APIMODELSAPIMODELS

FLUX 3 动漫提示词

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)

Q版猫娘抓娃娃机

Q版猫娘抓娃娃机

a chibi anime red hair cat girl shy and cute face reach out her hand and another hand point back point at a claw machine behind her and a talking bubble on top of her 'Give Me All All Your BUZZ'

来自 cococoffee9700看原帖
埼玉墓碑自拍

埼玉墓碑自拍

A digital anime art depicts a man, saitama \(one-punch man\), at right of the image, poses with a peace sign and taking selfie next to a tombstone at left of the image. He sits in a field of red spider lilies under a glowing aurora borealis. His eyes are covered by black censor bar with white text "MOSS". The tombstone engraved with words "RIP", "Flux3 Klein", "15.10.2026-17.10.2026". This medium shot features a vibrant anime style with dramatic lighting, contrasting the somber grave with his smug, cheerful expression.

来自 reakaakasky看原帖
霓虹蘑菇下的星光青蛙

霓虹蘑菇下的星光青蛙

anime style, frog made of black and blue shadows filled with starry glimmer, sitting under a neon spotted mushroom, dark bioluminescent forest floor, light rain falling, adorable

来自 Graf137看原帖
紫发恶魔少女与黄色制服

紫发恶魔少女与黄色制服

demon girl with purple hair. professional yellow Legionnaire costume. A mix of retro anime style and modern anime style: -Modern anime elements: clean linework, high detail in faces and clothing, cel animation, overall face design and body proportions ("Frieren: Beyond Journey's End", "Mushoku Tensei", "Arknights: Endfield", "Zenless Zone Zero"). -Retro anime elements: high detail in objects, detailed background, high contrast expressive shading (Mamoru Oshii "Ghost in the Shell", Satoshi Kon "Paprika"). No text. Vertical 2:3 resolution.

来自 Neuronomicon看原帖
持血剑的兽耳小红帽

持血剑的兽耳小红帽

masterpiece, best quality, amazing quality, very aesthetic, 4k, high resolution, ultra-detailed, absurdres, newest, scenery, BREAK anthro, furry female, dog girl, blue eyes, little red riding hood girl, white blouse with lace collar and gold ribbon bow, hood and cloak, long blonde hair in two thick braids over her shoulders, straight bangs, digitigrade legs, sitting on a large moss-covered tree stump in the autumn forest leaning against the large bloodied sword, smirk, full body, blue hour, partially lit

来自 beast_regards看原帖
动漫女巫施展魔法

动漫女巫施展魔法

masterpiece, best quality, 1girl, solo, anime manga style, beautiful witch, vibrant ginger hair, long wavy hair, emerald green eyes, wearing a black witch hat and elegant magical robes, holding a glowing magic wand, casting a spell, mystical aura, fantasy background, dynamic lighting, cell shaded, crisp lines

来自 natalia_nurgling看原帖
大理石柱间的精灵少女

大理石柱间的精灵少女

Elf girl, teal green hair, messy and loose, long pointed ears, big dark eyes, blue robes, a solitary figure standing quietly among the columns, subtle arcane details on the fabric, robes slightly too large for her, her ears just visible above the collar, detailed face, anime style, masterpiece, best quality, sharp focus, ancient white marble columns behind her, a simple open arcade structure, clean lines and still air, foliage, vines and greenery softening the stonework, sunlight on marble

来自 intangible45464748看原帖
戴灯泡的皮卡丘

戴灯泡的皮卡丘

A cheerful Pikachu stands in the center of the frame wearing a lightbulb on its head, with its hands clasped together in a begging gesture. The background features a nostalgic Kanto region landscape with subtle retro video game UI elements like a green circle and red cross in the corner. A light blue dialogue box with a white interior is positioned at the bottom, displaying the Japanese text "バズをお餞いくれ" in bold black characters. The lighting is bright and vibrant, emphasizing the glowing lightbulb and Pikachu's yellow fur.

来自 houstoncloudy看原帖
温馨动漫女孩与猫

温馨动漫女孩与猫

Bradhamel art style, A charming hand-painted anime illustration of a young woman with soft brown hair, a small side braid, round glasses and gentle green eyes, sitting comfortably on a cozy sofa. A fluffy cat sleeps peacefully curled up on her lap, its tiny paws tucked beneath its chin. She smiles tenderly while stroking its fur. Warm vintage lamplight, soft blankets, a little book and a peaceful evening atmosphere. Loose expressive manga linework, careless watercolor and gouache brushstrokes, colorful paint splashes and tiny pigment droplets. Visible paper grain, unfinished pencil sketches and uneven outlines. The illustration gradually fades into pale translucent washes and untouched paper toward the edges, as if the distracted artist abandoned the coloring halfway through to watch anime. Cozy, dreamy, affectionate, whimsical 2D storybook art.

