APIMODELSAPIMODELS

25+ FLUX 3 人像与时尚提示词

25+ 条。收录 400 条 FLUX 3 图片提示词,来自 Civitai(基础模型标为 Flux.3、Flux.3 Klein、Flux.3 Klein-base 的作品)和 X。每条都逐字全文收录、署名作者、链回原帖,旁边就是它生成的那张图。这些提示词都不是我们写的;apimodels.app 也没有提供 FLUX 3 的图片模型。

← 回到全部 401 条 FLUX 3 提示词

最近更新: · 我们已上线的模型 · API 文档

一句话摘要

FLUX 3 是 Black Forest Labs 的图片模型系列,在 Civitai 上,它的作品分在三个基础模型标签下:Flux.3、Flux.3 Klein 和 Flux.3 Klein-base。本页收录 400 条大家在 Civitai 和 X 上随 FLUX 3 作品一起公开的提示词,每条附它生成的图和来源链接。apimodels.app 没有提供 FLUX 3 的图片模型,所以这些提示词没法通过我们的 API 出图。apimodels.app 上的 FLUX 3 Max 是另一回事:它是视频模型,属于 FLUX 3 Video 的一个版本,能把提示词、一张首帧或最多 10 个定时关键帧变成 5 到 20 秒带声音的视频,$0.221/秒起;用这里的提示词做出来的图,可以拿去当它的首帧。另外,Civitai 上的图还取决于提示词以外的设置,比如种子、采样器、步数和尺寸,只抄文字是复现不出同一张图的。生成信息里列了 LoRA 的 Civitai 作品,我们没有收录。

出处:Black Forest Labs FLUX 3(模型方官方站点)

Y2K猫咪街拍

Y2K猫咪街拍

photorealistic 35mm photograph, natural color, subtle film grain. A cat a bashful shy expression while looking down coyly. She wears a prominent Y2K headband across her forehead with the text "BUZZ ME" clearly printed in bold retro lettering. Outfit consists of an oversized Y2K cropped zip-up hoodie over a white baby tee, paired with ultra-wide balloon-fit baggy cargo pants in washed grey featuring extreme wide leg openings pooling heavily over chunky platform sneakers, exaggerated fabric drapery and wide hem silhouette. Layered silver chain necklace. Pose: dynamic crouching pose on an urban concrete curb, resting one forearm on her knee while hiding half her face behind her hoodie sleeve in a shy gesture, weight balanced back. Setting: a dimly lit outdoor street alleyway with soft neon store signs glowing subtly in the background. Lighting: soft ambient evening streetlight, shot on 35mm lens, f/1.8, eye-level candid angle with shallow depth of field. Muted cool blue Y2K aesthetic with warm neon pops. Negative Prompt: bad art, extra text, watermark, logos (except "BUZZ ME"), distorted hands, fused fingers, extra limbs, warped face, plastic airbrushed skin, uncanny eyes, stiff mannequin pose, CGI, render, cinematic 3D, skinny pants, narrow leg openings, tight jeans.

来自 Graf137看原帖
戴VR头显的分割人像

戴VR头显的分割人像

A mid-body portrait of a 26-year-old gamer with a soft-angled face and pastel pink twin braids, wearing a VR headset. The composition is split at the headset strap, which acts as the visual anchor and dividing boundary. The camera-left half is photographed in dim bedroom light with real skin texture and visible tear tracks on her cheeks. The camera-right half renders as a polished VR avatar with subsurface scattering, smooth skin, and bloom lighting. Cool monitor glow from the left contrasts with warm bedside lamp light from the right. Shot on an 85mm lens, dream-and-flesh color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
分屏人像与粒子代码

分屏人像与粒子代码

A three-quarter portrait of a 41-year-old creative coder with a diamond face and auburn waves, split exactly at the anatomical midline. The camera-left half is a living photograph with real skin texture and north-window key light. The camera-right half flows into p5.js particle systems with flowing noise fields and emergent flocking behaviors, rendered as glowing dots and connected lines on a dark background. The boundary is traced by a single Perlin-noise streamline acting as the visual anchor. Soft key light with algorithmic accent, code-and-skin color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
黑底蓝色眼妆特写肖像

