What is it?
TickleDiffusion is a model aimed at creating tickle act in your art. If you're a total noob at writing a prompt, this model will do everything for you, even with a one-word prompt. Works great with single-one male and female characters. Alas, this version does not work well with tickling feet. I'm going to make a version for foot tickling, too. SOON™. More examples can be found on my Twitter and Boosty.
How to use?
Trigger Words - tkl, tickling, torso grab, disembodied limb, pov hands...
I use Clip Skip: 2, ENSD: 31337, vae: kl-f8-anime2.
More prompt you can found in the attached images.
Why not LoRA?
I also have LoRAs for body and foot tickling that work great. This model is a demo version of my LoRA's features. I'm going to update it and develop it further, and it will be enough for a normal user. If you want to speed up LoRA's release on CivitAI, then please support me on Boosty. I hand-selected over a thousand images for the dataset, cropped and cleaned them of text and other garbage that would affect the final result.
Description
A few fixes.
Updated style that I personally use.
FAQ
Comments (5)
hey i have a question can you do make this as a lora i want using this for realistic tickling pictures
@Hililuhwoop thats cool but my question is can i use this for realistic tickling too?
@NexaHoxi2023 I've tried using my LoRA in conjunction with realistic models and it works, but sometimes weird.
This is because the dataset for model training consists of art and not photos.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.