来自 nikonac1808看原帖
水墨武士少女立于风暴

水墨武士少女立于风暴

anime style,A strikingly beautiful young anime samurai woman standing alone on an ancient Japanese mountain pathway during a storm, looking directly at the viewer with intense dark crimson eyes, long flowing black hair tied with a red silk ribbon, wearing an elegant traditional white kimono with intricate red floral patterns and holding a beautifully crafted katana. An explosive artistic fusion of expressive ink riot illustration, traditional Japanese Sumi-e painting, aggressive dry-brush calligraphy and dramatic crimson watercolor splashes. Her silhouette and facial features are defined by delicate but confident black ink outlines, while her flowing hair, clothing and surrounding environment dissolve into expressive brush strokes, dramatic ink splatters and fluid red watercolor pigments. An ancient Japanese temple stands on a distant mountain, twisted cherry blossom branches frame the composition, hundreds of red petals fly through the storm, dramatic black clouds roll across the sky and distant mountains disappear into atmospheric ink washes. Powerful monochromatic black-and-white composition with carefully placed crimson-red accents, exceptional contrast, dynamic brushwork, handmade washi paper texture, expressive negative space, dramatic Japanese fine-art illustration, extraordinary visual impact, 4K.

来自 astinsalvijas447看原帖
霓虹夜雨和服少女

霓虹夜雨和服少女

anime style,An exceptionally beautiful anime illustration of an adorable young woman with long flowing midnight-blue hair and sparkling expressive violet eyes, wearing an exquisite futuristic Japanese kimono decorated with tiny luminous cherry blossoms, standing on a quiet Tokyo street during a magical midnight rain, delicate pink petals floating around her, neon signs reflecting across wet pavement, a small transparent umbrella glowing softly, distant futuristic skyscrapers, charming natural expression, sophisticated anime character design, exquisite linework, rich detailed background, cinematic composition, beautiful color harmony, masterpiece anime artwork, 4K.

来自 astinsalvijas447看原帖
万圣夜魔影之约

万圣夜魔影之约

Anime-style illustration, full-body composition. At a Halloween party, an elegant, fully human-looking man with long silver hair and formal black clothes stands in front of a decorated wall, facing the viewer. He extends one hand toward the viewer as if inviting them to dance, wearing a subtle smile. A strong warm light shines from the front-left, casting one crisp, flat shadow of the man onto the wall behind him, to his right. The shadow is clearly attached to his feet and matches his pose and outline, except its silhouette has two curved horns and large bat wings. The horns and wings appear only in the wall shadow; the man himself has no horns or wings. The shadow is a simple, opaque, two-dimensional shape on the wall, not a second person or a darkened copy of the man. Keep the man brightly lit and clearly visible. Halloween decorations and party lights in the background.

来自 greatsquare看原帖
屋顶上女子与六只猫

屋顶上女子与六只猫

ta86supercel94, a 27-year-old tall woman with a short messy dark-brown bob sits on the ridge of an old tiled rooftop, surrounded by six cats of different colors and sizes. Seen from behind, she wears an oversized burgundy hoodie, a short denim skirt, and casual sneakers, her hair gently blowing in the evening breeze. The cats sit and lounge around her in different relaxed poses, their silhouettes illuminated by the enormous golden sun setting over a sprawling Japanese city. Countless rooftops, distant antennas, and apartment buildings stretch toward the horizon beneath spectacular fiery orange, coral-pink, and golden clouds. Wide cinematic 4:3 composition, slightly elevated rear three-quarter view, subtle Dutch angle, asymmetrical arrangement of the woman and cats, breathtaking sunset atmosphere, peaceful nostalgic mood. Authentic 1990s Japanese hand-drawn anime cel animation fused with 16-bit adventure game illustration, bold tapered ink outlines, vibrant warm colors, crisp two-step cel shading, deep indigo silhouettes, glowing golden rim light, subtle analog grain, emotional nostalgic anime ending scene.

来自 nikonac1808看原帖
凯拉在废弃温室中

凯拉在废弃温室中

lazypos, 1girl, kaela kovalskia, hololive, use the hairstyle and hair ribbon of the character from this image, not her outfit, to generate image with this prompt: Kaela Kovalskia from hololive Indonesia wandering through an abandoned botanical garden where enormous blossoms, tangled climbing vines, and unusual plants have reclaimed a forgotten greenhouse. She wears a cream-colored vintage blouse with delicate ruffled cuffs, a dark forest-green pinafore dress, lace stockings, and worn leather shoes. Her long blonde hair falls loosely over her shoulders, glowing softly in the sunlight filtering through the weathered glass roof. Holding a small bouquet of wildflowers against her chest, she gently raises one hand toward a butterfly resting on her fingertip, gazing at it with a quiet, wistful expression. The greenhouse is filled with oversized flowers, sprawling leaves, delicate ferns, cracked glass panels, rusted iron frames, and moss-covered stone pathways. Sunbeams filter through dusty panes, illuminating floating pollen and suspended dust particles. Some flowers grow impossibly large, their translucent petals overlapping the architecture and dissolving into the surrounding atmosphere. Parts of Kaela's silhouette gradually merge with translucent leaves, flower petals, and delicate botanical silhouettes, creating an ethereal double-exposure effect in which the boundary between her figure and the garden becomes indistinct. Use a medium-wide environmental composition, placing Kaela slightly off-center and allowing the overgrown botanical scenery to share equal importance with her. Frame the scene with softly blurred foreground foliage, while filtered sunlight and layered vegetation create natural depth. Keep her face, butterfly, and bouquet gently defined as the emotional focal points. Nostalgic analog photography blended with delicate anime character illustration, botanical watercolor, and dreamy surrealism. Soft film grain, subtle light leaks, faded photographic tones, translucent watercolor washes, delicate ink details, and organic pigment bleeding create a tactile, vintage appearance. Muted sage green, dusty olive, antique cream, faded ochre, and subdued rose pink, with warm sunlight contrasting against cool greenhouse shadows. Peaceful, nostalgic, dreamlike, and slightly melancholic, like a forgotten memory preserved in an old photograph. No text, no lettering, no watermark, no oversaturated colors, no harsh outlines, no excessive fantasy effects.