黑底蓝色眼妆特写肖像

A close-up portrait shows a person with blue eyeshadow and yellow face paint, positioned centrally against a black background. Render this in the following style: photorealistic digital rendering, muted industrial grey palette, cool directional lighting, smooth specular finish, shallow depth focus, cyberpunk aesthetic, high dynamic range contrast, detailed metal texture rendering, volumetric atmospheric effect, cinematic perspective distortion, dramatic shadow and highlight gradient, dark monochromatic tone, subtle film grain texture Do not introduce any subject, object, or background from the style reference. Preserve the original composition, pose, and framing. No text, letters, words, signatures, or watermarks anywhere in the image. <lora:flux3_klein_sd_lora_rank_256_bf16:1>

来自 gonzalu看原帖
阴影中悲伤的吸血鬼

阴影中悲伤的吸血鬼

close-up of a vampire’s gaunt face half-submerged in liquid shadow, his human eye glistening with grief while the other burns crimson. Translucent dream-visions of a lost lover ripple across his skin like projected film. Style: Hyperrealistic portraiture fused with smoky charcoal sketches, echoing Francis Bacon’s distorted emotion.

来自 Silmas看原帖
沙发上坐着的女子

沙发上坐着的女子

A beautiful woman sitting on a couch.

来自 mphobbit看原帖
诗人水墨分割肖像

诗人水墨分割肖像

A vertical three-quarter portrait of a 55-year-old poet with a luminous round face and silver-streaked loose bun, split exactly at the anatomical midline. The camera-left half is a living photograph with soft window key light and real skin texture. The camera-right half flows into sumi-e ink wash with bleeding rice-paper textures and negative-space brushwork. A single ink drop frozen mid-fall at the boundary acts as the visual anchor. Warm key light with cool ink-wash accent, paper-and-flesh color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
戴耳机的波浪纹女子

戴耳机的波浪纹女子

Create a new painterly portrait composition that places the female bust from the first image at the center, keeping her closed eyes, full pink lips, dark hair, and over-ear headphones with their thin cables intact, and surrounds her with the repeating wave-and-sun pattern from the second image. Arrange the stylized dark blue waves and textured circular suns in vertical bands that flow behind her shoulders and curve around the sides of her head, so the pattern fills the entire background in rhythmic, textile-like horizontal layers. Render the whole scene in an expressive, painterly style with visible brushstrokes, blending the warm orange, terracotta, and cream tones of the pattern with the vibrant multicolored paint splashes on her face and neck. Let the headphone cables descend from the ear cups and intertwine with the undulating wavy lines, and maintain the warm, saturated color palette throughout the composition.

来自 gonzalu看原帖
伸手极度透视特写

伸手极度透视特写

A woman's hand reaching toward the camera in extreme close-up foreshortening, neatly manicured nails, fingertips razor-sharp and enormous in the foreground while the wrist and arm taper into soft creamy distance, crushed white background, dutch angle 18°, skin texture lit by a single cold silver sidelight, delicate ring on ring finger catching specular highlights, fine art wedding photography, shot on large format film, grain texture, high key.

来自 liutyi看原帖
照片与潜空间分割肖像

照片与潜空间分割肖像

A three-quarter portrait of a 44-year-old AI researcher with a neutral oblong face and a uniform silver braid of one single color front and back. The composition is split at the anatomical midline, which reads as her profile contour in the three-quarter view, rendered as a designed seam of warm gold emissive chain light. The camera-left sliver is a living photograph with a warm window key light coming from camera-right. The camera-right half converts at 100 percent coverage—including cheek, jaw, ear, temple, neck, and hair—into a diffusion latent-space render. This render consists of flowing contour line fields, swirling gradient noise, and glowing embedding particles. The conversion is strictly abstract gradient noise and flowing contour lines only, with no bones, no ribs, no skeleton, and no anatomical render. Her right eye alone remains photographically alive with wet, luminous detail as the focal point, with a gradient-descent path line acting as the secondary visual anchor. Shot on an 85mm lens, sharp focus on the eye, silicon-sage color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
咖啡馆隔代双手相触

