no key need, just prompt like normal.
512x768 "work in any resolution too."
tips or requests
Description
update your additional networks with is not working
FAQ
Comments (6)
Is it trained with the picture of the card?
trained on various cards
is it trained on just the card art or the whole card?
its trained on a entire card, a set of 10 cards with random images and colors.
the lora learn that the box art can be prompted, but the frame is common in every another any image.
Incredible, next step is to create a checkpoint with all the 11k cards, I want to play with fake HERO support lol
I wish there was an MTG version of this. The art of MTG is like the old Dragon Magazine covers. I like that old oil painting fantasy RPG style of Keith Parkinson and Larry Elmore and such.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.