If you like my work, drop a 5 review and hit the heart icon. it's free and keeps me motivated
All my models are officially hosted and maintained by me on Tensor.art . use my Exclusive and public model for free on tensor.art
06/06/25 : v2.0p released
Main focus of 2.0b is cleaning the style influence of v1.5
19/02/2025 : On-site Generation added
Private commission on fiverr
Get early access to my upcoming NSFW Lora on my Patreon.
early lora share on discord for patreon.
I was hoping you could support my work by joining any one of them and get early access to all my upcoming loras and other perks, such as fan requests and a Discord role.
Join my Discord Server
check the images for prompts
use lora at 0.7-1
Adetailer for faces
Img2img upscale
4x-ultra sharp
comment you idea or request
Suggested settings:
I had CLIP skip 1 on every image
Samplers : Eular A, DPM++ 2M
CFG : 5-6
I had ENSD: 31337 all of them
All of them had highres.fix or img2img at higher resolution.
I don't use restore faces
I use Adetailer for face details
4x-Ultrasharp upscaler
positive: masterpiece,best quality,amazing quality,absurdres,
negative: bad quality,worst quality,worst detail,sketch,censored,watermark, signature, artist name
Join my Discord Server
Description
FAQ
Comments (31)
So... what's the difference between 6 and 5?
We need a changelog!
Автор не считает нужным отчитываться. В чем-то он прав. Так что просто делай как я, скачай несколько версий и на одном сиде проверь.
@Gipno тратить время чтоб скачать и проверить когда автор может просто написать краткий changelog?
@aesthetic_factory ну вам никто не обязан это писать. Как я уже сказал, автор не считает это нужным. Тем более вам жалко потратить 1 минуту, чтобы скачать модель. Не думаю, что это проблема, скачать и сравнить несколько версий модели.
@Gipno я не спорю, да автор ни перед кем не обязан чем либо. но знаете ли, не у всех есть высокоскоростной интернет чтобы скачать 6гб данных за минуту, или вообще у кого то может быть ограниченный трафик. никто не требует от автора что то, он мог потратить 2 минуты своей времени на текст, просто чтобы во уважение и удобство своих фолловеров которые его/её поддерживает.
@aesthetic_factory You don't even follow them https://civitai.com/user/aesthetic_factory/following
@nvo76 я не говорил что "я фолловер"
@nvo76 чел, читай все комменты
Bro, please bring back the style from version 1.0, but make the model that does less mistakes and reading prompt better. I am missing time when this checkpoint was one of the best :*3
v1.0 definitely was the best. Everything after that has been basically a new worse model in a different way, and not a progression.
I agree 5.0 is a huge downgrade. Colors and everything are too strong.
It looks more like WAI: lesser fake-3d, and more randomness, and vibrant.
It looks like one of those things when you put an incorrect negative embedding, it becomes strangely vibrant.
What's new on V6?
I don't know what you are doing, but it's not working. The older versions are way better.
V6 forgot about the aglaea \(honkai: star rail\) that knew from V5.
I think this v6 don't know characters after February 2025.
v6 is a downscale in my opinion, not much different from v5 and probably older dataset since it couldn't recognize aglaea which released feb 2025
is there a version that looks like pony v2 cleaned?
Seems to love adding people to the background even with "solo" in the prompt
Still having this issue on version 7, especially when using regional prompter.
You need to fix the merging of characters into one and have fewer artifacts. If this happens, this model will become the best.
v3 and v4 are clearly the best in general terms. v3 is the perfect midpoint between v4 and v6. V6 is a little bit more sharp and v4 has less contrast and overall a creamier look.
v5 is complete trash ngl.
Keep up the good work tho, creator!
recommended VAE? or not needed?
its a checkpoint dude
i think XL_VAE_C F0.51 is good
@Meowzilla yes of course its a checkpoint thats exactly why i am asking. some checkpoints have a vae baked in and others do not have one baked in.
@nomorepublicindecency thank you! :)
Im a big fan of Neptunia if you have advanced compatibility turned on.
v4 seems best
Idk why people say V5 is trash. I compared 4, 5, and 6, and V5 has the most beautiful images.
Both V4's and V6's colors are too muted, while V5 is nice and vibrant.
The face and eyes are also sharper and more defined in my experience.
The only thing that is slightly worse than V6 is the realism and anatomy realism, but it's only a small difference, and honestly, who cares about high realism in anime style.
Agree, and for merge seems to have the best CLIP and prompt adherence.
Details
Files
Available On (6 platforms)
Same model published on other platforms. May have additional downloads or version variants.