If you ran into any troubles with my mix, please read description below. There is plenty of useful information. Also, you can contact me through message system on CivitAI.
Если вы столкнулись с каким-либо проблемами при использовании моего микса, пожалуйста, прочитайте описание ниже. Там есть много полезной информации. Также вы можете связаться со мной через чат на CivitAI.
Also try: DivineAnimeMix!
Introduction:
All models i've used before first release of DIvineElegance did not fit my requirements. I've decided to do something on my own to fix this issue. There was a lot of different mixes based on Anything v3 or v4 and i've taken it as a base as well.
At the beginning, it was a mix for my personal use, but it was warmly received by Unstable Diffusion discord community and some time later i've decided to share it here, on CivitAI.
Right now Divine series is my favorite models and quite unique thing for me. The main goal of DivineElegance - anime characters with as many details as it could have. Hair, clothes, backgrounds, all of this things should be beautiful, slightly more realistic and detailed.
I hope you also would like DivineElegance, so, stay tuned for newer versions of mix!
Recommendations:
Use danbooru tagging for prompts
Do ->not<- use "masterpiece" or "best quality" tags (v9+)
Sampler: "Euler a" at / 30 steps
Cfg scale: 8
Clip Skip: 2 <---- VERY IMPORTANT
99% of anime models only works with clip skip set to 2
ENSD (Eta noise seed delta) - leave blank
Negatives:
(EasyNegative:0.8), (worst quality, low quality:1.2), blurry, bokeh, dof, fog
Recommendations (Hires Fix):
Upscaler: Remacri
Hires steps: 10-15
Denoising strength: 0.3
Upscale by: any
Troubleshoot:
My colors are bleached out. What should i do?
You're missing a VAE file. For this mix i would recommend kl-f8-anime2 VAE. Just put it into SD folder -> models -> VAE folder. Then go to your WebUI, Settings -> Stable Diffusion on the left list -> SD VAE, choose your downloaded VAE.
If you want to get mostly the same results, you definitely will need negative embedding:
EasyNegative, it's better to use it at 0.8 weight.
Don't use EasyNegative (or other embeddings) with Style LoRAs, it may ruin the style!
Until you like the result, of course.
Don't use a lot negative embeddings at the same time. It definitely will ruin any possible output on all models! If you want to use negative embeddings - use one for quality, and one for hands (if you want a placebo for hands).
Technical support:
If you're ran into any problem with my mix, you can DM me on discord:
troubledarkness - discord nickname.
I'll be glad to help | Буду рад помочь.
Credits:
DivineEleganceMix Recipes (pre-v6)
Really BIG thanks to:
Nigurumi for BasilMix
WarriorMama777 for AbyssOrangeMixes and recipes
andite for Anything4.5/Cocoa/PastelMix
Ikena for Grapefruit
Cyria for ZeroTolerances
devdevice545734 for FaceBombMix
You, for reading, downloading and reviewing this mix
Description
New version has came out, so, lets find out what changed:
Mix:
- Removed Abyss Orange Mix 3 and Cocoa
- New merging techniques, block merge via supermerger
- Now there is a LoRA merged in instead of finetune for less time consumption, mostly the same result
Visuals:
- Less blurriness on foreground
- Sharper and cleaner textures
- More details in backgrounds
- Restored near-original saturation
Characters:
- Restored working concepts for some well-known characters
- Better reaction to unusual poses
Lighting:
- Restored original non-white color for lighting
LoRA:
- Better compatibility with "some" LoRAs
Fixes:
- Fixed "hands behind back" causing from behind view
- Fixed "backlighting" causing from behind view
There can be unexpected issues because of merging model through blocks, but i've not found anything.
Thanks for keeping an eye on Divine series models and thanks for reading this changes, i hope you'll like new version of DivineElengaceMix as well.
FAQ
Comments (21)
v8 Patch Notes.
New version has came out, so, lets find out what changed:
Mix:
- Removed Abyss Orange Mix 3 and Cocoa
- New merging techniques, block merge via supermerger
- Now there is a LoRA merged in instead of finetune for less time consumption, mostly the same result
Visuals:
- Less blurriness on foreground
- Sharper and cleaner textures
- More details in backgrounds
- Restored near-original saturation
Characters:
- Restored working concepts for some well-known characters
- Better reaction to unusual poses
Lighting:
- Restored original non-white color for lighting
LoRA:
- Better compatibility with "some" LoRAs
Fixes:
- Fixed "hands behind back" causing from behind view
- Fixed "backlighting" causing from behind view
There can be unexpected issues because of merging model through blocks, but i've not found anything.
