Triuni Style Lora (Illustrious / Pony)

详情

模型描述

请先阅读“关于此版本”部分中的版本详情和建议。

在 Civitai 上基于 furry 艺术家 Triuni 训练的 LoRA

或称为 triuni_h

遗憾的是,作者已删除了他所有的社交媒体(Twitter/FurAffinity/Hentai-Foundry/Newgrounds/Tumblr)

  1. 始终使用标签 by triuni

  2. 此 LoRA 基于 Pony V6 版本 / Illustrious

  3. 尽管使用了 Pony V6 基础模型,但在某些其他模型中生成的图像可能存在不一致,尤其是在眼睛和手部若在 Civitai 上使用,建议加入一些 Embedding / Face Fix 以获得更好的效果。对于使用 ComfyUI 的用户,最佳效果需结合 Hirefix / DDetailer 来改善角色面部图像。

  4. 请勿忘记使用与 Pony 模型相关的标签:

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up,
  1. 在 Civitai 上调整 LoRA 权重由您自行决定

  2. 如果您希望生成的图像没有任何形式的审查,请将 censored, out-of-frame censoring, mosaic censoring, bar censor 标签添加到负面提示中,因为部分由该 LoRA 生成的图像原作者绘制时带有审查处理。

  3. 为在 Civitai 上更好地使用 LoRA,我推荐使用 VAE (SDXL VAE) 和 Embedding (Pony PDXL Negative Embeddings),以减少手部和眼睛细节的错误。

  4. 请先检查与您选择的图像兼容的模型,某些模型会影响 LoRA 在最终图像生成中的效果,尤其针对以下类型:furry、人类、写实图像等。因此,同一模型并不总是适用于所有类型的图像。

  5. 若您希望强制图像生成器使用您选择的 LoRA 权重,请在正向提示中使用以下标签:

Illustrious (v1.0)

<lora:Triuni_Style_Lora_Illustrious:1>

Pony (v1.0)

<lora:Triuni_Style_Lora:0.8>

注意:LoRA 名称末尾的数字(如 0.8)代表 LoRA 权重。在图像生成器中我们有滑块调整权重,但若想“强制”使用特定权重,请在正向提示中使用与生成器滑块相同的数值,例如:

Illustrious (v1.0)

Civitai 滑块 = 1

正向提示: <lora:Triuni_Style_Lora_Illustrious:1>

Pony (v1.0)

Civitai 滑块 = 0.8

正向提示: <lora:Triuni_Style_Lora:0.8>

注意:该艺术家使用了多个知名 furry 角色(卡通、电影、漫画等),但在训练中我让 LoRA 保持灵活,便于与其他 LoRA 混合使用。我的建议是使用特定角色的 LoRA,以获得更满意的结果。

  1. 我添加了与艺术家最知名角色相关的标签,若希望生成接近该角色的图像,请务必输入角色名称。以下为每个角色对应的正向提示示例

Rimi

by triuni, rimi, furry, furry female, anthro, fox, canine, canid, brown hair, red eyes, snout, bedroom eyes, half-closed eyes, narrowed eyes, big breasts, huge breasts, curvy figure, eyebrows, eyelashes, eyeshadow, fluffy, fluffy tail, makeup, orange body, orange fur, tail, tan body, tan fur, thick thighs, wide hips, asian clothing, east asian clothing, japanese clothing, 1girl, solo, naughty face, blush, fox tail, animal ears, fox ears, looking at viewer, leaning forward, front view, grey background, gradient background, fox girl, sweat, snout, smile, seductive, standing, upper body, hands on waist

Miyu

by triuni, miyu, furry, furry female, anthro, fox, true fox, artic fox, canine, canid, big breasts, huge breasts, yellow eyes, white body, white fur, white hair, long hair, fluff hair, thick thighs, wide hips, 1girl, solo, sitting, sitting on knees, seiza, animal ears, japanese bedroom, japanese clothes, cleavage, snout, sweat, sweating, sweat dripping, off shoulder, large breasts, barefoot, open mouth, day, wariza, kimono, teeth, hair ornament, cum, tree, body fur, shimenawa, hot, bare shoulders, outdoors, no panties, full body, sky, smile, tail, hand up, animal nose, shide, thick thighs, long sleeves, sash, steam, steaming body

