Ai Image Prompts
产品介绍
从包含10,000+真实图像生成提示的精选库中推荐提示。兼容任何AI图像模型,包括Nano Banana Pro、Nano Banana 2、Seedream等。
资源信息
许可证信息摘自官方源,可能过时或缺失。商用前请以原仓库的 LICENSE 文件为准。
提示词库 · 2.3万
数据源:社区提示词库(MIT 许可,每日同步)共 15141 条提示词,按 11 个用途归类;同一条可归入多个用途,因此分类条数相加会大于去重后的总数。
命中 725 条
Hard Surface Modeling Retro Cassette Realism
game-assetA prompt for generating an extreme closeup, hard surface modeling final render. The style is retro cassette realism, applied to a futuristic device, emphasizing incredible detail, sharp focus, glass details, and surface imperfections.
展开提示词全文
Extreme closeup, hard surface modelling final render. In the style of retro cassette realism. A futuristic device, not a cassette tape. Incredible detail. Stunning design. Sharp focus and shallow depth of field. Glass details. Surface imperfections. 32k uhd
Stylized Miniature Diorama of Forest Waterfall Village
game-assetA detailed prompt for generating a premium stylized miniature diorama of a peaceful village built along a forest waterfall. The prompt specifies the scene elements (shrines, houses, villagers), the style (floating diorama, resin effects, handcrafted scale), and the camera angle (isometric, high angle, 4:3 aspect ratio).
展开提示词全文
SCENE: A peaceful village built along the edge of a tall forest waterfall with small shrines, wooden houses, rope bridges, and viewing platforms carved into the cliff. Tiny villagers carry baskets along stone paths while birds fly through the mist and moss grows on the rocks. STYLE: Create a premium stylized miniature diorama whose terrain base blends naturally with the environment so cliffs, forest floor, and flowing water form a continuous sculpted model. The entire diorama including the terrain base should float freely in mid-air without resting on any table or platform. Surround the diorama with a blank single-color environment that contrasts with the scene so the miniature stands out clearly. Use stylized resin waterfall effects, compact architecture, warm lantern glow, and dense foliage clusters. Include miniature villagers, birds, baskets, shrubs, and path props to reinforce handcrafted collectible scale. Lock the render to a 4:3 aspect ratio. CAMERA: Use an isometric camera angle positioned high enough to clearly view the terrain layers and structures. Frame the entire floating diorama so the full base and environment are visible from top to bottom and side to side with comfortable margins.
Ultra-Realistic 3D Cinematic Portrait with Tom and Jerry
game-assetA structured prompt designed to generate an ultra-realistic 3D cinematic full-body portrait of a smiling young boy posing with Tom and Jerry, emphasizing realistic fur textures for the animals while maintaining their original facial consistency, set against a soft blue studio background.
展开提示词全文
{
"prompt": "Ultra-realistic 3D cinematic full body portrait of a smiling young boy with Tom and Jerry. Boy wearing light grey sweater, blue jeans, white sneakers, relaxed crossed-leg pose. One arm around Tom. Tom standing upright with one hand on waist, other arm around the boy. Jerry sitting on Tom’s shoulder smiling. Tom and Jerry have realistic detailed fur but their faces remain consistent and unchanged. Age-appropriate boy height. Soft blue studio background, dramatic soft lighting, sharp focus, photorealistic quality. No cartoon look, no blur, no distortion.",
"style": "ultra-realistic 3D cinematic",
"camera": "full body portrait, sharp focus",
"lighting": "dramatic soft studio lighting",
"background": "soft blue studio background",
"quality": "photorealistic",
"constraints": [
"Keep Tom and Jerry faces consistent and unchanged",
"Age-appropriate boy height",
"No cartoon look",
"No blur",
"No distortion"
]
}
Anatomy Correction Prompt for Nano Banana 2
game-assetA simple negative prompt designed to stabilize human anatomy generation in Nano Banana 2, especially when dealing with complex poses like leaning, railing interaction, and full-body shots.
展开提示词全文
single person, correct anatomy, two legs, two arms
High-Resolution 2.5D Illustration Prompt for Winged Space Battleship
game-assetA detailed English prompt for generating a high-definition, 2.5D rendered illustration of a massive, futuristic space battleship with delta wings, flying in near-Earth orbit. The prompt specifies the ship's design, composition, armaments, and lighting details.
