Horikoshi Kouhei (My Hero Academia Mangaka) LORA trained on his artwork and scans of manga panels.
Works at 1.0 weight. Because of the dataset, traditional media and monochrome have strong effects.
Sample pics were generated using Based64mix-V3 model. I do not have a link, but it's around.
PSA: If your LORA requires a weight of 0.5 or something ridiculous like that, it's overcooked.
Description
Trained on NAI.
FAQ
Comments (2)
Still having trouble finding the Based64mixV3 model, google just gives me based64 encoding sites, and can't seem to find it on HF. Any ideas?
Found this one on the site, not really sure if this is the one that you're looking for https://huggingface.co/Dorhorshu532/Based64mix-V3/tree/main
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.