模型介绍: Best weight may vary This LoRA is definitely not the best option, but I prefer to preserve it so I leave it here ~~~~~~~~~~~~~~~~~~~~~~~o(* ̄︶ ̄*)o~~~~~~~~~~~~~~~~~~~~~~~ 示例提示词和参数: Prompt: (best quality), (masterpiece), detailed, 1girl, Tsunade, blonde_hair, brown_eyes, forehead_diamond, green_haori, grey_kimono, blue_pants, cleavage, loose_ponytails, looking at viewer, smirk, blush, large breasts, sitting, from above, sweat, indoors scenery, japanese_clothes, <lora:EMS-7851-EMS:0.600000>, , <lora:EMS-39696-EMS:1.000000> Negative prompt: (worst quality), (low quality), (bad anatomy), (bad fingers, combined hands, mutated fingers, extra fingers, poorly drawn fingers, bad hands, mutated hand, deformed fingers, missing fingers, missing hand), extra digit, fewer digits, text, error, cropped, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, out of focus, multiple views, (bad proportions), (multiple legs), (multiple arms), (missing arm), 3D, bad_prompt, inaccurate limb, lowres, (extra fingers, deformed hands, polydactyl) Steps: 30, CFG scale: 9, Sampler: Euler a, Seed: 1759209988, VAE: kl-f8-anime2.ckpt, ENSD: 31337, Size: 512x768, Model: EMS-39836-EMS, TaskID: 638679877641411930, Version: v1.6.0.8-2-g710787a, Hires steps: 8, Hires resize: 1024x1536, "EMS-7851-EMS: 614500170a6c, EMS-39696-EMS: 48cb1b4026c5", Hires upscaler: 4x-AnimeSharp, Denoising strength: 0.5, Clip skip: 2 |
1人表示很赞