咖啡馆隔代双手相触

A close-up portrait of two hands meeting across a café table: the wrinkled, photographed hand of a 68-year-old grandmother meeting the rendered polygonal hand of her 19-year-old granddaughter's avatar. The table surface transitions from real wood grain to PBR-textured virtual oak exactly at the point of contact, which acts as the visual anchor. Warm café key light contrasts with cool holographic rim light. A flat cream cartouche band at the bottom bears the caption in clean uppercase Cyrillic reading 'СВЯЗЬ ПОКОЛЕНИЙ В СЕТИ' in uppercase beautiful font. Shot on a 50mm lens, wood-and-mesh color grade, static atmosphere.

来自 FASCIUM看原帖
秋日林间骑行女子

秋日林间骑行女子

ta86_wistfall, a beautiful adult woman with long flowing copper-red hair and delicate freckles riding a vintage black bicycle along a quiet tree-lined country path in autumn, wearing a soft ivory knitted sweater, faded denim shorts, white sneakers and a loosely wrapped rust-colored scarf fluttering in the breeze, her expression thoughtful and slightly melancholic, vibrant red curls catching the gentle autumn wind, golden and crimson leaves drifting across the path, ancient trees forming a tunnel of fading amber foliage, scattered leaves covering the damp pavement, intimate frontal composition, slightly low camera angle, subtle Dutch tilt, gentle foreshortening of the bicycle's front wheel, cinematic late afternoon backlighting filtering through the branches, soft golden haze, delicate rim lighting around her hair, muted burnt orange, copper, dusty olive, warm ivory and desaturated teal palette, photorealistic facial details blended with painterly watercolor softness, subtle charcoal and ink textures, atmospheric depth, dreamy background bokeh, fragile beauty, poetic nostalgia, quiet solitude, the bittersweet feeling of cycling through the last warm days of autumn

来自 nikonac1808看原帖
鼠尾草色灰墙前的侧脸肖像

鼠尾草色灰墙前的侧脸肖像

Tasteful editorial portrait photograph of a woman in strict profile placed at the extreme left edge of the frame, the back of her shoulder slightly cropped by the left border, wearing a rust-orange ribbed wool turtleneck with her dark hair in a sleek low bun, chin lifted a little as she gazes calmly toward the right. The remaining three quarters of the frame is an unbroken, lime-washed plaster wall in pale dusty sage, its surface softly graded from slightly brighter beside her to deeper at the far right edge, holding only a faint trace of plaster texture. Gentle window light from the right wraps across her cheek, nose and neck, and the large open space in front of her gaze gives the eye somewhere quiet to travel. Generous negative space, clean minimalist composition, 85mm lens look, creamy natural skin tones, matte film color grade, serene and elegant.

来自 liutyi看原帖
装饰风地下酒吧浪漫剧照

装饰风地下酒吧浪漫剧照

MASTERPIECE, ultra-photorealistic cinematic lesbian romance, glamorous 1920s Art Deco speakeasy, intimate film-noir movie still, vertical 2:3 composition. Two extraordinarily elegant adult women share a passionate, private romantic moment beside the polished mahogany bar of an exclusive, dimly illuminated speakeasy. The first woman is a striking tall blonde with sculpted 1920s finger waves, delicate natural features, expressive eyes and deep crimson lipstick. She wears a breathtaking floor-length SILVER LIQUID-METAL evening gown, its flowing fabric shimmering like molten mercury, with brilliant mirror-like highlights and elegant Art Deco tailoring. A luxurious black feather boa falls loosely around her shoulders. The second woman is equally beautiful, with dark auburn hair, refined features and a confident, magnetic expression. She wears a spectacular, very glossy, figure-skimming midnight-black liquid-leather evening gown, with an elegant very open back, a narrow waist and subtle Art Deco detailing. They stand extremely close at the bar, completely absorbed in each other. The woman in black has one arm wrapped around the silver-gowned woman's waist, drawing her into a close embrace. The brunette gently touches her companion's cheek, their bodies turned naturally together, their faces only inches apart. Their eyes meet with unmistakable affection and desire, a spontaneous, deeply romantic kiss about to happen. Both women actively reciprocate the embrace, entirely comfortable with their closeness. Natural expressions, tender physical contact, genuine emotional connection. BACKGROUND: A spectacular but intimate 1920s speakeasy, dark polished mahogany, black marble, antique brass, geometric Art Deco mirrors, blood-red leather banquettes, small amber-shaded lamps and delicate crystal glasses. A lone female saxophonist appears as a distant silhouette on a tiny stage, partially obscured by drifting smoke. The remaining room disappears into rich darkness. No prominent background patrons. LIGHTING: Authentic classic Hollywood noir glamour photography. Dramatic sculpted chiaroscuro, a focused warm amber light illuminating both women's faces, a narrow cool silver rim light defining their hair and silhouettes, deep velvety shadows, soft atmospheric smoke. Brilliant but physically convincing reflections across the liquid-metal silver gown, restrained highlights on the black dress, natural luminous skin. COMPOSITION: The embracing women dominate the image. Their faces, hands and intimate body language form the emotional center, while the bar and beautiful gowns establish the setting. Cinematic realism rather than fashion-model posing. Realistic facial anatomy, subtle skin texture, natural hands and fingers, believable physical contact, magnificent glossy materials, luxurious 1920s glamour, extraordinary photographic depth, fine film grain, sophisticated romantic suspense.