Thanks for keeping an eye on Divine series models and thanks for reading this changes, i hope you'll like new version of DivineElengaceMix as well.
Thanks for your work.
Thank you, I've only used divine since I've started messing with ai images, nothing else comes close in my opinion ! Keep up the great work !
@GospinSin Good to know, the same for me. Was trying something the same as Divine series models for a long time. But most mixes has more pastel style or anime style with semi-realistic asian faces (2.5D mixes).
This SD model is absolutely cracked and the generations it makes are even more incredible. Keep up the great work man!
@Vastide I hope it so, good to know you've liked it.
Seriously unbelievable model. I get amazing generations first try with suggested settings. Getting more experimental lately and it tends to react well most of the time. Nice work!
@Hypotanime good to know! Thanks for reply. I was working on a new update to bring back old visuals i liked with features from newer versions. Also, there would be a lot of fixed in different directions, from backgrounds to character bodies and clothes.
HOHO ! V8 is out !
Hi, yeah, it was released 12 days ago. After a release, i can say that there is some problems need to be fixed in the next version.
1. Overbrightened images (or lighting)
2. Too anime-ish visuals, more similar to original anything model
3. NSFW concepts for naked female characters, for some reason, it's broken a little
Maybe there is something else, but that the only things i've found by myself and by looking for other people works. All of three tasks were done already, just need to complete work on visuals of a mix, but right now i'm not able to do it because i'm going through an examination in hospital until November 21s. It will need a lot of testing after that (maybe 3-4 days), and then i'll try to release a new version.
Thanks for commentary!
Hi, so does the latest mix have Anything v5 in it or is it still Anything v4.5? I ask because the description of the checkpoint page says one thing but the recipe list says another, so I'm just asking for clarification-
Hi, description tells about first steps of the mix, when it was using Anything v4.5, that was troll mix from Andite. Right now i'm using Anything v5 for DivineEleganceMix or DivineAnimeMix recipes.
https://tusiart.com/ https://www.liblib.art/
These two Chinese websites are stealing models from Civitai.
There is even more sites that doing that. Tusiart is a copy of TensorArt as i see, for example. Repost of my models isn't prohibited, according to license, but i want to change one thing later. At least, to credit me, if model is being used in mixes or on other sites. But thanks for pointing that out.
Is the model commercially available?
HI,
If question is about supporting or buying model - no, it's free.
If question is about selling merges, or using on paid generation services - no, as well. There is no such permission granted to anyone. About images - well, i guess only with my permission. But i'll be honest, there is no way to know if anyone uses my model contrary to these permissions.
If i got you wrong, just tell me. I'll try to explain.
@TroubleDarkness So, If I get this model for commercial use?
@fhdrheh explain what do you mean by commercial use at lest.
@TroubleDarkness Предоставление доступа к генерациям через условный патреон(за месячную подписку) считается коммерцией ?
@miys Привет, да, определённо считается. Только будем честны, нет способа доказать, что что-то было сгенерированно на определённой модели без необходимости предоставления автором оригинала сгенерированного изображения с сохранённой метой. Заставить автора это сделать почти невозможно, без предъявления обвинений в наурушении условий распространения модели, а делать это - кринж полный, будем честны. Да и как-то воздействовать на подобную торговлю не выйдет, а CivitAI в этом помогать не будет, вероятнее всего.
Я не хочу чтобы мой контент монетизировался только по одной причине - я делаю его беслпатно и хочу, чтобы всё созданное при помощи моих моделей, LoRA и т.п так же распространялось бесплатно и было доступно ->всем<- у кого есть доступ к генерации изображений. На данный момент, не включаю сайтов с генерациями, так как "работать" с ними - боль в заднице.
Запрет на продажу изображений указан в самой нижней части блока под кнопкой загрузки. Как и другие запреты. Но мониторить их соблюдение у меня нет как сил, времени, так и возможноости в целом. Не думаю, что эта самая возможность есть хоть у кого-то.
Если есть ещё какие-либо вопросы - задавай. Постараюсь ответить.
Надеюсь, ответил более-менее развёрнуто.
@TroubleDarkness Спасибо