展开提示词全文
High-definition, high-resolution 2.5D rendered illustration. An image showing a 300-meter-long super-large space battleship flying in near earth orbit space. The camera is looking at the spacecraft diagonally from the side. The space battleship is a spaceplane type with delta wings and tail, and is composed of futuristic organic curves that look like something out of a science fiction movie. Above the space battleship are the battleship's main rotating gun turret on the central axis, bridge, phased array radar, and anti-aircraft missile launchers. The nose of the spacecraft has a radome and is equipped with a leading tail wing. The fuselage is of the scramjet type, with all engines located at the rear of the fuselage. The lower half of the fuselage is painted red like a large naval vessel. Red flashing lights glow on the tips of the space battleship's wings, on top of the bridge, and on the bottom of the hull.
Giant Woman City Walk Video Prompt
game-assetA prompt for generating an ultra-realistic, cinematic video of a giant young woman walking through a modern city, towering over skyscrapers, emphasizing dramatic scale contrast, low-angle perspective, and dynamic lighting.
展开提示词全文
A giant young woman walking through a modern city street, towering over skyscrapers, her footsteps cracking the asphalt beneath her boots. She wears a brown leather jacket over a beige turtleneck sweater, a fitted brown skirt, ankle socks, and lace-up brown boots. Her long wavy hair flows naturally as she walks confidently. Below her, tiny cars and yellow taxis swerve around broken pavement, pedestrians scatter in shock at a crosswalk. Dust and debris rise dramatically around her feet. Cinematic perspective from low angle, dramatic scale contrast, ultra-realistic details, dynamic lighting, shallow depth of field, 8K resolution, highly detailed city environment, action scene, photorealistic, epic atmosphere.
3D Cartoon Avatar Character Grid Prompt (Pixar Style)
game-assetA prompt for generating a 3x3 grid of the same male 3D cartoon avatar character, rendered in Pixar style, showing nine different emotions and expressions, with colorful gradient backgrounds.
展开提示词全文
3D cartoon avatar character showing different emotions arranged in a 3x3 grid, same male character with brown hair, Pixar-style animation, each square showing a different expression (happy, shocked, angry, caring with a puppy, confident, thumbs up, sad, thinking, professional), colorful gradient backgrounds for each panel, soft studio lighting, smooth 3D rendering, expressive eyes and facial expressions, modern emoji-style character design, clean composition, ultra-detailed, vibrant colors, 8K.
Lego Explorer Chased by Polar Bear (Video Prompt)
game-assetA prompt for generating a dynamic, action-packed video using Gemini Nano Banana Pro and Kling 3.0. It describes a Lego explorer minifigure driving a snowmobile, chased by a polar bear, emphasizing cinematic effects, motion, and the crisp texture of Lego plastic.
展开提示词全文
lego explorer minifigure driving a mini snowmobile across icy terrain chased by a massive roaring polar bear, snow spraying dramatically behind, frozen mist in the air, intense cinematic backlight, dynamic tilted camera shot, blockbuster action style, exaggerated comic motion, crisp lego plastic texture, dramatic cold color grading
Dominance portrait generation prompt (Gemini - Thinking mode)
game-assetAn extremely detailed, structured prompt for Nano Banana (Gemini - Thinking mode) designed to create a 'dominance portrait' of an uploaded reference character seated on a divine throne. The prompt emphasizes strict likeness integrity, engineering the throne from the character's design language, and specific cinematic camera/lighting techniques (wide lens, partial obstruction, vertical divine beam).
