/upscale
(sharp focus:1.2), photo, attractive young woman, (beautiful face:1.1), detailed eyes, luscious lips, (eye makeup:1.2), body freckles, (medium breasts:1.0), (tight body:1.2), (dark hair:1.2), wearing (slitdress:1.2) in (red carpet event:1.2). (moody lighting:1.2), depth of field, bokeh, 4K, HDR. by (James C. Christensen:1.2|Jeremy Lipking:1.1).
I like the results of this one better... it does provide some good playboy centerfold pose base.
Same dataset as v1 but with 3 times the training steps and slightly lower TE and UNet LR.
Sample at from SD1.5 base and from protogenX53... to show how the LoRA can be used with different 1.5 based models.
Use kohya-ss/sd-webui-additional-networks (github.com) in auto1111 tu load the LoRA model. You can combine it with other LoRA for blended poses.