I can guide you with the best practices, tips, and tricks to achieve your goal in your Stable Diffusion training. Take a look at my Patreon, subscriptions for just $5. Exclusive content. Patreon. :)

About:

The purpose of this LoRA is to (try to) replicate a sideview of a Blowjob. Be careful, may generate body horror, disembodied penises and weird things.

It was trained in multires (512x768, 576x768, 640x768, 704x768 and 768x768), using simple captions and with 106 high-resolution images.

The demo images of the model aren't edited, but they are cherry-picked and generated using hires-fix (0.15 ~ 0.45 denoising).

Suggested Settings for Inference:

Model: analogMadness_v70 (but works with 97,2% of models)

Positive: a woman is sucking a cock, cute asian, short hair, wide open mouth, full facial shot <lora:bj_from-side_v1_by_IsnAI:1>

Negative: morbide-body-parts, wrong eyeballs, low res, lower quality

Sampler: DPM++ 2M Karras

Steps: 32

CFG: 6

Resolution: 512x768, 576x768, 640x768, 704x768 or 768x768

Note¹: You can get great image variations if you tweak resolution a bit.

Note²: In this version, it's not necessary to reduce weight, it's perfectly possible to generate the images using weight 1 in LoRA (all demos use weight 1), but maybe someone with more patience can discover some interesting things that I didn't have time to test (I'm also not very creative with prompts).

Triggers:

Main: a woman is sucking a cock, wide open mouth, open mouth

Angle: head and chest only, full scene shot, full face shot

Tags Used on Training: eye contact, deepthroat, throatfuck, kneeling, bbc, busty, teeth, indoor scene, outdoor scene, shot from below, open wide eyes

Note³: wide open mouth results in bigger cocks and throatfuck doesn't seem to work well.