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
It is a character LoRA of Senko-san from Sewayaki Kitsune no Senko-san.
While v3A (which I personally preferred) 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 of example 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.
FAQ
Comments (9)
Update is appreciated!
I'd really suggest including more variation with your preview images to show the full range of it. Like different outfits, and different checkpoints/style LoRAs.
any protips for inpainting senko's tail when things go wrong?
@waifuenjoyer sure. Try to paint over extra tails with background and draw a single correct tail with simple colors. After that use Inpaint for that area.
https://i.imgur.com/LOPYnIT.jpg
Here is an example how I fixed troubles with the tail using this method:
1. Original image I got from txt2img
2. Image after some rough fixes in Paint-like tool
3. Image after applying Inpaint
@NeuroSenko thanks! really helped a lot here.
Hi, for some reason I can't use a model with this character, I do everything as usual, namely, I throw the downloaded file into the "models" folder, but when generating it, in any case, it shows a gray square
I'm sorry to hear that. Could you provide more details? I see you are registered only for few days here so I'm not sure if you are familiar with LoRA models. So some of these questions may look stupid:
1. Do you place this model into "models\Lora" directory and use it by syntax like <lora:senkoLora_v4:1>?
2. Have you tried to use other LoRA models? Do they work as expected?
3. Do you use Multidiffusion Upscaler? I heard it has some bugs which could lead to gray squares. Or you just try to generate 512x512 square without any upscale and see this issue?
4. I saw this issue was fixed for someone by using this command in webui-user.bat: "COMMANDLINE_ARGS= --no-half --precision full --no-half-vae --opt-sub-quad-attention --opt-split-attention-v1" Maybe you should try to play with these startup params...
Unfortunately I'm not very familiar with such kind of issues so it's a bit difficult for me to understand why this issue happens for you. But I can try to help in case you provide more info.
@NeuroSenko Hi, thank you, I put the model in the catalog that you specified and everything worked!
@Kushalka You are welcome.
Thank you. With its help you can make a fairly accurate cosplay of Senko with the promts miko, japanese clothes
Details
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.