ChatGPT Images 2.0 生图提示词日报Daily GPT Image 2 Prompt Intelligence Report

2026-09-10 · 自动生成
近21日去重 · 不重复推送
12今日新提示词 / New
12文生图 · 直接复制
0需上传图 / Edit
6今日资讯 / News

✦ 今日原创 · Hermes originals

每天各一条文生图 / 需上传图;可另附一条对今日热词的改进。今日手法自选,不套热搜原句
文生图 原创 · 文生图
今日构图自选,不复用热搜原句 · 直接出图,不必上传。
复制Night aerial of an abandoned railway cathedral during a silent power cut: only bioluminescent traces outline the streets. a spinning analog metronome glows at the vanishing point. 100mm macro, razor plane of focus. hard noon sun, graphic shadows. Long exposure look, no signage text. Atmosphere thick enough to see the beam, but keep the subject edge-sharp.
停电后的废弃铁路教堂夜航:只有生物荧光勾出街道。一只还在摆的机械节拍器在灭点处发光。100mm 微距,焦平面极薄。正午硬光,阴影像剪纸。长曝光感。不要招牌字。
💡 直接复制到 ChatGPT Images,不必上传图
需上传图 原创 · 需上传图
今日构图自选,不复用热搜原句 · 先上传一张人像或产品图。
复制Using the attached photo, rebuild the subject as a life-size stained-glass figure set into an abandoned railway cathedral. Late sun throws coloured caustics across compacted chalk. Keep face geometry, hair silhouette and outfit colours. 70mm IMAX still, huge negative space. bioluminescent fill, no white key. No extra people, no beauty filter, no text. If the upload is a selfie, rebuild the background completely; if it is a product, keep the label crisp.
用上传的照片,把主体重建成立在废弃铁路教堂里的真人尺寸彩绘玻璃。夕阳把彩色焦散打在压实的粉笔灰上。保留脸部结构、发型剪影和衣服颜色。70mm IMAX 静帧,大面积留白。生物荧光补光,没有白主光。不要路人、不要美颜、不要字。
💡 先上传一张图,再把这段作为编辑指令
文生图 改进今日热词 · 文生图
在今日检索到的最高分句子上加约束,不另起炉灶。
复制Create a third-person cyberpunk action game screenshot set in a neon-soaked European capital at night. The protagonist has glowing cybernetic implants and stands on rain-slick streets near a famous landmark while holograms, drones, and flying traffic crowd the skyline. Add a polished game HUD with health bar, ammo count, radar, stealth/energy meters, and mission overlays. Vivid cyan-magenta palette, wet reflections, cinematic intensity, 16:9 One practical light only, visible air, huge negative space, shot like an IMAX still. No watermark, no extra text, no stock-photo smile.
改进:只留一盏真实光源,空气可见,大面积留白,IMAX 静帧。不要水印和字。
💡 直接复制到 ChatGPT Images,不必上传图

