differences between 0001:
Higher saturation
Brighter
More dynamic
uses 2 loras trained by me:
- QEW: https://civarchive.com/models/470285/qew-quasarcake-ebifurya-wlop-pony
- Bettershading: https://civarchive.com/models/470922/better-shading
Downsides may contain:
not good with loras
sometimes hands are off
Description
It's been a while since i made a new edition of this model. So heres whats new,
first i used Jupiter as model A then a unreleased version of Tsuki as model B with weights: 0,0,0,0,0,0,0,1,1,1,0,0,0,1,1,1,1,0,0,0 saved it as the name Jupiter+tsuki2.5
after that i used Jupiter+tsuki2.5 as model A and Jemnites model MIX-GEM-XLS-QromEW as model B with weights: 0,0,0,0,0,0,0,1,1,1,0,0,1,1,1,1,1,0,0,0
As i thought the style fit the 0002 lineup i chose to put it as 0002 version 3,
i hope you like it!
Also, i recommend using external SDXL vae.
FAQ
Comments (2)
Just so I/we don't waste bandwidth if this is the case, but are these alternate styles the "0002-v2test1 - 4" files on your huggingface?
heres the rest of the 0002v3-alt versions:
https://huggingface.co/sulph/0002v3