Kunaboto Style Lora.
"But we've already got 1000 Kunaboto Style Loras!"
"Good, now we have 1001! I drink whiskey, beer, and coffee. I even eat chocolate."
*As with any IllustriousXL or NoobAI XL models, you can add the latent knowledge that the model has of the artist with the corresponding tag. In the case of Kunaboto, that tag is "kunaboto". This will increase the strength of the style, but it may introduce change in detail or colour.
Update:
This model has been, by a large margin, my best performing one. It didn't necessarily need an update, but I've decided to make one. Dataset was refined by removing repetition, clearing censoring bars, getting rid of text bubbles, and adjusting some colors. Also, it was made as a LoCon.
Since I had the dataset prepared I ran it through CivitAI's training with their recommended settings for Illustrious.
Now that there are more (and better checkpoint mixes) I will also add that version since outputs can be good. Noob still has it ingrained, and I think it's working fine, but for those of you who really want a focus on the style, here it is.
The next lines of text were my initial reaction. I leave them, but with a strikethrough so as to be obvious that the statement is no longer true. History should never be erased. Have fun!
The results are not even close to what the Pony-trained one outputs, so I'm not going to upload it. Maybe at a later stage, and maybe with NoobAI as a base, I'd try on my own machine, but for now, the civit's training result gave me pretty much 2-3 very good image results, and all the rest (~20-30) were terrible.
Description
Dataset cleaned from censoring, text and text bubbles. Also, trimmed from repetition as much as possible, and last but not least trained as a LoCon rather than a LoRA.
FAQ
Comments (9)
Amazing.
Think you can recreate this art style as a lora?
I could give it a try. It's AI though so decent chance it might turn out bad, since training AI on AI content can be tricky.
Meantime, try this one: https://civitai.com/models/707597/pulenkompot-androjuniarto-style-ponyxl
Do you plan to make an illustrious version in the future as a test?
I have a Noob 1.0 version that works quite well with the right settings. I'll probably upload that soonish
Are you planning on releasing a Noobai/Illustrious version?
I made one for both IL and Noob 1.0. Then Noob 1.1 came out. After some testing I'll see which is worthy of uploading, and whether I should retrain on Noob 1.1.
But keep in mind, the "kunaboto" tag works very well for Noob and mixes based on it. Give it a try if you haven't.
@PulenKompot The kunaboto tag definitely gets the faces, shading, and coloring down, but a lot of other stuff isnt present without some kind of lora, such as the large nipples or the musclegut he's known for, at least when I tried (I used pornmaster and obsession mixes, obsession seems to work the best from my experience), there is 1 NoobAI kunaboto lora, but I find it doesn't work great and seems very inflexible compared to the one you've made for Pony.
Faces as well seem to suffer from 'sameface' syndrome, tho this may be remedied by adding IP tags depending on the character you're prompting, I haven't tested this.
Loras made for Noob 1.0 should work fine on 1.1 as far as I've seen, could always simply release the 1.0 version then either make a 1.1 version later or wait until Noob has another major update. ^^
@WildFire22 I see. The IL and Noob 1.0 Lycos I made have the same dataset as PonyV2. After the testing both IL and Noob have pros and cons. IL probably more consistently outputs the "musclegut" you mention, as well as the very characteristic pubic hair. The nipples are great on both versions. I could upload both, but it feels like that'd be a bit of an oversaturation of versions.