Z Image Turbo – Beta Release
This is a beta version, so you may need to generate the image multiple times to achieve optimal results. Further model retraining and improvements are planned to deliver higher-quality outputs in future updates.
The only difference between version 0.1 and version 0.2 is the number of training steps. Version 0.1 was trained for 5,000 steps, while version 0.2 was trained for 10,000 steps. Both versions are still experimental, so feel free to try each one to see which delivers the results you prefer.
To view the prompt and workflow, simply download the image and open it in ComfyUI.
What it is
This is a LORA that use in text2img
It's best at weights under <1.2 with accompanying keywords.
This is an example of how to use my program to search for sex images to train Lora
All the images used in training were extracted from the program along with the captions
Installation:
Download the file
Place it in models\Lora folder
Adjust the weights. 0.5 - 1.2 , 0.8 recommended
All images used in the training are from this software:
Used Filter:
Description
beta version