Trained on AnyLora
Note: i forgot to remove 'lipstick' from the prompt, so the preview images have a bit crooked hands. It should work a lot better without it :)
To get the best results, please read the description below:
Recommended weight: 0.6-0.8
Activation word: Kalifa
Upscale to 1.5
Using anime SD models is recommended, semi-realistic checkpoints should also work
I would appreciate any reviews, positive or negative!
I would very much appreciate you sharing your results ^_^
If you have any suggestions on what LoRA i should do next, please share them in the comments
Enjoy :)
Description
Version 1 - trained on 51 images
FAQ
Comments (3)
Hi I kept getting heavy black outline 2d cartoony characters instead of your more 2.5D style. I used the model and your prompt provided in the images, is there any other settings i missed? (vae choice etc, thanks!)
Hi! I use kl-f8-anime2.ckpt vae (https://huggingface.co/hakurei/waifu-diffusion-v1-4/blob/main/vae/kl-f8-anime2.ckpt), mainly with AnyLoRA/Anything v4.5 checkpoints.
Please check the weight of your LoRA(shouldn't be more than 0.8, ideally less than that) and your selected checkpoint. The vae shouldn't make that big of a difference if the "Automatic" option is selected, so it may be a good idea to check how it works with the vae i provided, or with some other that you like.
Hope that helps :)
@12user34kn276 Thank you!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.