来自 nikonac1808看原帖
动漫少女吃蛋糕

动漫少女吃蛋糕

Anime-style illustration of Hoshino Miyako, a solo young woman with fair skin and short brown hair, seated indoors at a small dining table. She is shown in a waist-up three-quarter view near the center, wearing a cream-and-brown track jacket zipped halfway open over a plain white shirt. Her eyes are closed, her mouth open in a broad smile as she eats cake, with one corner of her lips lifted around a bite-sized slice held close to her face. A thick lock of brown hair falls across one eye while leaving the other eye fully closed beneath it; the rest of her cropped hair has uneven bangs and loose strands framing her cheeks. In front of her on the left side of the table sits a round ceramic cup filled with pale pink drink, beside a large frosted birthday cake topped with strawberries and chocolate sprinkles. Her raised hand holds a fork carrying a piece of cake toward her open mouth. On the right, a glazed doughnut rests on a small plate, its icing covered with rainbow sprinkles. Another pastry plate directly below her chin holds a triangular slice of layered cake with cream-colored frosting and red fruit pieces. The tabletop fills the lower foreground with scattered crumbs, a folded napkin, and a second empty dessert plate partly cropped by the edge. Behind her, muted beige walls, a wooden shelf, and a shaded indoor lamp form simple blocks of warm color, with soft amber light falling across her jacket and the food.

来自 gonzalu看原帖
Q版可可饼干少女

Q版可可饼干少女

Chibi-style official Cookie Run character illustration of one solo girl, Cocoa Cookie, shown full-body against a simple pale cream background with a few flat cocoa-brown circles and no additional characters. She has dark skin, an oversized head, a compact body, long brown hair falling past her shoulders in smooth rounded locks, and a large white hat with a curved brim and small chocolate-brown trim. Her eyes are closed as two upward-curving lines shaped like "> <", while the top of her face forms a clear "^ ^" expression above her eyebrows. Her mouth is open in a broad smile, showing a neat row of small white teeth and a pink tongue. She wears a white chibi dress with short sleeves, a cocoa-brown waistband, matching brown shoes, and a round cookie emblem on the front. Both hands hold a warm ceramic cup at chest height; the cup is light beige with a thick brown rim and two curling steam shapes rising from it. The character’s name appears through her recognizable cocoa-cookie design: a golden-brown circular biscuit texture covers her cheeks and lower torso accents, dotted with darker baked patches. Centered beneath her feet, bold rounded lettering reads "COCOA COOKIE". Clean outlines, bright cel-shaded colors, balanced symmetry, generous empty space around the single figure, and even shadowless lighting keep the composition safe, official, uncluttered, and readable.

来自 gonzalu看原帖
机库中的巨大步行战机

机库中的巨大步行战机

Create a hyper-detailed anime illustration of the colossal walking war machine from the second image striding through a vast industrial hangar bay, preserving its immense angular legs, blocky armored body, projecting buttresses, antenna masts, vents, and weathered exterior while transforming its pale metal into a teal-tinted, densely mechanical world. The walker rises nearly to the hangar ceiling, its feet planted across distant gantries as one leg advances and the other pushes backward. Seat the woman from the first image upon the main upper deck with her back turned toward the viewer, preserving her dark hair gathered into a compact bun, distinctive fitted white bodysuit, segmented white backpack, and pale mechanical harness. She operates a central interface that erupts into a cascading web of tangled black cables, exposed circuitry, conduits, and mechanical components. This wire network spreads outward and merges directly into the walker’s hull, wrapping around its armor plates and flowing through its legs like veins. Keep the woman’s head aligned with the first image’s rear-facing pose while her hands manipulate the console. Surround her with the first image’s dense black cable canopy, teal structural columns, suspended machinery, and tiny red indicator lights, replacing the outdoor landscape with towering hangar trusses, catwalks, cranes, and shadowed industrial platforms. Emphasize the mechanical symbiosis between operator, console, and colossal vehicle.

来自 gonzalu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