Saeki Kaori from Bible Black is on NoobAI-XL (V-pred / Epsilon Illustrious) now!
PonyXL and SD1.5 versions are also available!
TAGS:
Main:
saeki kaori, 1girl, cowboy shot, solo,
blonde hair, long hair, high ponytail, purple eyes,
School uniform:
school uniform, collared shirt, white shirt, vest, suspenders, red skirt, pleated skirt, miniskirt, black thighhighs, red necktie
Casual outfit:
miniskirt, white panties, pantyshot, sleeveless shirt, solo, taut skirt, black thighhighs, turtleneck, white shirt, bare shoulders, no shoes,
Swimsuit:
two-tone one-piece swimsuit
Doctor Saeki from Bible Black: New Testament:
1) aged up, mature female, lab coat, grey shirt, collared shirt, purple necktie, pencil skirt, black skirt, black thighhighs
2) lab coat, crop top, black shirt, midriff, pencil skirt, black skirt, black thighhighs
NoobAI-XL quality prompt:
For NoobAI-XL Epsilon:
Put "masterpiece, best quality, good quality" in positive and put "low quality, worst quality" in negative.
If you have over-constrasted picture and cfg 4-5 didn't help you can try to put in negative "colorful" (Epsilon only)
For NoobAI-XL V-pred:
Check my guide about V-pred image generating!
The best model to use V-pred is Cyberfix! But pure NoobAI-XL V-pred is also good.
For Illustrious mixes and NoobAI-XL Epsilon: use NoobAI-Xl Epsilon (especially mixes based on NoobAI-XL Epsilon!).
PonyDiffusion v6 (XL) quality settings:
Style settings for PonyXL before prompt (important):
score_9, score_8_up, score_7_up, score_6_up, uncensored,
Negative prompt for PonyXL (recommended):
dutch angle, censored, watermark, jpeg artifacts, muscular female
SD1.5 quality settings:
masterpiece, high quality
in negative: worst quality, low quality
Upscaler / adetailer settings:
Hires.fix: Upscaler foolhardy_remacri (download and put in "ESRGAN" folder in "models" if you are using automatic1111 or forge).
Hires steps: 16-18
Denoising strength: 0.2-0.3
Adetailer: denoising strength 0.3. Detection model confident threshhold: 0.7-0.8.
Best sampling method is euler a SGM Uniform, 25-40 sampling steps. cfg 5-6, You don't need any DPM++.
For V-pred Euler A CFG++ (Cfg 1-1.5) is also good choice.
I will be glad to receive feedback from you and a demonstration of your art with my LoRA!
Description
SD1.5 animefull version.
FAQ
Comments (10)
Brother, this model illustrious no NoobAI model
What do you mean? These are NoobAI models - I trained one on NoobAI V-pred and another NoobAI Epsilon. NoobAI epsilon lora has reverse compatibility with Illustrious but can be used on noobai-xl epsilon too.
@Enigmata ok
Whenever I try to use the NoobAI model the end product looks like a very watery sketch. I can't find a workaround for this - any help would be appreciated
You need aestetic tags for positive and negative prompt which I use. You can find full list on noobai hugging face page.
Also I recommend euler A sgm uniform sampler or euler ancestral CFG++ with CFG 1.2-1.5.
Also if you use V-prediction model you need to enable rescale CFG or its alternatives (I mentioned in my article). If you use site... so it is better to use epsilon version because I think thousand years will pass before they add simple node to their comfy workflow - "rescale CFG". I saw stabilizer for noobai but it is not so good as rescale CFG.
Also I recommend to use cyberfix versions of NoobAI-XL (V-pred and epsilon) because they have better anatomy and more stable style (never met watery sketch problem on cyberfix). I hope that auction system will give chance to generate pics on cyberfix because shitmixes of premium users are awful.
@Enigmata Thank you for the swift reply and help - I'll try it when I get the chance to.
@onitensei94 btw if you don't have euler ancestral CFG++ you can use euler a SGM Uniform with CFG 5. It is also good.
New Testament???
She is. I made 2in1 version.
Use 2 last recommended prompts for it (they starts from "aged up, mature female...". It is for her adult version from New Testament. You can see her adult version in my preview too.
@Enigmata Thank you very much, I hadn't noticed.