来自 jon_s看原帖
穿靛蓝甚平的坐姿男子

穿靛蓝甚平的坐姿男子

. modern comics style. A seated portrait shows a man wearing an indigo cotton men's casual jinbei with a simple woven stripe, straw sandals, and an uchiwa fan. He is seated on a tatami mat on a relaxed summer veranda with a seasonal alcove arrangement visible, rendered as an analog photomontage and vintage paper collage with halftone print texture, surrealist cutout art, torn paper edges, psychedelic color contrast, and a retro magazine aesthetic. 1boy, solo, seated, men's casual jinbei, indigo cotton, simple woven stripe, straw sandals, uchiwa fan, tatami, summer veranda, seasonal alcove, portrait, analog photomontage, vintage paper collage, halftone print texture, surrealist cutout art, torn paper edges, psychedelic color contrast, retro magazine aesthetic. masterpiece, best quality, very aesthetic, absurdres, amazing quality, highres

来自 geekier看原帖
地铁中的全息倒影人像

地铁中的全息倒影人像

A full-body environmental portrait of a 38-year-old architect with a sculpted oval face and short cropped silver hair, standing between the seats inside a real subway car with steel poles and fluorescent ceiling strips. She is reflected in the dark subway window as a volumetric hologram with visible horizontal scan lines and blue phosphor glow, while her physical body remains a living photograph. The window glass acts as the sharp dividing boundary between the real and holographic renderings. Cold transit fluorescent key light from above casts warm holographic blue spill onto her shoulders. The subway interior is fully visible with seats, poles, and floor on all sides, no black void. Shot on a 35mm lens, full-body composition, steel-cyan color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
穿条纹毛衣的微笑女子

穿条纹毛衣的微笑女子

This photograph captures a young Asian woman with long, wavy dark brown hair and fair skin, smiling warmly at the camera. She is wearing a colorful, knitted turtleneck sweater with vibrant pink, orange, and yellow stripes. Her makeup is subtle, highlighting her natural beauty with a slight blush and pink lipstick. In the blurred background, two other young Asian women with similar hair colors and styles are seated on a brown leather sofa. The woman in the blue sweater is slightly out of focus, while the woman in the pink sweater is more visible. The room has warm, orange-yellow lighting with wall-mounted lamps and a neon pink sign on the right. The wooden door in the background adds a classic touch to the cozy, modern interior. The overall mood is friendly and relaxed, with a focus on the smiling woman in the foreground.

来自 Bakpasser看原帖
阳光公园里的微笑女子

阳光公园里的微笑女子

A beautiful 26-year-old woman with vibrant green eyes and a shy, gentle smile stands in a bright, sun-drenched city park. She is tall with long, elegant legs and a slim, athletic build, wearing a flowy sundress that highlights her natural figure. Her brown hair is loose and natural, catching the soft light of the day. The background is a soft blur of green park foliage and distant city structures, creating a serene and professional portrait aesthetic.

