← 返回首页
技能

Ai Image Prompts

tencent_skillhub · 1.0.9 · 下载量 1,827 · 热度分 55.6

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

作者:dophinl
分类:图文音视频多模态生成
版本:1.0.9
认证方式: 无需认证
是否付费: 官方未说明

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。

标签
图片生成

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

3D Stylized Caricature Portrait

game-asset

A complex prompt for turning a reference subject into a stylized 3D CGI caricature with specific anatomical exaggerations and professional studio rendering.

展开提示词全文
Convert the subject in this image into a stylized 3D caricature portrait with {argument name="head shape" default="an elongated rectangular head"}, {argument name="facial feature" default="oversized nose"}, and pristine studio lighting. Use a sculpted, high-end 3D CGI character treatment with strong identity preservation, while amplifying the facial geometry into a bold editorial caricature without changing the subject’s existing hair.

CHARACTER:
Build the subject as a tall, narrow caricature bust with a stretched rectangular skull, compressed cheek planes, and a dominant forward-projecting nose that anchors the face. Preserve the subject’s recognizable hair volume by translating it into clean, sculpted strands. Keep the clothing simplified, modern, and rendered as a refined secondary form that supports the face.

FACE & ANATOMY:
Use a severe angular facial construction with planar transitions, heavy brows, deep-set eyes, and a long tapered lower face. Keep the expression serious and slightly weary, with a rigid jaw rhythm and exaggerated vertical proportions. The body should remain secondary and understated, with the head as the primary silhouette.

ENVIRONMENT:
Change the background into a {argument name="background style" default="minimal 'dark gradient' studio"} that isolates the portrait and enhances its sculptural clarity. Keep the scene clean, controlled, and distraction-free, with no props or narrative clutter. Add subtle depth falloff so the character reads as a premium character-study render.

RENDER PIPELINE:
Use stylized 3D CGI with a polished sculpt-first workflow, crisp edge definition, and clean subdivision surfaces. Apply controlled toon-leaning shading with smooth form transitions, disciplined specular response, and no photoreal skin microdetail. Finish with a premium character-render presentation, sharp silhouette readability, and zero photographic camera artifacts.

COLOR & LIGHT:
Use a restrained palette of 'deep charcoal', 'cool blue' shadows, 'warm peach' skin tones, and 'soft amber' rim accents. Light the face with perfect studio key lighting, gentle fill control, and a balanced rim that carves the silhouette cleanly. Preserve high contrast in the facial planes while keeping the exposure elegant and even.

MATERIALS & TEXTURES:
Render skin as clean, stylized character material with smooth surface behavior and subtle form shading, not realistic pores. Hair should read as sculpted clumped strands with controlled gloss, and the beard (if presented) should appear as carved fiber-like grooves with crisp separation. Clothing and background surfaces must stay matte and simplified, supporting the caricature focus.

DETAILS:
Emphasize the rectangular head geometry through faceted forehead planes, sharp temple breaks, and a strong nose bridge structure. Add refined studio-finish highlights along the cheek planes, brow ridge, and hair contours to heighten the premium 3D look. Keep all detail language native to stylized character sculpture, with no comic halftone, no painterly brushwork, and no l

Cinematic Urban Car Meet Scene

game-asset

A detailed prompt for creating a hyper-realistic, nighttime JDM and American muscle car meet in a Los Santos style car park, featuring a cool female subject and atmospheric neon lighting.

