馃寱 LUMA
This is anime-based semi-realistic mix of checkpoints and loras tweaked for good results on simple prompts.
Model is fully compatible with all LoRA that I made and will make in the future.
All the content I make is and will be completely free on CivitAI.
But if you want to support me with a coin, then you can do it here.
Luma A3 VAE Style and Artist List (please note, that some of styles need additional tags like "retro artstyle, faux traditional media, painting", etc):
Please note that tags "3D, REALISTIC, ANIME" in positive or negative may change look of your images significantly.
I also higly recommend to add "score_5, score_4, ps1 screenshot, 3d, niji" in negative prompt, if you're not going to use it in positive.
Also note, that i added somme extra tags for concepts.
For example:
altshaped - will help you make humanoids with alternative anatomy
overdesigned - will help you make some complex designes if your prompt assumes it.
Complete style list:
123wsty
ahriman, anime
albrecht d眉rer
alex ross, retro artstyle
alexkonstad
aloija
amadmir
anato finnstark
anbudehuang
anime
arcane
asahiart, automaton, altshaped
boris valejo
brenda burke
burda, grungetech
castlevania, anime
cher-ro
cherrygig
chibi herowars
claude monet
cmg
darkest dungeon, high contrast
dashiana
degenesis
demessance
deus ex, cyberpunk
disney, flat color
dkeeno
drakonoart
dravacus
dreelrayk
eddie mendoza
edgerunners, anime
elden ring
fanfoxy
galyosef, 3d
ganas
greg rutkowski
hades
hearthstone
henry asencio
hizipro, high contrast
ikenaga yasunari
indus
ivan kramskoy
jandrew, scenery, environment
jeremy mann
jim lee
john william waterhouse
jojo
jose parra
jparked
junejenssen
kael ngu
krrat, overdesigned
kylepunk
laurelin
leicai
lol
lovk1y
makoto shinkai, style
mangamie
marta nael
max kostin
mengxuanli
mixppl
mohrbacher, divine, altshaped
mtg
nadar
nicolasaviori
niji
nixeu
othalam
painting \(medium\)
papsuev
peterskore
photo, realistic, cinematic
pixar, 3d
ps1 screenshot
qichao wang
rakavka
reliah
retrofuturism syd mead
saiya-art
sanchare
scenery, environment
shinjuku
smite
stephen pan
stumpyfongo
sugarpunk, cyberpunk
tck, altshaped
templarpainter
tixnen
tomas duchek
trufanov
valentin serov
valorant
vegera
viktor vasnetsov
vincent van gogh
warcraft
wayfinder
wlop
wroniec
wushenyou
yumingli
yuriiiko, flat color
zack snyder style, movie still, cinematic, realistic
zexiguo, overdesignedPDXL Usage:
Clip Skip 2
Sampler: Euler a / DPM ++3M SDE (SGM Uniform) (or any your favorite)
Steps: 30-40
CFG Scale: 5-7
Hires.fix: x1.5
Upscaler: 8x_NMKD-Superscale_150000_G for realism or R-ESRGAN 4x+ Anime6B for illustrations (Denoising strenght 0.2-0.3)
Hires steps: 15
Never use Face Restoration!
If you want better faces just use AutoChar or AfterDetailer.
Useful tags (negative):
wrinkled skin, double head, mutant, bad anatomy, fewer digits, bad hands, extra limbs, twisted, cropped, out of frame1.5 Usage
Recommended embeddings:
1. easynegative
2. easynegativev2
3. bad-hands-5
4. badhandv4
Recommended extensions:
1. AutoChar
2. AfterDetailer
My regular settings:
Clip Skip 2
Necessary negative prompt: easynegative or easynegativev2
VAE: kl-f8-anime2
Sampler: any
ENSD: 31337
Hires.fix: x1.5 and higher
Upscaler:
Option 1: 8x_NMKD-Superscale_150000_G for realism or R-ESRGAN 4x+ Anime6B for illustrations (Denoising strenght 0.3-0.5)
Option 2: Latent (any) or Lanczos (Denoising strength 0.5-0.6)
Never use Face Restoration!
If you want better faces just use AutoChar or AfterDetailer.
In 0.9X version you may use tags like 3d, realistic in positive or negative prompt to increase/decrease realistic shading. (Check sample images for examples)
But you may feel free to experiment.
Please note that this model is horny sometimes, so I strongly recommend to tag some clothes (pants or something) and using these tags in a negative prompt to avoid NSFW: nsfw, naked, nude
Useful tags (not necessary, just minor tweaks): volumetric light, realistic, dark atmosphere, volumetric light, caustics, absurdres, eye focus, contrast, dark scene, bright scene
Description
Upgrades, people, upgrades:
Completely reworked lighting, volumetric and caustic effects!
