Style LoRA of Tottotonero (とっととねろ太郎)
Like the style? Give the artist some support!
Clean two dimensional style with round eyes and puffy bits.
Drop a ❤️ & Review if you like what I make!
Your posts, reviews and comments motivate me!
Training Notes
Base Model: Pony Diffusion V6 XL
DIM / Alpha: 16 / 8
Training Batch Size: 1
Training Resolution: 1024
Optimizer: Prodigy
Image Count: 259
Steps: 1900
This LoRA can only work on Pony Diffusion V6 XL or models derived from it.
Description
FAQ
Comments (4)
Brilliant ❤️🔥
Nice, thanks :) is there a reason why you remove the character tags?
Hello!
I did so because for models that do not have a concept associated with the character's tag, the training might cause the tag to end up as an activation tag for the style, which is unintended.
In the case of pony and its derivatives, as it has a large number of characters learnt, I'm not sure if it is beneficial to leave the character tags in, so I have just stuck to pruning them.
@r3c I see, fair point. Will have to investigate this. I have seen very good style replications in pony with concepts and character preserved - though they might be less adaptable. The reason I'm asking is because it would obviously be better to not only get the style but also typical concepts & characters associated with the artist if it's free. Pruning them loses this benefit but then again you might be right that the resulting model would be less cooperative delivering the style if other characters and concepts are used.
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.