This is an omorashi lora, trained on Pony Diffusion V6 XL. It understands a lot of concepts: peeing, wetting, desperation, pulling clothing down or to the side, toilet use, different poses, different angles.

Here are the relevant tags it was trained on:

Core concepts

• peeing

• wetting self

• desperation

• leaking

  • Like "wetting self" but a lot less. Might need to experiment with the strength, or even add "wetting self" as a negative with low strength.

• trickle

  • like "peeing" but less

Pose/expression

• holding crotch

• standing

• sitting

• squatting

• lying

• kneeling

• relief

Clothing

• panties

• pants

• shorts

• leggings

• swimsuit

• clothing aside

• clothing down

• clothing around one leg

• ripped leggings

  • I only had a handful of images with this, it may not have learned it well.

Container

• toilet use

  • you can use this one on its own

• peeing in container

  • combine with a container from below

• holding container

  • Use if the girl should be holding the container. It may struggle a bit with this, there's only so many training images of this concept.

• cup

• bottle

• teacup

• bag

• bucket

• potty

Angle

• from above

• from below

• from side

• from behind

Some of these tags are "compositional". E.g. desperation is a general look of needing to pee, holding crotch means actually holding oneself. If you want the typical "potty dance" pose you need both tags. Similarly, "peeing in container" for the overall concept, then "cup", "bucket", etc for the type of container.

Generation tips

Euler a sampler, 30 steps. 896 width x 1152 height is the most common resolution bucket in the training data, but a variety of resolutions will work. Positive tags: source_anime, score_7_up. Negative tags: source_pony, source_furry. You can try other Pony quality tags, or a combination, but I found score_7_up alone gives the best results.