The trigger is "taySwift".
Now in flux!
SD 3.5: No idea what I am doing with sd 3.5 yet so consider this a beta. You probably shouldn't use this one yet.
The results are okayish but could maybe be improved with a better workflow or settings.
Pony: There are two versions available, Base Pony and Pony Realism.
Pony Realism is trained on Pony Realism and would give best results on realistic models.
The second version is trained on vanilla Pony and is not as accurate but works on all pony models.
If you enjoy my work please leave a like or a review. Pics of your work are also appreciated :)
Description
I have no idea what I am doing yet but I made this lora as a test for lora training with SD 3.5
FAQ
Comments (4)
Your Flux one is awesome which gives me little hope for 3.5 with the results you've gotten. I was hoping for better likeness with 3.5.
Same, though 3.5 is still extremely new and there isn’t as much excitement around it as there was around flux so there isn’t very little information on how to train properly.
3.5 is undistiled which is pretty big and hopefully some good sometimes will come out.
@JaneB yeah the fact that it’s undistilled is definitely a good sign. I was just blown away by the likeness in the loras with flux. I can’t image its due purely to the number of parameters?
her face can be pretty accurate but it's too strict and doesn't allow for different facial expressions