Please check out my Ko-Fi page if you are interested in supporting me.
LoRA commissions open! -> Fiverr
Trigger words for V1.18:
Simple prompt: <lora:straightjacketv118:0.7>(straightjacket, crossed arms)
For her clothes: <lora:straightjacketv118:0.7>(bulge, futanari, straightjacket, crossed arms)
Can always add "bondage", "bdsm", "bound" or "restraint" to improve the gens.
Note: This LoRA was made as a request and it was specifically to generate Futanari characters with a large bulge and straightjacket, but non futa is also possible. This was extremely hard to find images for, in the end I had to MANUALLY photoshop 21 images to make it work.
Follow the steps in the Model Version notes to reproduce my images.
Sample images generated from Nothing.
All negative embeds should be easy to find. and Prompts should be included in the sample images! Click the little icon in the bottom right to see it.
It was trained on 21 512x512 images scraped from the internet and 18 epochs using animefull_final.
Danbooru tags generated by Waifu Diffusion 1.4 Tagger and fine tuned for specific tags.
Description
Generate using a weight of 0.5-1 like this: <lora:straightjacketv118:0.7>
How I archived my generations:
txt2img: DPM++ SDE Karras with 20 steps,512x512(can go up to 768 for wide or tall), Restore faces on (using CodeFormer weight 1.0 in Settings), Hires-fix on, to 1024x1024 (double original size) with your preferred upscaler and Denoise 0.5, CFG 7. Then send to img2img!
img2img: DPM++ SED Karras with 20 steps, Restore faces on (CodeFormer weight 1), rescale to 1536x1536, CFG 7, Denoise between 0.3-0.7 (based on how much you want to improve the image), SAME PROMPT! Send to extras when happy with img2img/inpaint results.
extras: Scale to x3 (final will be 4608x4608, I suggest going lower, I did x3 just for the samples here) Upscaler 1 - FatalAnime 4x, Upscaler 2 - SwinIR 4x with 0.10 visibility, GFPGAN 0.10 visibility, CodeFormer 0.10 visibility.
FAQ
Comments (6)
Similar to straitjackets? Or a typo?
Not a typo, it's the official booru tag for it. Both exist.
@Part_LoRAs Ah, I see. I had a look. Straightjacket is an alias of straitjacket.
@jenxi Yes, and auto taggers tag it with straightjacket and not straitjacket which means NAI and other anime models got trained with straightjacket too. Using something that the base model used to train already understands, makes it easier for the LoRA to learn.
Is it possible to have other poses, like 'side', 'rear view', etc? I know people use to say: 'just test it',but I am renaming my LORAS to 'pose' prefix just to avoid creating batches of 5, 10, 15 renders with the same pose, even providing 'side', 'rear', and others.
This LoRA is quite limited tbh. I trained it with mostly frontal shots and very little 45º degrees angles. I wouldn't expect it to do anything else but SD being SD, maybe it can.
As I said in the description, I had to manually photoshop the 21 images cus its something that is hard to find examples of... I even searching in pixiv to get the images.