Home Artists Posts Import Register

Downloads

Content

I'm preparing a lora that inspired me to create a full-fledged model based on it, a kind of improved (in my opinion) version of abyssorangemix and Counterfeit, I'm still not able to do this with my 8gb of video memory, but in the future, if I can save up for rtx4080 or rtx4090, then I will make my own model, on the basis of which I will make all my lora.

I'm still waiting for a new open source version of stable diffusion, I'm pretty sure that it will be possible to make better arms and legs, more body position variations, which is the future for AI. But if no one is ready to switch to SD 2.2XL, then by then I will be making an anime model based on it myself from scratch, I'm almost sure that cutting copyrighted art will not affect the quality in any way if the entire base is based on yours arts.

We are waiting for serious changes in the quality of models and loras soon ( ͡❛ ω ͡❛)✌

P.S. I could post it on civitai so you can watch the prompts, but the site is dead as always (fixed https://civitai.com/models/42674/nadegata-style-realistic-volumetric-anime)

Files

Comments

No comments found for this post.