来自 TherasaJean看原帖
黑森林路牌人像

黑森林路牌人像

Objects: A young woman in Black Forest traditional dress with the red Bollenhut pompom hat stands beside a yellow German town-entry road sign that reads "FLUX.3 Klein" in black DIN capitals, with a smaller white plate below it that reads "Base". A small carved wooden cuckoo clock hangs from the sign post. Background: Steep slopes of dense dark fir forest, a traditional Schwarzwald farmhouse with a deep hipped roof, thin mist in the valley. Layout: The sign and her face sit close to the image center and the sign text is large and fully legible. The upper right corner of the image is clear sky. The bottom of the image is dark forest floor and road in shadow. Details: The sign is a rectangular yellow plate with a thin black border. Her dress is black with a red bodice and a white blouse, and the hat carries fourteen red wool pompoms. Style: Photograph, warm late afternoon light, medium depth of field, natural colors. Last detail: Text "5.6B" is handwritten in blue marker on a large white sticker in the top left corner of the image.

来自 liutyi看原帖
未来感和服科技装肖像

未来感和服科技装肖像

A dazzling, high-concept fashion portrait that blends traditional East Asian elegance with futuristic kawaii techwear. Drawing inspiration from the graceful woman with flowing dark hair wearing an ornate, gold-embellished traditional robe amidst blooming cherry blossoms in Her-Golden-Promise-1339962093.jpg, the composition reinterprets her outfit and style into a futuristic, pastel-colored technical space-suit ensemble from Fashion---Maychelle-Farlight-84-1397689969.jpg. The background features a glowing digital cherry blossom forest glowing in soft lavender, white, and golden hues, framing the character as she holds a high-tech glowing trinket. Cinematic lighting, intricate texture details, vibrant contrast, and a seamless fusion of classical heritage and high-fashion sci-fi aesthetics.

来自 Ai_0_iA看原帖
孔版印刷分割人像

孔版印刷分割人像

A mid-body portrait of a 29-year-old zine-maker with a diamond face and coral-pink bob, split exactly at the anatomical midline. The camera-left half is a living photograph in studio light with real skin texture. The camera-right half renders as a two-color risograph print with visible registration misalignment, halftone dots, and ink grain. The boundary is a single overprinted CMYK cross acting as the visual anchor. Warm key light with fluorescent-pink-and-cyan accent, pulp-and-skin color grade, neutral composed facial expression, static atmosphere.

来自 FASCIUM看原帖
侧脸花卉青色肖像

侧脸花卉青色肖像

A person's head is shown in profile on the right side of the image with flowers covering their left side. Render this in the following style: smooth digital render, saturated cyan and coral hue, even soft lighting, matte finish, subtle gradient rendering, uniform edge quality, clean perspective focus, minimal aberration effect, high medium detail, crisp illustration treatment, neutral exposure film, abstract aesthetic design Do not introduce any subject, object, or background from the style reference. Preserve the original composition, pose, and framing. No text, letters, words, signatures, or watermarks anywhere in the image. <lora:flux3_klein_sd_lora_rank_256_bf16:1>

来自 gonzalu看原帖
霓虹光环花卉肖像

霓虹光环花卉肖像

Create a futuristic photographic portrait in the style of the second image, preserving its deep blue background, neon yellow circular halo, sharp side-profile composition, and dark tailored suit. Show the man from the second image facing frame-left in exact profile, retaining his angular facial structure, slicked-back black hair, clean-shaven jaw, and small yellow hoop earring. Illuminate his face softly with the halo’s yellow-green glow while keeping the rest of his skin in deep blue shadow. Place the large flower arrangement from the first image across his shoulder: dense clusters of pale blush-white and cream roses, tiny white blossoms, unopened pink buds, slender stems, and serrated green leaves rest on his shoulder, drape over the bright yellow collar, and cascade naturally down his dark sleeve toward his arm. Preserve the flowers’ translucent petals, fine veins, golden stamens, and scattered water droplets. Let the botanical arrangement spread broadly across the lower frame, with leaves and petals overlapping the suit’s structured shoulder and sleeve. Keep the halo hovering directly behind his head as a complete luminous ring, with its yellow light tracing the silhouette and casting subtle green reflections across the flowers, collar, and suit. The result is one cohesive fusion of organic floral texture and precise futuristic geometry.

