Woafu's comment
Pictures made with "Anything" are the first ones I generated. While they can be ugly, I kept some of them for the prompts, even if the weights are often... stupid. Use at your own risk.
Image Generation Data
UI | Stable Diffusion A1111 (External link to GitHub) |
---|---|
Checkpoint | Anything |
Character LORA | Qiqi (External link to CivitAi) |
Prompt | (((horror, dark atmosphere))), <lora:qiqi1:1.4>qiqidef, zombi, zombie, |
Negative prompt | (worst quality, low quality:1.4), logo, text, badhandv4, bad anatomy, bad face, bad hands, bad body, bad feet, bad proportions, cut off hand, bad leg, more legs, missing fingers, more fingers, normal quality, jpeg artifacts, gross proportions, blurry, poorly drawn, text, error, missing fingers, missing arms, missing legs, short legs, extra digit, fewer digits, cropped, out of frame, low background, distorted perspective, face out of frame, out of focus, signature, artist name, watermark, username, (watermark, logo), Asian-Less-Neg, easyneg, extra ears, fewer fingers, extra fingers, (extra hands), bad hands, sketches, (low quality:2), (worst quality:2), skin spot, age spot, glans, skin blemishes, DeepNegative, lowres,bad anatomy, extra views, multiple views, extra panel, multiple panels, easynegative, |
Seed | 3611068207 |
CFG Scale | 7 |
Generation size | 400x700 |
Steps | 20 |
Sampler | DPM++ 2M SDE Karras |
Hires size | - |
Hires steps | 20 |
Hires upscaler | RealESRGAN_x4plus_anime_6B |
Extensions | Controlnet |