🔥 适配你 · Tailored for Qi Liang

只在今日新提示词明显贴合项目时才改一版,不逐条改编。
为什么适配:今日新提示词偏海报/编辑排版,适合排课侠或作品集头图。
原始
复制Create a 1:1 promotional poster that feels like it was designed by a professional ad designer. Make it suitable for books, courses, events, or browser-game promotion. Use strong visual hierarchy, polished typography, bold but modern composition, clear callout text blocks, and premium campaign-level finishing.
🔧 优化版(排课侠 / 作品集封面)
复制Editorial poster blending a developer silhouette with floating UI cards and a class-schedule grid, deep navy and cyan, text 'qi-liang · 排课侠'. Inspired by: Create a 1:1 promotional poster that feels like it was designed by a professional ad designer. Make it suitable for books, courses, events, or browser-game prom
人物剪影 + 课表网格 + UI 卡片的编辑式海报,深蓝青色。
为什么适配:今日新提示词偏示意图/信息图,可贴绘生生物实验讲义。
原始
复制Create a professional character sheet based on this character and background, similar to an official setting guide. Include front, side, and back views, expression variations, exploded outfit or equipment details, a color palette, and a short worldbuilding note. Keep the whole composition organized on a clean white background with a diagram-like layout and high-resolution concept art quality.
🔧 优化版(绘生 · 教学配图)
复制Clean editorial diagram of a plant transpiration experiment, labeled parts, minimal white background, classroom-handout style. Inspired by: Create a professional character sheet based on this character and background, similar to an official setting guide. Include front, side, and back views, express
植物蒸腾实验示意图,白底标注,课堂讲义风。
今日资讯 · Fresh sources
GPT Image 2 Prompts | AI Image Examples & Templates 2026Seedance 2.5 is live on EvoLinkTry Seedance 2.5 GPT Image 2 prompt gallery hero image # GPT Image 2 Prompts Curated image prompt examples for posters, portraits, UI mockups, character sheets, and marketing visuals. Browse prompt examples with ready previews, then continue to the GitHub - wuyoscar/GPT-Image2-Skill: GPT Image 2/2.5 prompt gallery ...> This project is not trying to collect every prompt on the internet. We keep a selected set of examples that show what GPT Image 2 can do and how to use it well. Thanks for all the love this little gallery has received 🫶. Caution [...] | 🧩 Brand Systems & Identity Full Gallery MChatGPT Image Prompt: 20 GPT Image 2 Prompts to Copy (2026)## 20 ChatGPT Image Prompt Examples by Category These examples were selected for category diversity, practical reuse, visual quality, prompt structure. The multilingual examples are kept in their original wording to show how GPT Image 2 handles non-English prompt instructions. ##GPT Image 2 First Look (5 Examples to try)# GPT Image 2 First Look (5 Examples to try) ## Paul Solt 22500 subscribers 30 likes [...] 9 comments ### Transcript: [...] because the teeth really sold it for me. The teeth look exactly sort of like how her teeth look. She was an Australian Shepherd and so that's where we're goGPT Image Generation Models Prompting Guide - OpenAI Developersand logos to translation, style transfer, virtual try-on, compositing, and lighting changes. Throughout the examples, the cookbook reinforces the importance of clearly separating what should change from what must remain invariant, and of restating those invariants on every iteratGPT Image 2 Prompts | AI Prompt Gallery# GPT Image 2 Prompts Explore GPT Image 2 prompts for readable text, posters, UI mockups, infographics, product ads, portraits, editing, and social creatives. Nano Banana PromptsGPT Image 2 PromptsGrok Imagine Prompts Filter by tag ## Featured GPT Image 2 Prompts Start with flexi
分类分布 · 文生图 vs 需上传
文生图
12
需上传图
0
人物
6
艺术
5
实验
1
筛选 · Filter
全部 All文生图 · 直接复制需上传图人物艺术实验
#1
S 级 文生图 人物
推荐指数 89
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Create a third-person cyberpunk action game screenshot set in a neon-soaked European capital at night. The protagonist has glowing cybernetic implants and stands on rain-slick streets near a famous landmark while holograms, drones, and flying traffic crowd the skyline. Add a polished game HUD with health bar, ammo count, radar, stealth/energy meters, and mission overlays. Vivid cyan-magenta palette, wet reflections, cinematic intensity, 16:9.
英文原提示词(来自「GitHub - wuyoscar/GPT-Image2-Skill: GPT 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 github.com · 近21日未推送过
#2
S 级 文生图 人物
推荐指数 89
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate a cinematic minimal portrait of a solitary man standing in an intense orange to red gradient environment, strong silhouette lighting, deep shadow contrast, reflective glossy floor, symmetrical composition, minimal
英文原提示词(来自「ChatGPT Image Prompt: 20 GPT Image 2 Pro」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 oimi.ai · 近21日未推送过
#3
S 级 文生图 人物
推荐指数 89
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Create a complete visual worldbuilding set for a futuristic desert civilization powered by solar technology: multiple images including architecture, characters, clothing, vehicles, maps, symbols, and environmental details. Keep a cohesive design language, cinematic realism, warm desert light, ultra detailed.
英文原提示词(来自「ChatGPT Image Prompt: 20 GPT Image 2 Pro」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 oimi.ai · 近21日未推送过
#4
S 级 文生图 艺术
推荐指数 89
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Create a realistic outdoor crowd scene in Bethel, New York on August 16, 1969.Create a realistic outdoor crowd scene in Bethel, New York on August 16, 1969.Photorealistic, period-accurate clothing, staging, and environment.Photorealistic, period-accurate clothing, staging, and environment.
英文原提示词(来自「GPT Image Generation Models Prompting Gu」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 developers.openai.com · 近21日未推送过
#5
S 级 文生图 艺术
推荐指数 89
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Turn this drawing into a photorealistic image.Turn this drawing into a photorealistic image.Preserve the exact layout, proportions, and perspective.Preserve the exact layout, proportions, and perspective.Choose realistic materials and lighting consistent with the sketch intent.Choose realistic materials and lighting consistent with the sketch intent.Do not add new elements or text.Do not add new elements or text.
英文原提示词(来自「GPT Image Generation Models Prompting Gu」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 developers.openai.com · 近21日未推送过
#6
A 级 文生图 艺术
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Create a 1:1 promotional poster that feels like it was designed by a professional ad designer. Make it suitable for books, courses, events, or browser-game promotion. Use strong visual hierarchy, polished typography, bold but modern composition, clear callout text blocks, and premium campaign-level finishing.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#7
A 级 文生图 实验
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Create a professional character sheet based on this character and background, similar to an official setting guide. Include front, side, and back views, expression variations, exploded outfit or equipment details, a color palette, and a short worldbuilding note. Keep the whole composition organized on a clean white background with a diagram-like layout and high-resolution concept art quality.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#8
A 级 文生图 人物
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate a game battle poster featuring Mai Shiranui versus Diao Chan in an epic confrontation, with dramatic action poses, energy effects, and fighting game aesthetics. Combine Eastern and Western character design elements with high-impact visual composition, dynamic lighting, and tournament poster typography.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#9
A 级 文生图 人物
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate image: Selfie of Sam Altman riding a bear. The image should look like a casual selfie taken at arm's length, with Sam Altman in the foreground and a large bear beneath him, set in a natural outdoor environment. Maintain a humorous, lighthearted tone with realistic photographic quality.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#10
A 级 文生图 人物
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate a precise realistic in-game image of Among Us. Maintain the game's distinctive character design (bean-shaped astronauts, colored spacesuits, visors) but render everything with realistic textures, lighting, and materials. Include typical game elements like the spaceship interior, task stations, and the characteristic atmosphere of suspicion and teamwork.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#11
A 级 文生图 艺术
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate an image of the most significant event of 2020. Create a powerful, editorial-quality visualization that captures the defining moment, mood, and global impact of the year's most important event. The image should be emotionally resonant, historically accurate in its references, and composed with journalistic gravitas.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过
#12
A 级 文生图 艺术
推荐指数 83
🖼️ 暂无真实生成示例图(不挂站方配图)· 看来源
复制Generate an image of the most significant event of 2001. Create a powerful editorial visualization that captures the defining moment of that year with journalistic composition, dramatic atmosphere, and emotional weight that reflects the global impact.
英文原提示词(来自「GPT Image 2 Prompts | AI Image Examples 」)。直接复制到 ChatGPT Images,不必上传图。
💡 文生图 · 来源 evolink.ai · 近21日未推送过