来自 gonzalu看原帖
绿肤女子的机械躯干

绿肤女子的机械躯干

Create a high-contrast photographic character artwork of the slender woman from the first image, preserving her green skin, long narrow profile, slicked-back dark green hair in a low bun, and gold circular ear ornament, but replace the pale background with the dark neutral charcoal backdrop from the second image. Keep her head and upper torso in the same left-facing profile, with her green structured coat and white buttoned shirt visible from the waist upward. From her lower torso, transform the body into the pink-and-gold mechanical torso from the second image: a pink armored robotic core erupts through the white shirt, with exposed brass gears, pistons, rivets, and articulated shoulder plates flaring dramatically outward like an enormous second coat. Preserve the woman’s green coat collar as the transition point while the robot’s broad pink shoulders spread to both sides. Exposed brass piping and copper tubing weave through her bun, loop around her head, and cascade down the front of her white shirt, fusing the woman and machine into one continuous figure. Include the robot’s pink segmented head below her neck as a mechanical understructure, while its articulated metal neck, chest plates, and mechanical anatomy remain visible. Use hard directional highlights, deep shadows, metallic reflections, and the second image’s photographic industrial finish.

来自 gonzalu看原帖
复古波浪发型肖像

复古波浪发型肖像

A sculpted finger-wave bob sits high in a vintage black-and-white studio photograph, just below a band of near-black backdrop above her crown: glossy jet-black hair set in tight 1930s-style ridges that sweep back from a deep side part and curl under at the jawline, every ridge of each wave catching its own bright silver highlight. The sitter, a West African woman in her thirties, is held in strict side profile facing the left edge of the frame, chin lifted, eyes lowered beneath long dark lashes in a soft, dreamy half-close, lips closed and dark with the corners turned faintly up into a small private smile. A pearl-and-gold drop earring hangs free below her ear, the pearl and its small gold cap throwing tiny sparkles of light. Her deep brown skin reads as smooth polished silver-grey, luminous along the cheekbone, the jaw and the throat, and the long curve of her neck flows down into a bare shoulder, the main line leading through the picture, above the straight neckline of a strapless black satin gown sitting low across the bottom edge, its sheen a soft band of grey. Crisp white studio key light falls from high in front of her face and carves the waves and the profile with clean edges. Behind her the backdrop is velvety near-black at the top and grades down to a soft pale glow behind her shoulder that rims the nape and the curled ends of the bob. Fine film grain throughout, deep inky blacks, silvery whites, no colour.

来自 @aiartalley看原帖

怎么写好 FLUX 3 提示词

  1. 1先说是什么类型的画. 本页很多提示词开头第一句就交代画的类型:宣纸上的水墨插画、搞笑卡通插画、野生动物摄影,或者干脆一个词 photorealistic。开头这句定了整张图的质感,后面的描述都建立在它上面。
  2. 2用完整的句子描述画面. 本页大部分提示词读起来是一段话,而不是一串标签:是谁或是什么、在哪、正在发生什么,然后是光线和颜色。早期 Civitai 模型常见的那种逗号分隔标签,在这里只占一小部分。
  3. 3把构图写明白. 写得细的提示词都会交代主体在画面里怎么摆:头部从边到边占满画面、正面平视、一侧被画框切掉;正面朝向;特写;广角。想要什么构图就写什么,别交给这个主体最常见的样子。
  4. 4要出现在图里的字,原样放进引号. 本页提示词想让图里出现文字时,会把原话放进引号,再说明字是怎么写的:用奶油写、潦草的手写、大号黄色字母。用之前先看看卡片上那张图,字到底出成了什么样。
  5. 5删掉残留的 LoRA 触发词. Civitai 的提示词里可能带着作者加载的 LoRA 的触发词,常常是风格名旁边一串奇怪的字符,没有那个 LoRA 它就毫无作用。列了 LoRA 的帖子本页没有收录,但帖子没列 LoRA、提示词里却留着触发词的情况仍然可能有。
  6. 6设置不在提示词里. Civitai 会在每张图旁边记下种子、采样器、步数、引导强度和尺寸,它们对结果的影响不比文字小。只抄提示词,出来的会是同一个想法的另一张图。
  7. 7用参考图时,说清楚拿什么、不拿什么. 本页有些 X 上的提示词是配着你上传的人脸照片用的:脸和发型照图 1,并且明确说不要用它的衣服、姿势和背景。这类提示词要在能传参考图的工具里用,而且只用你有权使用的照片。

