應該都能在 civitai 和 huggingface 上找到, 主要集中記錄一些會用到的 LoRAs 的 Prompts。 civitai 的搜索好難用,只有我這麼覺得嗎
使用的 checkpoints:#
AbyssOrangeMix 2#
很好用的 Model, 只用少量參數都能出很不錯的圖
How to use#
- VAE: orangemix.vae.pt
- Prompts 可以是長或短
越簡單越好。不要添加過多的細節提示。先從這個負面提示開始。
(worst quality, low quality:1.4) - Sampler: “DPM++ SDE Karras” 是不錯的選擇
- Steps: forTest: 12~ ,illustration: 20~
- Clipskip: 1 或 2
- Upscaler : Latenet (nearest-exact)
- CFG Scale : 5 或 6 (4~8)
- Denoise 強度: 0.5 (0.45~0.6)
如果使用 0.7~,圖片會變化太大。
如果低於 0.45,會出現區塊噪音
Use Models#
- AnythingV3.0 huggingface pruned
[2700c435]「543bcbc21294831c6245cd74c8a7707761e28812c690f946cb81fef930d54b5e」 - NovelAI animefull-final-pruned
[925997e9]「89d59c3dde4c56c6d5c41da34cc55ce479d93b4007046980934b14db71bdb2a8」 - NovelAI sfw
[1d4a34af]「22fa233c2dfd7748d534be603345cb9abf994a23244dfdfc1013f4f90322feca」 - Gape60
[25396b85]「893cca5903ccd0519876f58f4bc188dd8fcc5beb8a69c1a3f1a5fe314bb573f5」 - BasilMix
「bbf07e3a1c3482c138d096f7dcdb4581a2aa573b74a68ba0906c7b657942f1c2」
AbyssOrangeMix3#
AOM2 的改進版,但是很多人還是用的 AOM2, 導致我在 AOM3 上出不了相似的結果。
The AOM3 is a merge of the following two models into AOM2sfw using U-Net Blocks Weight Merge, while extracting only the NSFW content part.
(1) NAI: trained in Danbooru
(2)gape: Finetune model of NAI trained on Danbooru's very hardcore NSFW content.
How to use#
Prompts
- **Negative prompts ** 越簡單越好。
(worst quality, low quality:1.4) - 使用 "3D" 作為負面會導致在 "sketch" 級別的粗略草圖風格。使用時要小心,因為這是一個非常強的提示。
- 如何避免 Real Face
(realistic, lip, nose, tooth, rouge, lipstick, eyeshadow:1.0), (abs, muscular, rib:1.0), - 如何避免 Bokeh
(depth of field, bokeh, blurry:1.4) - 如何去除 mosaic:
(censored, mosaic censoring, bar censor, convenient censoring, pointless censoring:1.0), - 如何去除 blush:
(blush, embarrassed, nose blush, light blush, full-face blush:1.4), - 如何去除 NSFW 效果:
(trembling, motion lines, motion blur, emphasis lines:1.2), - 基本的負面提示範例對於動漫女孩 ↓
- v1
nsfw, (worst quality, low quality:1.4), (realistic, lip, nose, tooth, rouge, lipstick, eyeshadow:1.0), (dusty sunbeams:1.0),, (abs, muscular, rib:1.0), (depth of field, bokeh, blurry:1.4),(motion lines, motion blur:1.4), (greyscale, monochrome:1.0), text, title, logo, signature - v2
nsfw, (worst quality, low quality:1.4), (lip, nose, tooth, rouge, lipstick, eyeshadow:1.4), (blush:1.2), (jpeg artifacts:1.4), (depth of field, bokeh, blurry, film grain, chromatic aberration, lens flare:1.0), (1boy, abs, muscular, rib:1.0), greyscale, monochrome, dusty sunbeams, trembling, motion lines, motion blur, emphasis lines, text, title, logo, signature,
- v1
- Sampler: “DPM++ SDE Karras” 是不錯的選擇
- Steps:
