It is a character LoRA of Senko-san from Sewayaki Kitsune no Senko-san.
While v3A (which I personally preferred before I made v4) tends to work good only with some specific models, v4 seems to be more universal. I still couldn't get any good result with 2.5D models but maybe it's just because I didn't have any experience with them.
I had to delete tags from metadata of preview images because they contain some words civitai doesn't like. Here are couple examples of images which contain PNG-info:
https://files.catbox.moe/tfu9h5.png
https://files.catbox.moe/g4uj7n.png
https://files.catbox.moe/tvxd9d.png
My favorite model for this LoRA is Anything v5 but it should work fine with other 2D models.
I tried to train new version of the LoRA using different base models, but ended with that original NAI works the best as a base model for learning.
Description
The model was converted to safetensors format and size was reduced to 144MB. I recommend to use this LoRA with weight about 0.8
Style changes comparing with version 1 should be minor.
FAQ
Comments (3)
Works pretty great by itself but if you try anything with another lora it just does not wanna cooperate unless I go to like 0.5 or sometimes 0.6. Sadly, at that point it looses a bit of coherency. Is this intentional?
Edit:
After some more testing I've found it's not this lora that is at fault, it's my prompt or rather the position of the prompt. Lora is alright.
Could you send me a links to loras which work bad with this one please? I'm working on another version of that lora and would like to check if the issue is still here.
Also I saw there is alternate lora for Senko-san which was published couple of days ago, you can try this one:
@NeuroSenko Sadly the other one is even less flexible and sadly doesn't reach the fidelity yours has. I've had no luck with the headpat one from here, the zankuro style one does not work at all, even going to 0.5 and below. The covering eyes lora from here works pretty well though just as an example. It's rather odd, if you want I can upload you an example.
Figured it out myself :)