I’ve created a few loras in the past but I haven’t in a while. So I’m thinking if anyone has something they’d really like to see and want to assemble a collection of input images, I’m happy to run the training. I’m the most familiar with photorealistic models, so something derived from those would probably be best.
Obviously nothing even remotely underaged, I’m also not super interested in doing anything scat/gore related. Otherwise, I’ll whip one up and we can see how it goes. No promises that it’ll be amazing, but it can be fun to try.
As far as the tagging goes, if they come untagged I’ll use a combination of blip/deepbooru.
Not gonna remove this or anything yet, but I’d appreciate if you kept things like this to the discussion thread in the future.
I’m very new to this. I just installed some software the other night. What’s a lora?
In simple terms, it’s a small modification to a model which usually produces some kind of specific change, commonly an art style, a pose, a hairstyle, something like that. This link describes it in more depth: https://softwarekeep.com/help-center/how-to-use-stable-diffusion-lora-models
You can find lots of them here: https://civitai.com/
If you see a ‘Lycoris’, it’s like a fancier lora.
This is really cool. One of the first things I tried to do when image generation came around was to make a story book, but the characters changed too much between prompts. Loras seem to fix that.
