metadata
license: creativeml-openrail-m
language:
- en
tags:
- stable-diffusion
- text-to-image
based on Anything V4,
Made by several Loras, and block merge with community models
Using these negative prompt have actually resulted in significant improvements
https://huggingface.co/datasets/gsdf/EasyNegative
https://civitai.com/models/4629/deep-negative-v1x
Although I tried to reduce negative prompt, but gave up because the difference was difficult to evaluate.
Maybe this is enough.
(EasyNegative:1.4), (NG_DeepNegative:1.4), (worst quality:1.4), (low quality:1.4) , (monochrome:1.1)
(masterpiece), (best quality), (illustration), (beautiful detailed), (highres), 1girl, solo, face foucus
Negative prompt: (EasyNegative:1.4), (NG_DeepNegative:1.4), (worst quality:1.4), (low quality:1.4) , (monochrome:1.1), lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, normal quality, jpeg artifacts, (signature, watermark, username:1.4), blurry, bad feet, multiple breasts, (mutated hands and fingers:1.5 ), (long body :1.3), (mutation, poorly drawn :1.2) , black-white, liquid body, liquid tongue, disfigured, malformed, mutated, anatomical nonsense, text font ui, malformed hands, long neck, blurred, lowers, bad proportions, bad shadow, uncoordinated body, unnatural body, fused breasts, bad breasts, huge breasts, poorly drawn breasts, extra breasts, liquid breasts, heavy breasts, missing breasts, huge haunch, huge thighs, huge calf, fused hand, missing hand
Steps: 32, Sampler: Euler a, CFG scale: 8.5, Seed: 601115233, Size: 640x640, Model hash: 79fb704aeb, Clip skip: 2, Eta: 0.67