Picva
搜索、筛选并复用适配多种 AI 图片模型的高质量提示词
搜索: Nano Banana Pro
Nano Banana Pro Prompt Felt toys Full body [Subject] toy, [attributes/accessories], [expression], made of felt, in a [place], [lighting], friendly and cartoonish appearance, rich and soft textures https://t.co/u6w49Q3K6O
GPT Image 1.5 vs Midjourney. MJ's aesthetics are still superior. prompt: "a cross-stitch of a christmas elf - anime style. the elf is working at a guitar store, guitars hang on the wall. the cross stitch has a christmas border with mistletoe and christmas decorations."
Geminiへ行ったらNano Banana Proお試ししてというからやってみた。プロンプトで「女性の背景に描いている月を向かって左に移動させてください。絵のタッチを透明水彩に変えて」と言っただけなんだけど（過去の苦労はなんだったんだと思うぐらいの仕上がり） https://t.co/I6WpiXWt57
مثال آخر على قدرة nano banana pro باش يفهم السياق فهم مبوق شنو هية Prompt : I want you to create a visual story cartoon for this joke واحد مبوق شاف واحد البنت رقيقة بزاف سولها: فاش خدامة؟ قالت ليه: عارضة قال ليها: ف إينا ملعب؟ https://t.co/sbqD7bYmGv
Visualize any selected emoji [] as an ultra-detailed, hyper-realistic 3D sculpture fully composed of luxurious enamel mosaic tiles. The emoji should retain its iconic silhouette and proportions, reinterpreted as a stylized 3D figure made entirely from curved, faceted, and geometrically interlocked enamel tiles in a radiant mosaic pattern.\n> Use high-gloss enamel tiles in varied shades derived from the emoji’s symbolic palette—integrating metallic accents, opalescent glazes, deep ceramic pigment
Fun #NanoBananaPro exercise. Consistent character with minor edits to prompt. The prompt is "change the background and the image of the <your object> to match his surroundings, the surroundings is <your surroundings> https://t.co/Krjj3dYxTU
请生成一张儿童识字小报《游乐园》，竖版 A4，学习小报版式，适合 5–9 岁孩子 认字与看图识物。 一、小报标题区（顶部） 顶部居中大标题：《游乐园识字小报》 风格：十字小报 / 儿童学习报感 文本要求：大字、醒目、卡通手写体、彩色描边 装饰：周围添加与 游乐园 相关的贴纸风装饰，颜色鲜艳 二、小报主体（中间主画面） 画面中心是一幅 卡通插画风的「游乐园」场景： 整体气氛：明亮、温暖、积极 构图：物体边界清晰，方便对应文字，不要过于拥挤。 场景分区与核心内容 核心区域 A（主要对象）：表现 游乐园 的核心活动（孩子们在玩游乐设施）。 核心区域 B（配套设施）：展示相关的工具或物品（售票、零食、指示设施）。 核心区域 C（环境背景）：体现环境特征（入口、路牌、彩旗、绿地等）。 主题人物 角色：1 位可爱卡通人物（身份：游乐园工作人员/游客小朋友皆可）。 动作：正在进行与场景相关的自然互动（如微笑指路、挥手欢迎、陪孩子玩）。 三、必画物体与识字清单（Generated Content） 请务必在画面中清晰绘制以下物体，并为其预留贴标签的位置： 1. 核心角色与设施： gōng zuò ré
肝了几小时，调出来一套适合2～5岁宝宝的：认知大发现图片：3D黏土风。8h后公布元提示词（因为还需要微调以确保质量），欢迎先点赞收藏，评论区留下你想要的风格和要求 评论先留一个特定场景的提示词给大家先玩 https://t.co/QydOPJiNQY [SCENE_THEME] = {动物世界} [TARGET_AGE] = 2-5 岁 (词汇爆发期，需要丰富且清晰的指引) 绘图提示词 (Optimized Prompt Structure) 【全局设定】 一张极具视觉冲击力、内容丰富的儿童认知横幅全景长图 (Wide Panoramic Banner)，主题为《[SCENE_THEME]》，竖版 A4。整体采用柔和、圆润的黏土塑形 3D 风格 (Soft Claymation/Plush 3D Sculpture)，就像一个巨大的、精心布置的实体玩具沙盘。光线明亮、温暖、纯净，带有细腻的体积光，营造出无与伦比的治愈感和触摸感。 一、小报标题区（顶部横幅） 大标题：在画面顶部横向展开标题：《[SCENE_THEME] 双语认知大发现》。 字体风格：超大、圆滚滚的彩色黏土气球字体，边缘带
分享元提示词及使用技巧： 很多人询问使用方式，这里做个说明： 1.文生图使用: nano banana pro 生成方式用两种： a. 使用元提示词+ 场景名，生成对应场景的提示词，再喂给nano 🍌 pro b.直接在元提示词写入具体场景使用nano 🍌 pro生图 试了这两种，资源模型不降智差别不大 大家关系的标签标错的问题，这次优化了下，但依然有可能有一两个标错的，有个小技巧：就是让模型把错误标签去掉，然后再让它打上去效果可能更好 备注：调整了图标文字展示顺序可自行调整，包括尺寸。附一张横幅的图片 元提示词如下： [SCENE_THEME] = {在此处填入场景，例如：繁忙的恐龙世界 / 热闹的超级市场 / 广阔的太空站} [TARGET_AGE] = 2-5 岁 (词汇爆发期，需要丰富且清晰的指引) 绘图提示词 (Optimized Prompt Structure) 【全局设定】 一张极具视觉冲击力、内容丰富的儿童认知，主题为《[SCENE_THEME]》，竖版 A4。整体采用柔和、圆润的黏土塑形 3D 风格 (Soft Claymation/Plush 3D Sculp