DPM++ SDE Karras: Test: 12~ ,illustration: 20~
DPM++ 2M Karras: Test: 20~ ,illustration: 28~ - Clipskip: 1 或 2
- CFG: 8 (6~12)
- Upscaler : Detailed illust → Latenet (nearest-exact)
Denoise strength: 0.5 (0.5~0.6) - 簡單 upscale: Swin IR, ESRGAN, Remacri 等等…
Denoise strength: 可以設置低一些。 (0.35~0.6)
Use Models#
- AOM2sfw 「038ba203d8ba3c8af24f14e01fbb870c85bbb8d4b6d9520804828f4193d12ce9」
- AnythingV3.0 huggingface pruned [2700c435]「543bcbc21294831c6245cd74c8a7707761e28812c690f946cb81fef930d54b5e」
- NovelAI animefull-final-pruned [925997e9]「89d59c3dde4c56c6d5c41da34cc55ce479d93b4007046980934b14db71bdb2a8」
- NovelAI sfw [1d4a34af]「22fa233c2dfd7748d534be603345cb9abf994a23244dfdfc1013f4f90322feca」
- Gape60 [25396b85]「893cca5903ccd0519876f58f4bc188dd8fcc5beb8a69c1a3f1a5fe314bb573f5」
- BasilMix 「bbf07e3a1c3482c138d096f7dcdb4581a2aa573b74a68ba0906c7b657942f1c2」
- chilloutmix_fp16.safetensors 「4b3bf0860b7f372481d0b6ac306fed43b0635caf8aa788e28b32377675ce7630」
- Counterfeit-V2.5_fp16.safetensors 「71e703a0fca0e284dd9868bca3ce63c64084db1f0d68835f0a31e1f4e5b7cca6」
- kenshi_01_fp16.safetensors 「3b3982f3aaeaa8af3639a19001067905e146179b6cddf2e3b34a474a0acae7fa」
YesMix#
作者用超過 10k 張 NSFW 圖片訓練的畫色圖的 Model, 不幸的是用力過猛了,所以又混合了一些別的 Model, 包括:
- Hassanblend by sdhassan (0.15)
- Anyhentai by asdpro123 (0.1)
- AbyssOrangeMix2 - Hardcore by Havoc (0.15)
- 7th-anime-v3 by syaimu (0.2)
- Kenshi by Luna (0.15)
- Grapefruit by Ikena (0.1)
- ColorbombMix by mocker (0.1)
- Counterfeit v2.5 by rqdwdw (0.1)
以及:
- Moxin by simhuang to enhance face and style.
- Fashion Girl by me to enhance face.
- Akiryo's Mai by me to slightly finetune face, body shape and skin texture.
使用的 LoRAs:#
很多內容複製於位於 civitai 的原始鏈接
garterbelt
ppfs.safetensors| Panty Pull / Clothes Pull | Clothing Lora 446 https://civitai.com/models/8406/panty-pull-clothes-pull-or-clothing-lora-446
panty pull, clothes pull
shiny_skin to trigger, use tan / dark_skin to deepen the skin tone.
Some prompts about lighting and skin can be used:
oiled_skin / glistening_skin / high_contrast / perfect_lighting / etc.
However, since they are not been tagged in training images, they cannot be guaranteed to be always effective.
主要標籤: upshirt, underboob
標籤以生成看起來在襯衫裡的效果: hiroki (yyqw7151), see-through, from below, lifted by self, looking at viewer, looking down,
see-through leotard
共通的 trigger 為 1girl,1boy,sex
配合from front可以生成正對女性的視角,from side 同理
sex from behind,pov 可以生成 pov 視角
from normal 可以生成常規的視角
breast grab, grabbing from behind ??
但是我發現很多 LoRA 不加到 Prompt 裡, 而是只用 Prompt 也可以獲得類似的效果甚至我覺得畫的更好了。
對於 see-through leotard 來說可能就是裙子是否透視吧。