Stable diffusion is the biggest time sink, well after you find civitai and want to try some “one more” lora and promt. Wish my GPU wast 9 year old…
possitive promt: 4k, photo, realistic, photo of a beautiful 30 y.o woman, nude, (full body), ((detailed face)), (detailed skin), pale skin, petite, (ginger hair),(twintail), (looking at camera:1.6), (detailed eyes:1.3),(smile:1), squatting, bathroom, leaning backwards, sunset, depth of field, c-cup, normal nipples, normal areola, pink nipple, “lora:Breasts_Helper_Trail_v2:0.6”, inniepussy1,pussy,front pussy,cameltoe,slit_pussy,open_pussy, “lora:InniePussy1 v4:0.4”, Negative prompt: lowres, blurry,jpeg artifacts, asymmetric eyes, bad anatomy, cloned, crooked teeth, deformed, dehydrated, disfigured, double nipples, duplicate, extra arms, extra fingers, extra legs, extra limbs, long fingers, long neck, malformed limbs, missing arms, missing legs, missing teeth, more than five fingers on one hand:1.5, more than two arm per body:1.5, more than two leg per body:1.5, giant breast, giant boobs, Steps: 40, Sampler: Euler a, CFG scale: 7, Size: 512x768, Model hash: fc2511737a, Model: chilloutmix_NiPrunedFp32Fix, Lora hashes: “Breasts_Helper_Trail_v2: 470f04826a09, InniePussy1 v4: 4de35566a92e”, Version: v1.5.1
I tried just for a few minutes “easy-diffusion” is supposed to use less vram. Cant quite comment on that, though it is slower then AUTOMATIC1111. Power useage is also way lower then using easy compared to 1111. So havnt actully used it to for real. Image generation is already slow as it is…