Download
1 variant available
fp16 SafeTensor
mafuyu-loha-final.safetensors
Half precision, best balance • 18.79 MB
Verified: 3 years ago
1,4800 1 2 3 4 5 6 7 8 9,0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
(190)
Mar 12, 2023
Other
I re-retrained using experimental method, loha
This seems to do a better job of learning the details of the character, too.
The first four photos show latent upscaling, while the last two show no upscaling.
Show more

1310 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.7K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
Warning: The latest version is not a LoRA model; this is a LoCon model. It will not work without the following extensions:
GitHub - KohakuBlueleaf/a1111-sd-webui-locon: A extension for loading locon model in sd-webui
I used official cutout character illustrations to train this LoRA.
The base model is the animefull-final-pruned model.
Trigger word: 'iom'
The example images were generated using merge model based on AbyssOrnageMix2
set this LoRA's weight to 0.4~0.8. my recommendation is 0.6.