展开提示词全文
Hyper-realistic vertical photograph 4:5 ratio. {argument name="location" default="Underground car meet on the third floor of a concrete multi-storey car park in Los Santos at night"}. Bare pillars covered in graffiti, exposed conduit pipes along low ceilings, tyre marks scarring the oil-stained floor. JDM and American muscle cars with wide-body kits, aggressive splitters, big wings, and vibrant pink, blue and purple underglow lighting reflecting on the concrete.
A beautiful young woman with long dark wavy hair is the main subject, standing in full pose beside a slammed deep blue Annis Elegy on air suspension. She is wearing a {argument name="outfit" default="cropped olive-green bomber jacket, wide-leg beige cargo pants"}, and has a cigarette tucked behind her ear. One hand in her pocket, confident and relaxed pose, looking straight at the camera with a cool, unbothered expression.
Behind her, a loose group of guys in streetwear leaning against cars and talking. A {argument name="background car" default="bright yellow Nissan GT-R"} and other modified cars visible in the background. Chain-link fence, half-torn "NO PARKING" sign, security cameras on pillars.
Lighting: Dramatic nighttime scene with mixed neon underglow, overhead fluorescent lights, and city lights glowing through the open side of the car park. Cinematic atmosphere.
Photography style: Photorealistic, Rockstar Games visual quality, ultra-detailed, sharp focus on the woman and foreground cars, rich colors, cinematic lighting, 8K sharpness, natural skin texture. No plastic look, no AI artifacts.

3D Game Character Face-off

game-asset

A detailed prompt for a high-quality 3D render of two distinct game characters in a combat stance against a minimalist background.

展开提示词全文
Highly detailed 3D render of {argument name="characters" default="two AAA game characters in a dynamic face-off pose"}: {argument name="character descriptions" default="a muscular ancient warrior with rugged armor and a heavy spiked mace, and a stealthy ninja in dark layered attire holding a katana and chained sickle"}. Ultra-realistic anatomy, sharp textures, cinematic sculpting, and realistic fabric physics. Both characters facing each other in combat stance, full-body visible, strong silhouette contrast. Clean white minimalist background, soft studio lighting, subtle ground shadows, high contrast, ultra-sharp details, professional game concept art quality, {argument name="aspect ratio" default="16:9"}. Negative: no background scene, no extra characters, no blur, no noise, no text, no UI elements

3D Game Character Transformation

game-asset

Generates a detailed 3D game character based on the pose of a reference photo, specifically creating an old western gold miner.

展开提示词全文
based exactly on the pose in my photo, create a highly detailed {argument name="style" default="3d game character"} of an {argument name="character" default="old western gold miner"} in an {argument name="setting" default="American wilderness"}. same composition

Neo Goku in the Machine City

game-asset

A prompt for generating an image of Neo Goku arriving in a futuristic city populated by machines, created with Nano Banana Pro.

展开提示词全文
{argument name="character" default="Neo Goku"} - Arrives at the {argument name="location" default="city of machines"}

3D Gaming Character Composition

game-asset

A detailed instruction to generate a 3D gaming character based on a specific composition, featuring a monk and an Indian landscape.

展开提示词全文
use the exact composition of @walk for a highly detailed 3D gaming character of a {argument name="subject" default="buddhist monk with a buddha statue strapped to his back"} walking through a {argument name="landscape" default="indian landscape with palmtrees in the back"}. background landscape is even and does not rise towards the edges

Persian Court Robe Portrait

game-asset

A prompt designed to dress a reference subject in regal Persian court attire with complex pleats and silk textures, set in a luxurious palace interior.

展开提示词全文
Ultra-detailed portrait of {argument name="subject" default="the same person from the reference image"}, wearing an {argument name="attire" default="elegant Persian court robe with intricate pleats"}, rich silk fabric, and ornate embroidery inspired by Safavid Empire royal attire, {argument name="color scheme" default="deep jewel tones (emerald, maroon, gold)"}, fine textile patterns, layered drapery, regal posture, soft cinematic lighting, sharp facial resemblance, realistic skin texture, shallow depth of field, symmetrical composition, luxurious palace background with Persian motifs, high resolution, photorealistic, 85mm lens style, ultra aesthetic, highly detailed fabric folds and textures

3D Gaming Character Battle Scene

game-asset

A prompt for generating 3D gaming characters, specifically Spartacus and a ninja, on a minimalist background.

展开提示词全文
3D gaming character {argument name="character name" default="Spartacus"} with {argument name="weapon" default="axe"} and a ninja with a katana in {argument name="background" default="white minimalist background"} neat and clean 16:9

Chibi Penguin on a Motorcycle

game-asset

A playful 3D-style prompt for a chibi character in a penguin onesie sitting on a vintage cruiser motorcycle at sunset.

