Notice of Hiatus

Dear model makers and lovers,

Since I started playing around with SD in February, I have met many amazing people and have done many amazing things I would never have thought possible before. We really do live in exciting times, and I feel privileged to be able to share these moments with you. Thank you, I love you all very much!

Unfortunately, I must tell you that I intend to take a break in order to focus on myself, gather inspiration, and travel. I have been so laser-focused on AI creation these past months that I might have lost touch with the beauty of the real world which is unhealthy in my opinion.

I do intend to come back at a later date in the future, refreshed, and able to deliver you all fresh, original, high quality EDG-grade content. In the meantime, please do keep my content alive, like the models you haven't liked yet, and review the models you haven't reviewed yet.

Take care, have fun!

Best,

EDG

This document serves as a guide to any beginners looking to start AI art (mainly anime) using stable diffusion (Local and online)

Online Link:

Google Doc: https://docs.google.com/document/d/15B7doxj0OLxpcIcQvEVQFlHojnzwtyr-G86cZbvzWf4/edit?usp=sharing

V1.1 Changes:

• Moved Downloadable after Launch Arguments, Before Prompt Guide

• Reworked Outfit and Scene section of Prompt Guide

• Added sample images for Outfits with Scene

• Rearranged Extensions, gave them priority highlight colors

• Changed Vae comparison image

• Simplified Model links to hyperlinks with name of model

• Added table of contents
  

Feel free to:

• comment suggestions/changes

• copy and paste any of the tokens/tags within the document
  

Document includes:

• How to run stable diffusion locally and online

• Launch arguments

• Prompting guide and samples

• Webui parameters (That I use with a 3060ti)

• Downloadables: VAEs, Models, Lora/LoCon/Lycoris, extensions, etc.

• My workflow and sample images with png info
  
  My Discord: RadioFM#1369
  Patreon: https://www.patreon.com/user?u=23055206

Redream is a (kind of) real time diffusion UI using Automatic1111's API and ControlNet extension.

Select a zone of your screen, and redream in real time. (2 fps on a RTX 3090 with xformers)

(Don't forget to select an Inpainting model and a high denoising strenght if you want to use mask)

Redream the reality.

Link : GitHub

Found my upscaler useful? Support me on ko-fi https://ko-fi.com/bericbone

An upscaling method I've designed that upscales in smaller chunks untill the full resolution is reached, as well as an option to add different prompts for the initial image generation and the upscaler.

The idea is to gradually reinterpret the data as the original image gets upscaled, making for better hand/finger structure and facial clarity for even full-body compositions, as well as extremely detailed skin.

This version is optimized for 8gb of VRAM. If the image will not fully render at 8gb VRAM, try bypassing a few of the last upscalers. If you have a lot of VRAM to work with, try adding in another 0.5 upscaler as the first upscaler. Differnent models can require very different denoise strength, so be sure to adjust those aswell. There are preview images from each upscaling step, so you can see where the denoising needs adjustment. If you want to generate images faster, make sure to unplug the latent cables from the VAE decoders before they go into the image previewers.

For those with lower VRAM, try enabling the tiled VAE and replacing the last VAE decoder with a Tiled VAE decoder. This can also allow you to do even higher resolutions, but from my experience, it comes at a loss of color accuracy. The more tiled VAE decoders, the more loss in color accuracy. There's still some color accuracy loss even for the regular VAE decoding, so if you wanna use as many as I do is up to your preferences and the checkpoints you work with. You can reduce this by using fewer upscalers.

The detail refinement step needs a very low denoise strength. try not to go above 0.2. Might need as low as 0.03 or lower. This step is to add a layer of noise that makes skin look less plastic, and to add clarity.

There's some logic behind why the scaling factors are gradually decreasing which I won't go into it too much. basically, the lower scale factor the more smaller details are being worked on in relation to the denoise strength.

If you want to lower the schedulers from karras back to normal, please be aware that you need to reduce the denoise strength drastically. different schedulers applies noise differently.

ComfyUI custom nodes needed to use this:
https://civitai.com/models/20793/was-node-suite-comfyui
https://civitai.com/models/33192/comfyui-impact-pack
https://civitai.com/models/32342/efficiency-nodes-for-comfyui

ESRGAN (HIGHLY RECOMMENDED! Others might give artifacts!):
https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth

VAE: https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main