Futa Heavy Flaccid: for shots of futas with flaccid penis where you can personally feel the weight
I am aware that there are more flaccid models for futas out there (or even in general) but I wanted to give my own spin to it.
This is a lora in which you create relatively large flaccid penises, mostly downturned and veiny. It is consistent between different models and can be combined with other loras. It is capable of creating foreskin and urethra, but for more consistentcy in that you can try to combine it with the Futa Foreskin Taker pov model.
Pony V2 is trained on 92 images using pony for 1 step at 65 epochs.
Noob/Illustrious V1 is trained on 98 images using NoobaiV1 for 1 step at 100 epochs.
Instructions: Set at strength 0.3-1 Feel free to use any base model you feel like, but it is recommended to use the base model matching the lora.
Use the following trigger tag: fhf
State of the penis: veiny penis, dark penis
Position: from above, from side, from below (you can use fftaker if you use this one for help)
Optional tags: huge testicles, foreskin, urethra, precum, close-up, lower body focus, pubic hair, Experiment!
Somewhat gacha: penis ring
Pony versions: When using low strength it sometimes fails to understand fully how a penis head is supposed to look like. I would recommend (if that occurs) to do inpainting with a model like Futa Foreskin Taker, which seems to be better for generating good foreskin and urethra. Or, as mentioned before, try to combine the two!
If it gives erections then I would recommend including in tags the following: (fhf), (1girl), (flaccid), (veiny penis) and (foreskin) if you want that.
Noob/Illustrious version: It has a tendency to show you an angle from below. When this happens put (from below) in negatives and (full body) in positive.
Later versions may have more consistency, more penis positions, different thickness/lengths. In any case the bottleneck is finding a good imageset. Feel free to send good images my way. I tend to be a bit particular about which flaccid images work well.
If you like my loras and would like to engage more, we now have a discord server! https://discord.gg/R4A57rKdcN
Description
Version exclusively for SDXL. Use a base model of your choice. Slightly less consistent but doable.
FAQ
Comments (6)
is sdxl meant for pony? or any sdxl?
I only use pony and autismmix, but feel free to try any SDXL model. It is trained using animagine.
i love this lora
SDXL and/or Pony? Can we get some clarification. Please and thank you.
Lora works with any XL checkpoint. It was trained using animagine XL which was before pony therefore i listed base XL. V2 (whenever theres a big enough dataset) will be trained on pony. Personally i recommend using autismmix over pony but thats my preference. Pony is still a very powerful checkpoint.
Or correct the details panel, at least.