hyperfusion LoRA 550k images

This LoCon LoRA was extracted from the hyperfusion-v7 finetune instead of trained on its own. It does seem to be more consistent than v6.

Because it is an extracted LoRA I had to keep the dim at a larger size to preserve its concept knowledge, sorry for the larger than usual file size.
*I am limited by the technology of my time*

Training details are under the finetune link above.