Much more small details on surfaces and textures!
Much cleaner and sharper image!
Baked VAE!
Slightly more realistic shading.
Even less horny.
Better lora compatibility.
Slightly colder gamma.
Easier prompting and better results.
FAQ
Comments (22)
Is the VAE included in update or do i need to download seperately
VAE is included in new version, i noted it in version info.
@sadxzero聽thanks for some incredible models and loras, you are an absolute star, I hope more will come of disney, wow and style as you are SO talented <3. Regarding 0.98, would it be possible to also have a non baked vae version? I have a set of VAE and I like to experiments with them on the models :) only if it's possible and won't ruin the model. cheers!
@zx96聽Oh, thank you for such flattering words, i really appreciate it. About version without VAE: am i wrong, or models with baked VAE working with any other VAE too? I mean, you may choose any VAE you have, and it will work fine, but when you set VAE to none model will use baked VAE instead. I maybe mistaken here, but i think it should work in this way.
@sadxzero聽oooohhh thanks for the explanation, that means i can try that. The other thing they say is that if it has a baked vae, the lora may not work too well on it, but maybe I am just being paranoid. Congrats for the stellar work once again and all your loras <3 !!!
Hello!
First and foremost, i wanted to thank you for this incredible model. The quality is amazing and arts made from it are such a joy to the eye.
I'm new to exploring Stable Diffusion and automatic1111, and i've been trying to recreate some of your arts for better understanding on how the thing works. For example, i've picked up your elf art and tried to build it myself using all the settings you've posted. I've almost managed it
but the eyes keep messing up. Is there something that i'm missing, or is there any trick to it? Thanks!
Hi there! Thank you for such sweet words. About eyes, i noticed that you are using face restoration, please turn it off and use it only never or almost never. Model itself provides good eyes, which you can improve in inpaint if it's really necessary. But for now just turn off restore face option.
@sadxzero聽Holy fucking christ bro thanks for the tip it actually worked!!! The eyes are working!!!
How did you recognize that my version has a face restoration on it? Pretty sure i've sent you a PNG and you can read generational data through JPEG.
Thanks for the answer!!!!
@moodybluesrip598聽Hehehe, glad to help, mate! No, can't read data from your pic, but when i zoomed in i recognised some stuff from face restoration like bad eyes, flat lips and poor skin texture and flat light on face.
@moodybluesrip598聽On top of removing face restoration, the best way to increase facial quality is to upscale the image, and then inpaint only the face at the new higher resolution with like 35-60 denoise depending on how much you want it to resemble the original. Although do note that this will take a long time on older GPUs.
@Alyxra聽Thank you for the advice, much appreciated!
I've been in love with Luma ever since I started using it not too long ago. I'm in equal parts excited to try out the new version, as well as terrified at the thought I might lose the quality of 0.97.
The solution is, of course, to make space to have both versions.
Yeah, i remember your feedback and kind words. Thank you for your loyal passion to this model! About new version i can say that it's not huge improvement of 0.97 but more-like spin-off with some minor tweaks. Actually, at first i was concidering to make a separate model page with this model, but then decided to continue numbering. So 0.98 is definetely more detailed than 0.97, but it's still not as good as i want it to be. So i guess i will make some new iterations on way to 1.0 version.
And btw, now i'm thinking about extract some data and mix with loras, which i trained special for 0.98. And i hope i will make something similar to 0.98 look but in lora form. I already made some moves in this way and now i need to test it all.
It's fantastic. I like .97 better for some reason though
horny is good. we like horny. embrace the horny
this model is perfect, but the hands' form is kinda bad right? Hope this could be improved 馃挴馃挴
Can you add the 0.98 version without vae?
Sorry, don't have one anymore.
This is one of the best models out there, absolutely amazing art-style! One question though, I've been having some trouble with sex scenes, doesn't seem like it works too well, any thoughts?
Hey, great model, my personal favorite on the site. Btw, sorry, don't know where else to comment about this, but why did you delete your post about dpm++ 2m alt karras sampler? I saved the site so I can download it later, but now its gone, what happened?
Hi there, thank you for your feedback! I unpublished and archived it, cause there was too much people saying that it's not working and stuff. I'm kinda tired lately of guiding guys on basic stuff and pin point every mistakes, so decided just to unpublish it. But i may bring it back if you want.
@sadxzero聽I saw you republish it, thank you!