So, I tried to make a multipurpose lora with images of various boots and in the end it was ok enough, but I was curious to see how the lora I made would stack up against a wildcard which ended up being more dynamic with generation and saved me from training and tagging hundreds of images, which I do plan to do just for fun and see what's the difference.Anyhow, wildcard is a good start, works much better with flux as prompts the wildcard manages are more suited to it, there's still a bit to do and test for v2 as well as making it work at least as well for Pony.

Boots dynamic wildcard

Good starting point, still need to define a few things and add stuff, which I'll won't be able to do until 2025.

Files with SHA256: 890b396ff603a11926e4929b3a6c06a0f5374bd8eda227a560f7da17bc36ae7e