展开提示词全文
Create a dominance portrait of the uploaded reference character seated on a divine throne that feels engineered from their own world and design language. The camera observes the ruler from a partially obstructed viewpoint, as if the viewer is looking through the monumental structure of the throne itself. Likeness Integrity (HIGHEST PRIORITY) Maintain exact: • Head-to-body proportions • Eye shape, spacing, stylization • Surface materials • Silhouette and structural design Do not humanize. Do not reinterpret anatomy. Do not add unintended accessories. Perspective exaggeration must come only from camera placement, not structural redesign. Identity always wins. Throne Adaptation System (UNIVERSAL LOGIC) The throne must feel like a structural extension of the character’s design universe and must feel divine. Rules: Extract design language from the character: Dominant material logic Geometric language Color World energy Translate, never copy. Echo shapes subtly in silhouette Reflect material family, not exact textures Use tonal harmony, not color duplication Patterns may appear in architecture, structure, or ornamentation but must remain subordinate to the character’s identity. Camera & Perspective • Camera positioned slightly below the throne platform but offset laterally • 16–20mm wide cinematic lens • The throne structure partially obstructs the frame • Large architectural element of the throne occupies the foreground edge of the frame • Character visible through the throne’s structural opening or between its elements • Horizon slightly tilted 5–8 degrees • The viewer feels as if witnessing the ruler from behind the architecture of power Foreground elements must feel massive and close to the lens, framing the character like a monumental gateway. Composition • The character sits centered within the throne structure but partially framed by its geometry • Throne architecture creates a natural visual tunnel toward the character • Foreground throne elements remain slightly out of focus • Character remains the focal authority within the frame Lighting • Singular vertical divine beam descending from above the throne • Volumetric haze revealing light shafts • Strong rim light outlining the character silhouette • Controlled specular spikes along throne edges • Foreground throne elements remain darker, creating contrast against the illuminated ruler Depth & Focus • Eyes in absolute clarity • Character face and upper torso remain sharp • Foreground throne structure slightly blurred from proximity to lens • Background dissolves into atmospheric layers The image must feel layered in depth: foreground architecture → seated ruler → atmospheric background. Background • Vast negative space above the throne • Atmospheric tension • Light beam
Sherlock Holmes Ornate Cupboard Diorama Prompt
game-assetA complex, structured prompt template for Nano Banana 2 to generate a hyper-realistic render of an ornate cupboard diorama. The diorama must contain three shelves, each depicting a scene from 'Adventures of Sherlock Holmes'. The prompt uses variable guides for materials, style, lighting, and camera angle, and provides strict directives on composition and structure (must be open-front, no doors).
展开提示词全文
Variable Guide
1. OVERALL_DIORAMA_THEME = {argument name="diorama theme" default="Scenes from Adventures of Sherlock Holmes by Arthur Conan Doyle, in the order in which they are presented in the book"}
2. CUPBOARD_BASE_MATERIAL =
3. CUPBOARD_LINING_MATERIAL =
4. RACK_MATERIAL =
5. ORNAMENTATION_STYLE = [e.g., Gothic carved wood, Art Deco brass, Elven interlaced vines]
6. ART_STYLE = Hyper-Realistic Render
7. TOP_SHELF_SCENE = {argument name="top shelf scene" default="[Describe specific actions of miniature figures defining the scene]"}
8. MIDDLE_SHELF_SCENE = [Describe specific actions of miniature figures defining the scene]
9. BOTTOM_SHELF_SCENE = [Describe specific actions of miniature figures defining the scene]
10. LIGHTING =
11. CAMERA_ANGLE = A straight-on, front-facing view, captured from a moderately high angle looking distinctly downward.
12. ASPECT_RATIO = 3:4
Generation Directives
(Note: If any variable is left blank, intelligently infer the most appropriate detail based on the OVERALL_DIORAMA_THEME)
Directive 1: Framing & Composition (Highest Priority)
A direct, front-facing photograph captured from CAMERA_ANGLE. The composition must perfectly center the open display structure, looking down at a distinct angle into the shelves to clearly reveal the floor plates and top-down depth of the miniature objects within the dioramas, while ensuring the exterior side panels remain completely hidden. The entire free-standing structure must be completely visible with negative space around it against a simple background. The aspect ratio is 3:4.
Directive 2: The Main Structure (Strict Constraints)
Render the image in the style of ART_STYLE. The central object is a highly detailed, permanently open-front curio display structure. It must not possess doors, hinges, or door frames of any kind. The exterior is constructed from CUPBOARD_BASE_MATERIAL and detailed with ORNAMENTATION_STYLE. The interior back wall is lined with CUPBOARD_LINING_MATERIAL, and the individual shelves are made of RACK_MATERIAL.
Directive 3: The Miniature Dioramas & Figures
The structure houses a cohesive miniature world divided across three shelves, adhering to the OVERALL_DIORAMA_THEME. Crucially, each shelf must contain miniature figures engaged in a specific narrative action.
* Top Shelf: Miniature figures are depicted [TOP_SHELF_SCENE].
* Middle Shelf: Miniature figures are depicted [MIDDLE_SHELF_SCENE].
