What Is It?

This LoRA is trained to provide a dark, dramatic, fisheye, music video style to your realistic image generations. It has been tested with standard 1girl prompts but not tested with NSFW, or groups. It may have effect on faces, and the image style will lean towards model type poses. The fisheye lens effect has a tendency to affect scale and can cause giant women sometimes. As always I did not cherry pick the showcase images, so natural errors are left in.

LoRA vs no LoRA

an amateur photo of a polish woman, 5'8, 110lbs, wearing an unbuttoned white blouse, in a dark room, slight fisheye, low-angle

How To

All of the samples provided use only this LoRA and Face Detail node.

• euler/beta @ 32 steps OR

• res_3m/kl_optimal @ 28 steps.

• Model Strength: 0.80

Suggested Weight: 0.4 - 1.0

Training Notes

Trained with 18 uncaptioned images at 768, 1024. No effort was made to train NSFW concepts.

While I have been able to use this with no keyword applied at all, psxrrstyle will enforce the effect, using prompts like 'fisheye lens', 'dark room, harsh shadows' can be helpful.