banner
HoshinoAya

HoshinoAya

github
twitter

Play with Stable Diffusion

应该都能在 civitai 和 huggingface 上找到, 主要集中记录一些会用到的 LoRAs 的 Prompts。 civitai 的搜索好难用,只有我这么觉得吗

使用的 checkpoints:#

AbyssOrangeMix 2#

很好用的 Model, 只用少量参数都能出很不错的图

How to use#

  • VAE: orangemix.vae.pt
  • Prompts can be long or short
    As simple as possible is good. Do not add excessive detail prompts. Start with just this negative propmt.
    (worst quality, low quality:1.4)
  • Sampler: “DPM++ SDE Karras” is good
  • Steps: forTest: 12~ ,illustration: 20~
  • Clipskip: 1 or 2
  • Upscaler : Latenet (nearest-exact)
  • CFG Scale : 5 or 6 (4~8)
  • Denoise strength: 0.5 (0.45~0.6)
    If you use 0.7~, the picture will change too much.
    If below 0.45, Block noise occurs

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 的改进版,但是很多人还是用的 AOM2, 导致我在 AOM3 上出不了相似的结果。

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 **is As simple as possible is good.
    (worst quality, low quality:1.4)
  • Using "3D" as a negative will result in a rough sketch style at the "sketch" level. Use with caution as it is a very strong prompt.
  • How to avoid Real Face
    (realistic, lip, nose, tooth, rouge, lipstick, eyeshadow:1.0), (abs, muscular, rib:1.0),
  • How to avoid Bokeh
    (depth of field, bokeh, blurry:1.4)
  • How to remove mosaic:
    (censored, mosaic censoring, bar censor, convenient censoring, pointless censoring:1.0),
  • How to remove blush:
    (blush, embarrassed, nose blush, light blush, full-face blush:1.4),
  • How to remove NSFW effects:
    (trembling, motion lines, motion blur, emphasis lines:1.2),
  • Basic negative prompts sample for Anime girl ↓
    • 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,
  • Sampler: “DPM++ SDE Karras” is good Take your pick
  • Steps:
    DPM++ SDE Karras: Test: 12~ ,illustration: 20~
    DPM++ 2M Karras: Test: 20~ ,illustration: 28~
  • Clipskip: 1 or 2
  • CFG: 8 (6~12)
  • Upscaler : Detailed illust → Latenet (nearest-exact)
    Denoise strength: 0.5 (0.5~0.6)
  • Simple upscale: Swin IR, ESRGAN, Remacri etc…
    Denoise strength: Can be set low. (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#

作者用超过 10k 张 NSFW 图片训练的画色图的 Model, 不幸的是用力过猛了,所以又混合了一些别的 Model, 包括:

  1. Hassanblend by sdhassan (0.15)
  2. Anyhentai by asdpro123 (0.1)
  3. AbyssOrangeMix2 - Hardcore by Havoc (0.15)
  4. 7th-anime-v3 by syaimu (0.2)
  5. Kenshi by Luna (0.15)
  6. Grapefruit by Ikena (0.1)
  7. ColorbombMix by mocker (0.1)
  8. Counterfeit v2.5 by rqdwdw (0.1)

以及:

  1. Moxin by simhuang to enhance face and style.
  2. Fashion Girl by me to enhance face.
  3. Akiryo's Mai by me to slightly finetune face, body shape and skin texture.

使用的 LoRAs:#

很多内容复制于位于 civitai 的原始链接

garterbelt
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.

Major tags: upshirt, underboob

Tag to gen looking inside of shirt: hiroki (yyqw7151), see-through, from below, lifted by self, looking at viewer, looking down,

see-through leotard

共通的 trigger 为 1girl,1boy,sex

配合from front可以生成正对女性的视角,from side 同理

sex from behind,pov 可以生成 pov 视角

from normal 可以生成常规的视角

breast grab, grabbing from behind ??


但是我发现很多 LoRA 不加到 Prompt 里, 而是只用 Prompt 也可以获得类似的效果甚至我觉得画的更好了。

对于 see-through leotard 来说可能就是裙子是否透视吧。

Resources#

Loading...
Ownership of this post data is guaranteed by blockchain and smart contracts to the creator alone.