* Bottom Shelf: Miniature figures are depicted [BOTTOM_SHELF_SCENE].
Ensure the scale clearly reads as miniature dioramas housed within a furniture piece.
Directive 4: Atmosphere & Lighting
The scene is illuminated by LIGHTING. The lighting should cast appropriate micro-shadows within each shelf to emphasize the depth, the figures, and the miniature scale of the dioramas.
Lego Set Design Prompt Template
game-assetA prompt template for generating a Lego set design (including the box and instructions) based on a specific subject (place, movie, image reference) and set type (adventure, architecture). The example shows it was made using Adobe Firefly with Nano Banana 2.
展开提示词全文
Create a Lego set for {argument name="subject" default="[place, movie, image reference]"}, {argument name="set type" default="[type of set, adventure, architecture]"} set, include box and instructions
Ornate Box Diorama Prompt (Cyberpunk Pub)
game-assetA highly detailed, structured prompt template (AUTO-THEME MASTER PROMPT) for generating an ultra-realistic miniature diorama inside an ornate box, specifically themed as a Cyberpunk pub. The prompt dictates architectural integration, material style, lighting, and extreme macro detail.
展开提示词全文
Scene:
ORNATE BOX DIORAMA — AUTO-THEME MASTER PROMPT
========================
SMART VARIABLE GUIDE
THEME_LOCATION = {argument name="Theme Location" default="Cyberpunk pub"}
AUTO RULES:
• If THEME_LOCATION is filled → automatically infer:
PRIMARY_ACTIVITY
STAIR_TYPE
CEILING_DETAILS
THEME_MATERIAL_STYLE
PROP_LIST
THEME_LIGHT_SOURCE
BASE_MATERIAL_COLOR
THEME_PALETTE
FOCAL_SUBJECT
All inferred values must be logically consistent with THEME_LOCATION.
• If THEME_LOCATION is blank → first invent a visually rich miniature environment theme, assign it to THEME_LOCATION, then automatically infer all other variables from it.
• Always prioritize strong environmental storytelling, believable materials, and handcrafted miniature realism when inferring values.
========================
SUBJECT:
A two-level miniature THEME_LOCATION where the ornate open box itself forms the architecture, the base container acting as the full ground floor chamber with the box walls serving directly as the structural walls. The sides of the box must contain functional doors and windows, with the front-facing side featuring a wide open doorway and visible windows. No side of the box may be removed, cut away, or missing. The box height:width:length ratio is 1:4:7 (the width four times the height). The scene features PRIMARY_ACTIVITY actively interacting with the environment. The structure must be fully integrated into the box material itself — not a room inserted inside, not a display frame, not a cutaway model. The ground floor must be constructed from the same material as the box exterior.
ENVIRONMENT:
The base of the ornate box forms the ground floor surface and foundation. Its interior sides transform into the architectural walls containing carved doorways and window openings seamlessly cut into the box sides and aligned with exterior ornamentation. A mezzanine or upper level is built above with structural elements attached directly to the box walls and connected via STAIR_TYPE. The inside of the open lid functions as the vaulted ceiling decorated with CEILING_DETAILS.
DETAILS:
Intricately carved ornate box styled according to THEME_MATERIAL_STYLE. Exterior ornamentation must continue logically into interior architecture. Doors and windows must be precisely cut into the box sides following decorative carving patterns. Include believable miniature structural joins where beams meet the container walls. Include ultra-fine theme-specific props such as PROP_LIST. Include floating dust, micro-surface wear, tactile handcrafted realism, and environmental storytelling details.
STYLE:
Ultra-realistic cinematic miniature diorama photography, museum-grade handcrafted realism, extreme macro detail, physically believable materials, strong architectural integration of container and structure.
