This WAN Lora is based on parts of the Ratatoskr Helheim / Flux Database
its made to animate Ratatoskr image outputs with WAN 2.1 I2V model
I also recommend using any Ratatoskr based Pictures to generate Videos with it.
the T2I Version is also in early alpha and can downloaded / tested here
trained now at 1100 Pictures and 132 Videos (V1)
the main goal is to train a whole Wan 2.1 model if you like to support me please generate some videos with it to get some video input for further training also it would be nice to support my work with a like.
Description
So now we leaving the alpha and beta stage let me introduce V1
more stable
still creative
better in nsfw
trained now at 1100 Pictures and 132 Videos
recommended strength between 0.9 and 1
FAQ
Comments (8)
Great! Think it needs a bouncing jiggle physics test ;)
bet it can do it ;-)
For Online generation without Censorship
https://tensor.art/models/873261284991235398/Ratatoskr-WAN-2.1-Lora-I2V-V1
Hey! That cheetah sticking her tongue out looks mighty familiar lol
I used your Lora - Checkpoint? - in SDXL and it worked pretty damned well. It's quite effective, and I really should upload some stuff from the SDXL side of things. You do really solid work.
My wishlist (and my birthday is coming up) is something similar but with human faces. Your work tends to give animal faces and ears. I have found no way around that other than image to image, and it still likes to put those animal faces on .
At any rate, once I get myself better situated in Wan I plan on giving this a go!
I'm probably using it wrong, with Fusion x (the model, not the Lora), it doesn't give any tangible results, it doesn't do what I tell it to do, nothing comes out even though I write the triggers.
Well done! This solved my problem with furry faces morphing into human.
it gives me zero effect