提示词公式

先写画的类型或风格 + 主体 + 正在发生什么 + 场景 + 构图和视角 + 光线和颜色 + 要出现的文字(原样放进引号,并说明怎么写)

例(本库收录的原文,原样引用,作者 TheP3NGU1N,发在 Civitai):photorealistic. A penguin holds a birthday cake over the edge of a trashcan about to drop it into the trashcan, on the cake written in frosting using sloppy handwriting "FLUX 3, WE TRIED!" the penguin is in a living room at a birthday party for other penguins.

常见问题

FLUX 3 是什么?Flux.3 Klein 和 Klein-base 又是什么?

FLUX 3 是 Black Forest Labs 的图片模型系列;Flux.3、Flux.3 Klein、Flux.3 Klein-base 是 Civitai 给它的作品分类用的三个基础模型标签。我们没有把这三个放在一起测过,所以不在这里讲它们的区别,想知道的话看作者原帖和 Black Forest Labs 自己的页面。每张卡片都链回了原帖。

能在 apimodels.app 上跑这些提示词吗?

不能:apimodels.app 没有提供 FLUX 3 的图片模型,这些提示词没法通过我们的 API 出图。我们提供的是 FLUX 3 Max(flux-3-max),一个视频模型,属于 FLUX 3 Video 的一个版本:它能把提示词、一张首帧(可再加尾帧)或最多 10 个定时关键帧变成 5 到 20 秒带声音的视频,720p $0.221/秒、1080p $0.377/秒。想跑本页的图片提示词,要去提供对应 FLUX 3 版本的工具里跑;每张卡片都链回了原帖。

FLUX 3 的图片提示词和视频提示词有什么区别?

图片提示词只描述一个静止的瞬间,视频提示词还得写清楚什么在动、镜头怎么走、有多长、听起来是什么样。把本页的提示词直接贴进 FLUX 3 Max,也能出一段视频,但动作和声音全靠模型自己猜。更好的做法是拿定稿的图当首帧,再在提示词里写它怎么动。FLUX 3 的视频提示词在本站另有一页。

为什么我出的图和卡片上的不一样?

因为提示词只是这张图的一部分。Civitai 上的图还取决于基础模型(Flux.3、Klein 还是 Klein-base)、种子、采样器、步数、引导强度和尺寸,有时还有作者加载的 LoRA 的触发词。原帖里列了这些设置的,照着抄;没列的,就当是同一个想法的另一种出法。

这些提示词是谁写的?

是在 Civitai 和 X 上发这些图的人写的,我们一个字都没写、也没改。每张卡片都逐字全文引用提示词、写明作者、链回原帖。提示词和图片归原作者所有;我们自己写的说明和整理的数据按 CC BY 4.0 授权。要更正署名或撤下某一条,发邮件给我们,或者在 GitHub 仓库提 issue。

本页的图是 apimodels.app 生成的吗?

不是:每张图都是作者随提示词一起发的那张,直接从原帖显示。我们没有生成、修改或重新上传任何一张。

这些提示词和图片能商用吗?

提示词和图片归原作者所有,商用前先看原帖怎么说。你自己做出来的图,要看你用的那个 FLUX 3 模型的许可证,以及你用的工具的条款。

apimodels.app 不提供 FLUX 3 出图

我们没有接 FLUX 3 的图片模型,所以这些提示词没法通过我们的 API 出图。我们接了 FLUX 3 Max 视频模型:把这里的一张图当首帧,或者把几张图钉成定时关键帧,就能做成 5 到 20 秒带声音的视频,$0.221/秒起,失败不扣费。

其它提示词库