet if you think that this violates your copyright, please create an issue and we will take them down.
This project is licensed under the Creative Commons Attribution 4.0 International License - see the LICENSE file for details.
This repository is community-maintained and not affiliated with OpenAI.