I make them myself, that why their quality is pretty random. I will try to make the sisters

Why not, I have another thing to do right now, but I will eventually make it somewhere this week.

@Lancerx 

well, I can try

@Lancerx 

I don't really watch tutorial so I don't have any to recommend, but I can tell you how I do.

----------------------------------------------------------------------------------------------------------------------

well it take between 10-15 minutes.

some time I have to repeat that processes 3-4 times to get a good result

I can recommend this guide for collab:

https://civitai.com/models/22530/guide-make-your-own-loras-easy-and-free

if you take this Lora for example,

I had 31 images, so i used 10 epoch(2000 steps), if I Had used 50 images it would be 8 epoch(1600 steps)

also, I use 10 epoch for any dataset with 40 and less images

here the link for the dataset_config:

https://drive.google.com/file/d/1-45pSJ3Lu1DTfJ08j28wjFJ6ltHy84HR/view?usp=sharing

here the link for the Lora_Training_config:

https://drive.google.com/file/d/1--8keVVWUNtSytLJ4QOlpcBtuUqNkzxj/view?usp=sharing

----------------------------------------------------------------------------------------------------------------------

To start most of my lora are trained on several images ranged from 22-42 Images and in some special case (mostly anime), on 100 images.

When I'm creating a Lora, a search for any image from the show (like png or artwork)?

And I add some fanart (just a few).

Most of the fanart have a style that it's close to the show style.

Sometimes it's very hard to do when the show is not widely known, I have very hard time finding good images. In those case I had to go to some unholy website to find some of the art, but the least of those in your dataset, the better.

----------------------------------------------------------------------------------------------------------------------

notes:

1-More images don’t really mean a better Lora, it really depends on what is the subject.

2-No-human (furry, animals) subjects are more complicated to make, some time you will need to restart your training multiple times.

3-When you are training your subject, put in the dataset multiple “full body” images of your subject.

4-If your subject is a character that rarely change outfit, you should make sure that the dataset of said character doesn’t have too much different outfit.

5-upscale your training images if their quality is low.

Maybe, some styles are more difficult than others.

I will try again to see if it can work