LIGHTING:
Primary lighting motivated by THEME_LIGHT_SOURCE combined with subtle environmental secondary lighting entering through box windows and openings. Volumetric dus
提示词模型无关,可直接用于 Midjourney / DALL·E / Flux / Stable Diffusion 等任意文生图模型。
信息边界说明
本页信息摘自官方文档,仅作技术信息呈现,不代表本站的判断或推荐。价格、配额、可用性以官方最新说明为准。你离开本站后的任何行为(申请 Key、付费、部署、商用),本站不做任何承诺与担保。
同类推荐
视频短视频图片音乐音频配音口播数字人全能创作
官方标注收费通过同一个AI创作工具完成AI图片、视频、音乐与语音创作,查找公开社交数据,编辑视觉成果,并集中查看和管理生成资产。Beatra 全能创作支持AI图片生成(文生图)、AI视频生成(图生视频)和AI视频编辑,通过AI音乐生成把想法做成歌曲或纯音乐,也可用AI语音生成(文字转语音)、多语言旁白和AI语音克隆完成跨媒介内容;还能查找抖音数据、小红书数据,以及 YouTube、TikTok、Instagram 与 X 上的公开帖子、笔记、评论、账号和趋势,把社交媒体数据用于竞品调研和创作。它是一款面向AI内容创作与多模态内容创作的AI创作套件,用起来就是一个把图片、视频、音乐与语音收在同一处的AI创作助手,适用于社交内容、产品展示、营销素材、课程、播客、短片和跨媒介项目;还能在不同媒介间复用原始素材、已有成果和查到的公开资料,查看制作进度、返回文件、实际用量和积分记录。
AI数字人口播与虚拟主播视频
官方标注收费用一张人像和一段短脚本或语音,制作有明确讲述内容的数字人口播视频。可使用现成录音,也可以先选择音色,把脚本生成旁白,让画面中的人物随声音完成稳定、自然的口播。适合 AI 主播、虚拟主播视频、数字人讲解、让照片说话、照片说话视频、产品讲解视频、数字人培训视频、公告和社交口播,并在交付后检查人物一致性、声音清晰度、口型同步和动作稳定性。
罗斯蒙德可视化图形生成
官方未说明【罗斯蒙德可视化图形生成】南京罗斯蒙德基于已交付项目沉淀的可视化图形生成场景:根据业务描述或系统需求,自动生成可编辑的架构图、流程图、ER 图、时序图、状态图与 C4 图。采用「统一图模型」一次建模、多格式输出:结构化文本(Mermaid / PlantUML,可嵌入 Markdown、进 Git 版本管理)、可编辑图形文件(drawio / Excalidraw,可拖拽修改节点与样式)、静态图(SVG 自渲染,PNG 给转换命令),并输出图形说明(图的含义 + 关键节点解释)。内置样式系统统一配色与命名,内置布局引擎自动排布,校验脚本拦截孤立节点、悬挂连线与语义缺漏。触发词:画架构图、画流程图、画 ER 图、画时序图、画状态图、C4 图、系统架构图、部署架构图、数据流程图、实体关系图、生成 drawio、生成 Excalidraw、导出 mermaid、导出 PlantUML、导出 SVG、导出 PNG、可编辑的图、图转可编辑、架构图可编辑、流程图二次修改、画图、出图、画个图。
架构图一键生成
官方标注收费需申请 Key使用 ContextWeave 生成或修改架构图、流程图、思维导图和复杂信息图。适用于需要从代码、文件或自然语言中提取结构与关系并实际产出 CW、SVG、HTML、可编辑 PPTX 或原生 VSDX 的请求;未特别说明时,PPTX 与 Visio 使用原生形状和连接器导出。不适用于统计图表、手绘插画或像素级排版。
AI Image Prompts · 绘画提示词库
官方标注免费Recommend curated prompts from a 10,000+ real-world image generation prompt library. Works with ANY AI image model — Nano Banana Pro, Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E 3, Flux, Stable Diffusion, and more. Use this skill when users want to: - Find proven image generation prompts (any model) - Get prompt inspiration for portraits, products, social media, posters, etc. - Create illustrations for articles, videos, podcasts, or marketing content - Browse categorized prompt templates with sample images - Translate and understand prompt techniques
架构图一键生成
官方标注收费需申请 Key使用 ContextWeave 生成或修改架构图、流程图、思维导图和复杂信息图。适用于需要从代码、文件或自然语言中提取结构与关系并实际产出 CW、SVG、HTML、可编辑 PPTX 或原生 VSDX 的请求;未特别说明时,PPTX 与 Visio 使用原生形状和连接器导出。不适用于统计图表、手绘插画或像素级排版。
什么是技能
技能(Skill)是封装好的单项能力,比如文字识别、文档摘要、关键词提取。开发者把调用逻辑和提示词打包,使用者像用插件一样加载即可,无需了解底层模型。启枢技能 收录大量技能,可直接试跑或编入方案。