Overview

This LoRA is for close-up facial cumshots. It is meant to generate these with thick sticky cum. It can somewhat generalize to wider camera angles but it works best with input images where the face fills a large part of the frame. See example images below.

I will update this text over time, I don't know all to write right now.

Shortcomings

• When the penis is far from the target's face cum streams can become suspended mid-air. The model is very keen to connect the streams between penis and face.

• Generated penises don't look good. Overcome this by using a penis LoRA or having one in the input image.

• Sometimes the model wants to make cum appear from above. To stop this avoid works like over, above etc.

• The trained concepts are not decoupled from camera angles and framing. It does some things a lot better from certain angles.

I want to improve this LoRA but I expect there will pass some time between releases.

Generation

This model is image-to-video! The input image sets the scene, lighting etc. Including a penis shooting cum in the input image gives a lot of control.

Input image

The model is a lot better in certain angles. I recommend starting from something similar to the framing in these images:

Recommended settings

All the recommended settings given here are with Lightx2v being used. I have not explored settings without it.

Strength: 1.0

1.0 works best for both the high and low-noise models.

CFG

High-noise: 1.0 - 4.0

Higher CFG values for the high-noise model makes the cum thicker. This is linked to the resolution. Higher resolution needs higher CFG to increase the thickness. The sweet-spots I've found are:

• 512 x 512: 1.3

• 720 x 720: 3.5

You may need to lower the CFG when working with unusual lighting to keep the image from degenerating.

Low-noise: 1.0

For the low-noise model I leave it at 1.0. I haven't seen much difference when changing it.

Steps: High 4 + Low 4

4 steps for both high and low works well. Additional steps on low-noise can give more detail in the cum. I've not seen much improvement by going higher than 6.

→ If cumshots "erase" each other you may have too few steps for the high-noise model. Try increasing.

Sampling shift: 3 - 7

Experiment with sampling shift to find what works best for your prompt. I don't know what it does. Please post feedback if you find any patterns and I will add it here.

Prompting

Prompt using natural language and use keywords from the tags list (see tags used below). Describe the scene and other things as you would normally for Wan 2.2.

My go-to prompt is some variation like these:

the video shows an erect penis shooting monstrous amounts of cum into woman's face with fast rhythm. multiple thick slimy cum shots. every cum shot sticks thickly to her face. her whole face gets buried in slimy cum. the cum is sticky.

the video shows an erect penis shooting abundant amounts of cum into woman's face. the separate cum shots spurt with very fast rhythm from the penis. countless thick slimy cum shots erupt. every cum shot sticks slimily to her face. her whole face gets buried in sticky cum.

Cumshot rhythm / cadence

It usually responds to phrasings like these:

• slow rhythm cumshots

• fast rhythm cumshots

• very fast rhythm cumshots

Cum amount and thickness

It usually responds to phrasings like these:

• monstrous amounts of cum

• abundant amounts of cum

• huge cum shots

• big shots of cum

Start image without penis

To have a penis appear during the video, prompt it into existence. For example:

• penis enters from the left

• penis appears on right

My workflow

I create and render my start images in Blender. I include a penis that shoots the first cumshot. By changing the size of the shot you can often control the size of shots in the generated video. I use motion blur to control the force or velocity of the shots.

Most of my published videos have the workflow embedded. The released model uses different file names, so if the workflows are missing the LoRA files simply replace them with the high and low of this release.

About the dataset

• The dataset was 26 clips of 5 second length. High-noise model was trained at 256 x 256, and the low-noise at 512 x 512.

• The average number of tags was 11.8 per sample, and the average number of words was 35 per sample.

Tags used

These are the most frequently used tags of the dataset. Use words from these in your prompts; it's not necessary to use the whole tags verbatim.

• cum spurts from penis

• cum shoots from penis

• cum drips from penis

• cum is sticky

• cum on face

• much cum on face

• cum on tongue

• man strokes penis

• penis touches mouth

• woman is happy

• woman is calm

• woman smiles

• woman tongue is extended

• woman mouth is open

• woman eyes are closed

• woman blinks eyes

• woman eyes flutter

• woman grimaces

• woman looks up

• woman looks into camera

• top-down view

• side view

• profile view

• front view

• static shot

• handheld camera

• erratic camera movement

Post any questions you have and I will try to answer. Enjoy!

P.S. If I used your resource in any of the sample videos please let me know so I can credit you. I've used only this LoRA in almost all my generations while testing but something may have slipped through.