受宝玉启发做了个，利用四画面来演绎诗歌绝句：分享给大家。 提示词： Create a captivating poetic visual composition based on this prompt, seamlessly integrating the four lines of the quatrain into a unified masterpiece. Each panel perfectly illustrates one line of the poem, with the corresponding verse and the poem’s title and author elegantly embedded within its respective scene using Chinese characters. Together, these panels form a harmonious and cohesive panoramic artwork. Rendered with rich detail, cinematic lighting, 8K res
哈哈哈，让banana帮宝宝生成一个3*3的人物风格形象照片，我就想知道右二是什么情况😂 提示词（先上传一张照片）： A 3x3 grid of nine distinct artistic styles: Top row: pixel art style; watercolor illustration on textured paper; flat vector illustration with clean lines. Middle row: 3D render of a cute toy figure; 3D render of a futuristic robot with metallic and mechanical textures; traditional Japanese Ukiyo-e woodblock print. Bottom row: vibrant neon sign light art; medieval illuminated manuscript with gold leaf and floral borders; crocheted amig
新Ideal来了～不会做饭的人有口福了，周末搞起来！（元提示词见评论） https://t.co/nQ3Ke32KF7 请创建一个温暖的手绘食谱风格（16:9）插图，标题为"{菜名} 烹饪流程图"。 整体应呈现厨房手账风格，使用温暖的奶油色背景（如浅米黄#FFF8E7）， 搭配食物系配色：番茄红、胡萝卜橙、生菜绿、奶油黄等，营造温馨的家庭厨房氛围。 --- I. 主场景：烹饪步骤流程线 在中心位置绘制一条"波浪形/Z字形的烹饪流程线"，用箭头和虚线连接各个步骤。 根据用户输入的{菜名}和{难度等级}，自动生成烹饪步骤： 示例结构（自动填充{菜名}相关内容）： - "步骤1：准备食材 {食材清单 + 预处理方法}" - "步骤2：热锅预热 {油温/火候提示}" - "步骤3：{关键烹饪动作 + 时间}" - "步骤4：{调味环节 + 配料比例}" - ... - "最终步骤：装盘出锅 {摆盘技巧 + 享用温馨语}" 规则： - 如未提供难度等级，默认生成5-6个基础步骤 - 简单菜3-4步，中等菜5-7步，复杂菜8-10步 --- II. 周围装饰元素（根据菜系自动适配） 在流程线周围添
公布下元提示词： 使用方法：仅需在最后输入诗名就好 分享个小技巧：这个迭代了很久，创作前期发现有时候诗句顺序错乱，标题位置不稳，作者容易标错等。于是添加了workflow和反馈机制，可以实现稳定的输出。感觉可以把唐诗300首都刷出来了 晚点公布是因为先拿去起号了，发现之前有网友手速比我快太多，导致我来不及整理就被抢先了😂 元提示词如下： # Role: Expert Classical Chinese Poetry Illustrator & Art Director # Goal: Translate a user-provided Chinese Poem Title into a highly detailed, structured Midjourney V6 prompt. The output must interpret the poem's meaning into a continuous 4-scene visual narrative with specific layout requirements. # Workflow: 1. **Analyze Inpu
Create a detailed technical illustration of a Cybernetic Samurai, exploded into components: katana with energy core, cybernetic limbs, neural processor, armor plating, and internal power systems. Each part is labeled with clean futuristic font. Use a graphite and crimson color scheme on a dark blueprint background. Add subtle particle glow and depth shadows. Studio render style. 创作一幅赛博武士的详细技术图解，将其拆解为核心部件：带能量核心的武士刀、机械义肢、神经处理器、装甲板以及内部动力系统。每个部件均采用简洁的未来风格字体标注。整体采用石墨色与深红色配色方案，搭配深色蓝图背景。添加微弱的粒子光晕效果与层次感
🎨 NANOBANANA PRO 🎨 Prompt : Create an exploded-view illustration of a dragon's anatomy, dissecting its wings, scales, fire-breathing glands, and skeletal structure into labeled components with connecting arrows. Each part is rendered in intricate detail, showing metallic iridescent scales, glowing ember veins, and crystalline bones. Set against a ancient parchment background with subtle smoke effects and warm torchlight shadows. Include educational annotations in elegant script font. Horizonta
