Cumflation Concept LoRALoRA request queue on my discord: <a target="_blank" rel="ugc" href="https://discord.gg/DAfjhRfyEM">https://discord.gg/DAfjhRfyEM</a>This model was trained using 225 images using the NAI model. This was my first attempt at making a model trained on something more extreme, and it is tricky to get good images with due to the results being far from what most models normally generate. So don't be surprised if it takes a moment till you get something decent or something that is not an abomination xDI did some testing and suggest to not go above 0.7 strength, but 0.5-0.6 seems to be a decent spot. It will require lots of inpainting to get the images to where you want them to be. I suggest to turn the model off while inpainting stuff that isn't the belly to get the rest cleaned up.Trigger word is cumflation, but you can also add stuff like bloated belly or excessive cum if you want it to be a bit more consistent. But like I said don't expect to get clean gens, you will need to inpaint a lot.Example image 1 was generated using AnyHentai with 20 steps, image 2 with AOM2 hard. The first image also used this LoRA: <a target="_blank" rel="ugc" href="https://civarchive.com/models/18419/murkys-suspended-on-penis-lora">https://civarchive.com/models/18419/murkys-suspended-on-penis-lora</a>Images were then upscaled on 0.2 denoising strength and 150 steps without the LoRA being active to clean it up a bit.I'm curious to see what people will create with this, also if anyone has suggestions to get this to give better results without inpainting please let me know.Note: This was made exclusively on anime images and on an anime checkpoint. Feel free to post images on realistic, semirealistic etc. checkpoints, but do not rate it on that! This was made to be used on anime checkpoints so keep reviews for that!

Files with SHA256: fc7595fabac5b16a0e316ebc4c8b9783668a0b3426026b682010b4526303b00f