It took a while, but here's the next model :)
I'm looking forward to seeing what you create with it.
Description
The new model is a merge of many models whose styles I liked. It creates very nice images both with and without the Loras style.
The model is untrained, just in case anyone asks. ^^
I'm excited to see what you can create with it. The focus is on anime and furry images. I don't know if it can do anything else, since I mostly only work with furry images. :)
The VAE is included in the model but if you have problems use AAA Anime VAE SDXL
FAQ
Comments (6)
Unfortunately I see more and more models and lora's that have no description or it is extremely minimal. 😶
What kind of description do you want?
@nexorus1984 No need for anything fancy, really. 🤷♂️ It's just that there are a lot of checkpoints available now, so most people aren't going to download it without knowing what it does differently than the one they're already using.
I mean, some good checkpoints are bad with small details, but better than the others for prompt fidelity. People wouldn't be able to know that without a description or extensive testing.
So... maybe a sentence or two mentioning what styles or types of images it is good at generating? Or the idea you had in mind when creating that checkpoint?
And maybe just basic guidance on what values of cfg, steps, and schedulers work well with it (arguably, the example images can provide that info though).
yoo
looks pretty cool
i compared it with other checkpoints and its one of the best, dk really why but i wanna see next version
work for dragon?
Even though this model is nearly one year old at this point it is legit one of the best on here imo.