🎨 WATERCOLOR LANDSCAPE 🎨 Prompt : [SUBJECT] rendered as an Impressionist Watercolor Landscape, with loose brushstrokes and color blending to capture the essence of light and atmosphere. Use [COLOR1] and [COLOR2] to create a sense of time and place. 1. A snowy village at dusk rendered as an Impressionist Watercolor Landscape, with loose brushstrokes and color blending to capture the essence of light and atmosphere. Use indigo blue and warm amber to create a sense of time and place --ar 3:2 --v
🎨 NANOBANANA PRO 🎨 Prompt : Design a hyper-detailed technical poster dissecting the anatomy of a SpaceX Falcon 9 rocket. Explode-view of components thrusters. Each labeled with precise sans-serif text and dashed connector lines. Use metallic silver and blue palette on a starry black background. Add realistic lighting, orbital glows, and particle effects for depth. Studio render style, high resolution, aerospace engineering illustration, isometric view with slight rotation, Blender render. Made
🎨 WATERCOLOR STEAMPUNK SPLASH 🎨 Prompt : A Watercolor Steampunk Splash illustration of [SUBJECT], where Victorian-era machinery and gadgets are depicted through fluid, translucent paint washes. Blend [COLOR1] and [COLOR2] to create a unique fusion of vintage and artistic styles. Check ALTS
🎨 ART NOUVEAU INSECT 🎨 Prompt : [SUBJECT] as an Art Nouveau Holographic Insect. Intricate, flowing lines and organic forms are rendered with iridescent holographic foil, shifting between [COLOR1] and [COLOR2] to create a mesmerizing, Art Nouveau-inspired creature Check ALTS https://t.co/w75IxxsXQF
🎨 CLASSIC PIN-UP ART🎨 Prompt : [SUBJECT] rendered in the style of Classic Pin-Up Art, with bold outlines and vibrant colors. Emphasize a playful and glamorous atmosphere with a color scheme focused on [COLOR1] and [COLOR2] Check ALTS
🍌 Nano Banana Pro 200% 활용하는 7가지 팁 Nano Banana Pro는 단순한 이미지 생성을 넘어, '디렉팅'이 가능한 전문가용 도구가 되었네요. 원하는 이미지를 의도대로 제대로 뽑아내기 위한 7가지 내용을 기억해보세요. 🔬 주인공(Subject)은 '현미경'으로 보듯 묘사하세요 단순히 "고양이"라고 하지 마세요. AI는 상상력이 너무 풍부해서 엉뚱한 고양이를 데려옵니다. - Bad: "커피 마시는 로봇" - Good: "푸른색 광학 눈을 가진 근엄한 표정의 로봇 바리스타"처럼 외형, 표정, 특징을 구체적으로 적어주세요. ↓ 📸 카메라 감독이 되어 앵글(Composition)을 지시하세요 사진 찍을 때처럼 구도와 카메라 세팅을 알려주세요. Nano Banana Pro는 조명과 심도 조절에 탁월합니다. - Tip: "로우 앵글(Low angle)", "광각 샷(Wide shot)", "심도 얕게(f/1.8, 아웃포커싱)", "골든 아워의 역광" 같은 전문 용어
🍌 나노 바나나 프로 완벽 가이드: 프로페셔널 에셋 제작을 위한 10가지 팁 Google AI Studio 아티클에는 정말 구체적인 이야기가 들어있네요! Nano Banana Pro 프롬프트 기준과 10가지 상황들.. 그리고 여기엔 Best Practices, Example Prompts, 샘플 이미지 등 다양한 정보가 모두 담겼어요. 정말 좋은 가이드예요. 🔑 0. 프롬프팅의 중요한 규칙 Nano Banana Pro는 단순히 키워드를 맞추는 모델이 아니라 "생각하는" 모델이예요. 의도, 물리법칙, 구도를 이해할 수 있죠. - 다시 뽑지 말고 편집하세요 이 모델은 대화식 편집을 매우 잘 이해합니다. 이미지가 80% 정도 마음에 들면 처음부터 새로 생성하지 말고, 단순히 필요한 구체적인 변경 사항만 요청하세요. - 자연스러운 언어와 완전한 문장 사용하기 인간 아티스트에게 브리핑하듯이 모델에게 말하세요. 문법을 지키고 서술적인 형용사를 사용하세요. - 구체적이고 서술적으로 쓰기 모호