Text by @sevenof9247

https://civarchive.com/articles/3990

juggernautXL_juggXIByRundiffusion inpainting

How to use:

Recommended Negative Prompt:

(worst quality, low quality, illustration, 3d, 2d, painting, cartoons, sketch), open mouth

or another negative prompt

load the model

go to tab "img2img" -> "inpaint"

you have now a view options, i only describe one tab "inpaint"

1. put any image there (below 1024pix or you have much Vram)

2. press below "auto detect size" (extention: sd-webui-aspect-ratio-helper)

3. draw a maske (you can resize the pencle or reset)

4. write a prompt, what you want to see in that mask

5. now you have several options

a) Masked content

1. fill and latent nothing: best suited for large changes with high denoising strength (>0.8)

2. original and latent noise: for minor changes with Denoising (0.4 up to 0.9)

b) Inpaint area

1. Whole picture: take more the weight from the whole image

2. Only masked: take more weight of your prompt

sampling steps ~40, Mask blur 4-8,

After multiple rounds of testing, the suggested drawing parameter settings are:

• Steps ≥ 30

• Sampler: DPM++ 2M Karras

• CFG scale: 7

• Size ≥ 1024x1024

Other stuff:

Merge between

juggernautXL_juggXIByRundiffusion

sd_xl_base_1.0_fp16_vae

sd_xl_base_1.0_inpainting_0.1