A version of Megachubsdxl LORA that is meant for generating images of megachubs and fat men in general with Ponydiffusion. Meaning its capable of generating a lot more and varied images, compared to base sdxl. Ponychub was trained was trained on different set of images this time. Mainly semi realistic images.

Place this in the Negative prompt:

score_2_up, scored_1_up, worst quality, low quality, normal quality, messy drawing, amateur drawing, lowres, bad anatomy, bad hands, source_furry, source_pony, source_cartoon, comic, source filmmaker, 3d, blurry, cropped

There is an issue where it generates images with a strange rock pattern on the underside of the belly. It was present in the last version too. Adding a low score to negative prompt can remove it. Feel free to experiment with how much positive score prompting.

Place this in the Positive prompt as a starting point:

score_9, score_8_up, score_7_up, score_6_up, megachubsdxl, morbidly obese, [subject], 50mm. cinematic 4k epic detailed 4k epic detailed photograph shot on kodak detailed cinematic hbo dark moody, 35mm photo, grainy, vignette, vintage, Kodachrome, Lomography, stained, highly detailed, found footage

In some case, reducing the amount of score in the prompting can increase the size of the person. Other times it isn’t necessary.

V1.01

• Minor changes, can actually differentiate between other races and trained with a little bit more training data.