phantom_assassin

phantom_arcane

Sorry, I made a mistake and overtrained the model quite a bit. It's still possible to get good results, but you'll need to experiment with the prompt and the LoRA weights. Setting the model strength to around 0.5 is usually enough to prevent it from simply reproducing images from the training dataset.

don't forget to share what you create. You get the buzz and feedback from the community, and I genuinely enjoy seeing what people make with my models.

It also helps me better understand the model's strengths and weaknesses, which makes it easier to improve future versions.