>>45785713yeah i made a lora just yesterday and i've been playing around with it to try to figure out good CFG/upscaler/tag settings. here are some other pics i've posted
>>45752395>>45752871>>45755513>>45757318https://files.catbox.moe/hu3mex.jpghere's a link to the loRA itself, with the caveats that i don't really know what i'm doing when it comes to training (but it seems to work), the base model is abyssorangemix2_sfw so it should work best with that, it should help to specify
>rune \(dualhart\),in the prompts because that's what they're tagged as on danbooru where i scraped the pictures from, and i think you have to use the 'additional networks' a1111 extension instead of the built in loRA support unless the incompatibility bug with the latest training scripts is finally fixed
https://files.catbox.moe/sf5meh.safetensors 