Experimental lora trained on ~2500 images parsed from <a target="_blank" rel="ugc" href="http://rule34.xxx">rule34.xxx</a> same_size_vore tag.Theoretically, it can accept any tags that can be found in images with same_size_vore tag on rule34, but don't count on a stable result on everything.I can highlight the following:same size vore, after vore, vore - recommended to include in every gen.struggling prey, belly bulge, stomach bulge, squirming gut - bulge/struggling tagsinternal view, inside view, x-ray, xray view, unseen character - inside view tags (very unstable results)

Files with SHA256: 03323aa3f345a34631c377ad45078cb85f715bb1486e8ca9e71176902b385eae