Additional Info:
This was the first Lora I ever trained.
It definitely isn't the best
I will have some better lora's coming in the next few weeks.
It was created to look more like the chainsaw man anime so most images have an anime look to them because of this you can generate some screenshot esque looking images e.g.
These are generations but the look like screenshots from the anime
Base model used for training the Lora
https://civarchive.com/models/66/anything-v3
Recommend generation model (Anime)
https://civarchive.com/models/3095/wilds-mix
Recommend generation model (Real)
https://civarchive.com/models/4201/realistic-vision-v13
Model version of the Lora
https://civarchive.com/models/10094/power-from-chainsaw-main-checkpoint
Video Created With TikTokDifusion Using This Lora
Description
FAQ
Comments (5)
hello, very good, your loRA wanted to know how you do it, do you have a link where you learned from, thank you very much
Do you mean how to train loras in general or how to train this in particular?
I used this repo https://github.com/kohya-ss/sd-scripts/issues but an easier to use repo would be https://github.com/bmaltais/kohya_ss
@tingtingin Thanks you are my idol haha
can you please share your settings for power video of tiktok diffusion. I tried making similar thing but ended up just getting blurry face, nothing changed in body. I would like to know your prompt, denoising strength etc and how to affect whole body instead of just face.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.