FANTAZEE - Fantasy Oriented Manipulator LoRa for SDXL
Images in gallery are NOT cherrypicked, just a bunch of semi-random prompt made with my collection of wildcards
All the images used Restart sampler but it works with any sampler
Hope you enjoy and have fun
if you like my LoRas just publish some image made with my models to help me gain buzz
Training LoRa necessitate a lot of coffee. If you think that what I do is useful and want to help me keep doing this you can buy me a coffe here: https://ko-fi.com/clumsy_trainer
Description
FAQ
Comments (1)
This one is pretty neat, it follows what I say more accurately than the model itself. And the results are a bit more predictable that way. That said, don't go over 0.8 anything higher and you get a wonderland effect where everything is simply crazy (though it probably depends on models), I was using a straight lace model (ultrium v4) which tends to be pretty sharp but not that creative. This one adds the creativity in there. Though I don't think its as sharp, and this is tricky to use you may need numbers like .75 or something small like that.