Yuuka

by triuni, yuuka, furry, furry female, anthro, fox, canine, canid, big breasts, large breasts, black hair, black nose, brown eyes, curvy figure, grey body, grey fur, slightly chubby, thick thighs, voluptuous, wide hips, 1girl, nipples, on side, animal ears, pillows, solo, lying, glasses, looking at viewer, smile, nude, blanket, black-framed eyewear, bangs, round eyewear, bed sheet, animal nose, long hair, snout, blunt bangs, naked sheet, body fur, fox ears, parted lips, navel, completely nude, bite, biting lip, biting own lip, furniture, inviting under covers, lifting covers, sagging breasts, self bite, sweat, sweating, glistening body, shiny body

以下为初始配置,但您可自由使用其他任何配置:

ILLUSTRIOUS

模型

WAI-NSFW-illustrious-SDXL - v12.0

LoRA

Triuni Style Lora - v1.0 (illustrious) - 0.5 ~ 1.0 (建议使用 1)

高级设置

CFG Scale - 3.0 ~ 9.0 (建议使用 6)

采样器 - DDIM / Euler a / Euler / Heun / DPM2 / DPM++ 2M Karras / Uni PC (建议使用 Euler a)

步数 - 20 ~ 60 (建议使用 30)

Clip Skip - 1/2

VAE - SDXL VAE

正向提示

masterpiece, best quality, amazing quality, very aesthetic, high resolution, ultra-detailed, absurdres, newest, scenery, depth of field, scenery, volumetric lighting, by triuni, rimi, furry, furry female, anthro, fox, canine, canid, brown hair, red eyes, snout, bedroom eyes, half-closed eyes, narrowed eyes, big breasts, huge breasts, curvy figure, eyebrows, eyelashes, eyeshadow, fluffy, fluffy tail, makeup, orange body, orange fur, tail, tan body, tan fur, thick thighs, wide hips, 1girl, solo, naughty face, blush, fox tail, animal ears, fox ears, looking at viewer, front view, grey background, gradient background, fox girl, sweat, snout, smile, asian clothing, east asian clothing, japanese clothing, seductive, standing, upper body, hands behind head, presenting armpit, sweaty armpit

负向提示

bad quality, worst quality, worst detail, bad quality, worst quality, worst detail, sketch, censor, multiple tails, modern, recent, old, oldest, graphic, cartoon, text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured, long body, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, very displeasing, (worst quality, bad quality:1.2), bad anatomy, sketch, jpeg artifacts, signature, watermark, username, simple background, conjoined, ai-generated, worst quality, low quality, off-topic, old, early, chromatic aberration, abstract, jpeg artifacts, scan artifacts, logo

PONY

模型

WAI-ANI-NSFW-PONYXL - v13.0

LoRA

Triuni Style Lora - v1.0 (Pony) - 0.5 ~ 1.0 (建议使用 0.8)

高级设置

CFG Scale - 3.0 ~ 9.0 (建议使用 7)

采样器 - DDIM / Euler a / Euler / Heun / DPM2 / DPM++ 2M Karras / Uni PC (建议使用 Euler a)

步数 - 20 ~ 60 (建议使用 30)

Clip Skip - 1/2

VAE - SDXL VAE

正向提示

score_9, score_8_up, score_7_up, by triuni, toriel, undertale, 1girl, solo, furry, tail, pussy, furry female, flower, all fours, animal ears, looking back, anus, goat girl, horns, big butt, big breasts, barefoot, goat ears, smile, sweat, from behind, looking at viewer, dress, dress lift, blush, purple eyes, bottomless, no panties, long sleeves, clothes lift, thick thighs, wide hips, flower field, black background, simple background

负向提示

zPDXL3, 3D, 2D, censored, out-of-frame censoring, mosaic censoring, bar censor, human

由于该模型基于 Pony Diffusion,负向提示中无需过多指定,因为 Pony 模型本身非常灵活,但我始终保留 zPDXL-neg 标签(/model/332646/pony-pdxl-negative-embeddings)。

通用建议:

  • 若可能,在使用此 LoRA 时请注明作者 Triuni 的署名。

  • 请勿出售或商业化此 LoRA,如有可能,请通过您的 Patreon 支持该作者。

  • 请勿移除由 LoRA 生成图像的元数据。

此模型生成的图像

未找到图像。