This model is designed for use with the embedding, it has been trained using OneTrainer with pivotal tuning.

I trained this LoRA using screenshots from an RTS videogame to reproduce scenes of battle.

• fbf - focuses on armies of undead, abominations are particularly present

• fbf_ud - focuses on armies of undead

Notes

• ComfyUI workflow for the samples generation should be embedded in the images, in case you want to check more settings.

• It's intended to use with this embedding, it's been trained with OneTrainer with pivotal tuning!

Changelog

fbf (v1.0-beta)

This version reflects more what I wanted to create but controlling the bleed-over is hard...

I think the next versions will be centered around one particular type of unit.

fbf_ud (v1.0-alpha)

I made quite a few mistakes training this LoRA, which was intended to reproduce big armies of undead fighting against humans...

Still, I think it can produce fun and weird results, so I decided to share it anyway ^_^

For the next version I'll probably tune down humans and remove captioning the undead (any advice is welcome!).