I'm running a Stable Diffusion bot with the new model @X (formerly chiobu) was using the other day (download.) It's finetuned on 120 Marseys following this guide. It's not quite as good as the Dreambooth model at composing a scene, but it does make very cute Marseys.
Just start your line with "!sd prompt_here":
!sd marsey made of ice
!sd marsey harry potter
!sd marsey singing at a concert, cartoon
Works up to five prompts per comment
GitHub here for any interested codecels. I rent my GPUs, so this'll just be on for the next couple days or so. I'll bring it back up if there's a significant bump in model quality
edit: some coding hiccups if your prompt didn't run, post it again
Jump in the discussion.
No email address required.
Top left is perfect if you swap the black/orange in the tail to white/orange
Jump in the discussion.
No email address required.
Thanks for the input! We'll be sure to take your perfect design and make it even better.
Jump in the discussion.
No email address required.
Thanks buddy
Jump in the discussion.
No email address required.
Don't thank me, thank yourself. If it weren't for you being born, I wouldn't have this great honor.
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context
More options
Context