Please share what you generate with this, feedback is helpful so I can iterate on this LoRA!

What Is It?

This LoRA is trained to stylize images of women for soft-modeling photos. Photos that would leave something to the imagination rather than hardcore explicit poses. This brings in more candid lighting, more varied faces and natural playful poses.

How To

All generations in the samples are used with this LoRA applied only, and a face detail node. Generated using euler/beta @ 32 steps OR res_2s/bong_tangent @ 20 steps. You may notice at strengths near 1.0 and higher the model can cause some issues with hands, and distorted people in backgrounds. I've left these in the examples.

Suggested Weight: 0.4-0.8 produces very natural results, strengths above can produce great results inconsistently.

Training Notes

Trained with 19 uncaptioned images at 768. No effort was made to train NSFW concepts.

While I have been able to use this in ComfyUI with no keyword applied at all, psxzstyle is the keyword if you need it.

I experimented with bringing this one into Chroma as well as Krea, if anyone has experience with prompting and generating please let me know how I can generate better samples.