由于模型属于搬运 以下简介 属于机翻 大家自行补习
WLOP 风格的 LoRA
wlop已经有一个模型检查点,您绝对应该检查一下。然而,LoRA 正在成为一种新的方式,基本上可以随时随地混合模型并添加艺术风格和角色,还将多个模型与许多不同的预先存在的模型相结合。Wlop 风格已经在提示中使用了很多,尤其是在 Midjourney 中,所以我做了这个。
对于预览图片,我使用了 Abyss Orange Mix V2 和0.6 的权重。触发词当然是wlop
。也适用于DreamShaper,但您无法在其上获得完美的样式(它是在动漫模型上训练的)。我还使用了 CLIP skip 2。
如何在 auto1111 中使用 LoRA:
-
更新webui(
git pull
像这里一样使用或者重新下载) -
将文件复制到
stable-diffusion-webui/models/lora
-
确保更改重量(默认情况下它
:1
通常太高)
(masterpiece:1.0), (best quality:1.0), (ultra highres:1.0) ,(8k resolution:1.0),(realistic:1.0),(ultra detailed1:0), (sharp focus1:0), (RAW photo:1.0),
wlop, aeolian, jewelry, cute girl, eastern, red hair, red eyes
<lora:wlopStyleLora_v1:0.75>,pureerosface_v1 <lora:lcePrincessInGhost_v10:0.3>
Negative prompt: (Easy Negative:1.4),(low quality, worst quality:1.4), (bad anatomy), extra digit, fewer digits, (extra arms:1.2), bad hands, by (bad-artist:0.6), bad-image-v2-39000, low quality, worst quality, lowers,monochrome,grayscales,skin spots, acnes, skin blemishes, age spot,6 more fingers on one hand,deformity, bad legs, error legs, bad feet, malformed limbs, extra limbs, <ng_deepnegative_v1_75t>< badhandv4>
ENSD: 31337, Size: 896×1280, Seed: 151499052, Model: salutemix_v1, Steps: 32, hashes: [object Object], Sampler: DPM++ SDE Karras, CFG scale: 7, Clip skip: 2, Mask blur: 4, Model hash: 3177a3a2a0, Denoising strength: 0.4, Tiled Diffusion method: MultiDiffusion, Tiled Diffusion overlap: 48, Tiled Diffusion upscaler: R-ESRGAN 4x+ Anime6B, Tiled Diffusion batch size: 8, Tiled Diffusion tile width: 96, Tiled Diffusion tile height: 96, Tiled Diffusion scale factor: 2
wlop, 1girl, artist name, bangs, black background, black hair, blurry, china dress, chinese clothes, crying, dark, dress, dutch angle, earrings, eyelashes, closed eyes, fantasy, gears, highres, instrument, jewelry, lips, long hair, making-of available, music, original, parted bangs, playing instrument, simple background, solar system, solo, tears, violin
<lora:wlop-000030:0.6>
Negative prompt: mole, anime, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), watermark, text, error, blurry, jpeg artifacts, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, artist name, (worst quality, low quality:1.4), bad anatomy
ENSD: 31337, Size: 576×576, Seed: 3432025314, Steps: 28, Sampler: DPM++ SDE Karras, CFG scale: 10, Clip skip: 2, Mask blur: 4, Model hash: 0873291ac5, Denoising strength: 0.1, SD upscale overlap: 64, SD upscale upscaler: R-ESRGAN 4x+
(masterpiece:1.0), (best quality:1.0), (ultra highres:1.0) ,(8k resolution:1.0),(realistic:1.0),(ultra detailed1:0),
wlop,dalcefo,realistic,amazon girl warrior, fine detailed face, blue eyes,jungle,cinematic light,<lora:wlopStyleLora_30Epochs:0.7>,eastern, detailed beautiful skin,solo, absurdres, detailed beautiful face, petite figure, detailed skin texture, pale skin, thigh gap, detailed hair, random hair style, detailed eyes, glistening skin,castle, pureerosface_v1
Negative prompt: (Easy Negative:1.4),(low quality, worst quality:1.4), (bad anatomy), extra digit, fewer digits, (extra arms:1.2), bad hands, by (bad-artist:0.6), bad-image-v2-39000, low quality, worst quality, lowers,monochrome,grayscales,skin spots, acnes, skin blemishes, age spot,6 more fingers on one hand,deformity, bad legs, error legs, bad feet, malformed limbs, extra limbs, <ng_deepnegative_v1_75t>< badhandv4>
ENSD: 31337, Size: 1024×1024, Seed: 1362915345, Model: dalcefoV3Painting_dalcefoV3Painting, Steps: 25, hashes: [object Object], Sampler: DPM++ SDE Karras, CFG scale: 7, Clip skip: 2, Mask blur: 4, Model hash: c0646f97ce, Denoising strength: 0.35, Tiled Diffusion method: MultiDiffusion, Tiled Diffusion overlap: 48, Tiled Diffusion upscaler: R-ESRGAN 4x+ Anime6B, Tiled Diffusion batch size: 8, Tiled Diffusion tile width: 96, Tiled Diffusion tile height: 96, Tiled Diffusion scale factor: 2
wlop, 1girl, black gloves, black jacket, character request, commentary, earrings, ghostblade, gloves, hand up, jacket, jewelry, long hair, looking down, red hair, solo,
<lora:wlop:0.5>
Negative prompt: (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), watermark, text, error, blurry, jpeg artifacts, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, artist name, (worst quality, low quality:1.4), bad anatomy
ENSD: 31337, Size: 512×768, Seed: 646616586, Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 10, Clip skip: 2, Model hash: e4b17ce185, Hires upscale: 1.6, Hires upscaler: R-ESRGAN 4x+, Denoising strength: 0.31
wlop,dalcefo,realistic,painting,male,fine detailed eyes <lora:wlopStyleLora_30Epochs:0.6>
Negative prompt: easynegative, (worst quality, low quality:1.4),
ENSD: 31337, Size: 640×640, Seed: 2351232376, Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 7, Model hash: d62d1a7201
wlop,1 girl,solo,beautiful face,very long hair,upper body, undercut, dalcefo,realistic,japanese samurai in armor, fine detailed face,samurai armor,cinematic light,<lora:wlopStyleLora_30Epochs:0.6>,
Negative prompt: easynegative, (worst quality, low quality:1.4), signature, text,watermark, artist name,borrowed character, text focus, two girls
ENSD: 31337, Size: 640×640, Seed: 2827755954, Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 7, Model hash: d62d1a7201
LORA
Lykon
2023 年 1 月 23 日
30 epochs
SD 1.5
wlop
288.08MB
隐藏内容需要回复可以看见
楼主辛苦了,谢谢楼主,楼主好人一生平安!这个逼装的我给82分,剩下的用666的方式打给你!
卧槽!无奈本人没文化,一句卧槽行天下!
1
奈何本人没文化,一句卧槽行天下
卧槽!无奈本人没文化,一句卧槽行天下!