展开提示词全文
Ultra-realistic, cinematic vertical 9:16 shot of Gugo Gaga, a small chibi-style character with short black hair and gray eyes, wearing a {argument name="character outfit" default="plush penguin onesie"} with a giant zipper detail at the neck. The character is in an extremely happy, joyful mood with eyes closed and a wide, laughing mouth. Gugo Gaga is sitting on a {argument name="vehicle" default="vintage teal and chrome cruiser motorcycle"} parked on a {argument name="setting" default="winding coastal highway"}. The lighting is soft golden hour sunset, casting a warm glow over the scene with a blurred ocean and mountain background. 8k resolution, highly detailed textures on the plush fabric and polished chrome, commercial photography style.

Magical fantasy world illustration

game-asset

A prompt for generating a magical fantasy world scene specifically set in a land of witches.

展开提示词全文
{argument name="theme" default="Magical Fantasy World (Land of Witches)"}

PS1 Retro Gaming Style Prompt

game-asset

A comprehensive prompt designed to transform an image into a low-poly PlayStation 1 era video game screenshot, featuring authentic technical artifacts.

展开提示词全文
Transform the entire image into an authentic screenshot captured directly from a {argument name="console era" default="late-1990s PlayStation 1 (PS1)"} video game running on original hardware. Re-render every single element of the scene — every object, character, surface, background, lighting, and detail — as {argument name="geometry style" default="low-polygon 3D models"} with clearly visible triangular facets, flat angular geometry, and no smoothing or subdivision whatsoever. Use extremely low-resolution textures (256×256 pixels or smaller) that are blurry, pixelated, and heavily compressed, with obvious tiling seams, mipmapping artifacts, and poor UV mapping.
Apply classic PS1 affine texture mapping across all surfaces so textures visibly stretch, warp, wobble, and distort unnaturally with any change in perspective or camera angle (especially noticeable on floors, walls, and curved objects). Remove all anti-aliasing entirely — every edge must be sharply jagged and aliased with prominent stair-stepping. Use only basic flat shading or simple Gouraud vertex shading with no dynamic lights, no real-time shadows, no specular highlights, and no advanced effects. Simulate the PS1’s 16-bit color depth and 5-5-5 RGB limitations by introducing visible checkerboard or ordered dithering patterns in all color gradients, skies, shadows, and semi-transparent areas to prevent banding.
Include typical PS1 graphical quirks: subtle vertex snapping/jitter/wobbling on moving or angled geometry, minor polygon pop-in or Z-fighting where appropriate, and a limited draw distance that may softly fade into a low-detail fog. The overall resolution should feel like a raw {argument name="resolution" default="320×240"} or 640×480 PS1 framebuffer output, giving the whole image a slightly soft yet crunchy retro pixel look when viewed at modern sizes. Preserve the original scene composition and camera angle exactly, but convert everything into this pure PS1 3D style as if the photo itself were modeled and rendered inside an actual PS1 game. No text overlays, no HUD elements, no subtitles, no menu graphics, no scanlines, no CRT curvature, and no modern post-processing of any kind — just a clean, direct screenshot from the console’s video output.

Rugged Urban Portrait

game-asset

A prompt for a gritty, ultra-realistic portrait of a man in an urban alley with cinematic lighting.

展开提示词全文
Ultra-realistic portrait of a {argument name="subject" default="rugged man"} with sharp facial features, messy hair, and a thick beard, intense eyes with a cold expression, slightly bruised face with subtle dirt and scratches, wearing a {argument name="clothing" default="dark worn-out jacket and textured clothing"}, cinematic lighting with deep shadows, moody atmosphere, dramatic contrast, 35mm lens, shallow depth of field, background {argument name="environment" default="blurred urban alley"}, high detail skin texture, masculine energy, raw and gritty aesthetic, 4K, ultra detailed

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。