Making models is time consuming. Do you like what I do? Feel free to buy me a coffee ☕
This is my new amazing hires sized Model based on Nihil’s Geiru Toneido Dataset! If you want your own style/person trained, just DM me with a dataset of ~20 imgs on discord Flonix#4022 and ill train it for you!
trained on high res vertical images
set the weight lower than 1 to get it to follow your prompt more.
Use with SD upscale on 0.3 denoise and Swin IR4x
Example Prompt, made with Additional Networks:
GTT woman in a hair saloon
Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 2780232970, Size: 608×912, Model hash: 08acb74861, Denoising strength: 0.24, ENSD: 31337, Mask blur: 4, SD upscale overlap: 64, SD upscale upscaler: 4x_foolhardy_Remacri, AddNet Enabled: True, AddNet Module 1: LoRA, AddNet Model 1: GTT(5c8d9557bdc7), AddNet Weight A 1: 0.4, AddNet Weight B 1: 0.4
How to use LoRA’s in auto1111:
Update webui (use
git pull
like here or redownload it)Copy the file to
stable-diffusion-webui/models/lora
Select your LoRA like in this video
Make sure to change the weight according to the instructions (by default it’s
:1
)
Text here with @Lykon‘s scheme