Many people who create a checkpoint are now dealing with FLUX models and will ask themselves why I am publishing an SDXL model.

The question is easily answered. I believe that SDXL is far from being exhausted, it is more flexible and of course easier to handle for the normal user at home.

The focus of my model is on photorealism, details and flexibility.

NSFW is just as possible as SFW and even if the model is designed for realism, anime, cartoon and other styles are also possible.

I tried to create a large number of different photos. Whether individual tokens or sentence-by-sentence descriptions...both work.

For most of the example photos I used CFG-Scale 5.5, Steps 25, DPM++ 3M SDE, Clip skip 1. I also achieved good results with other sampling methods. VAE is integrated.

If you like the model, I would be very happy if you give it a "thumbs up".

Please post your results directly here on the model page.

Creating a test point is very time-consuming. You can support me with a few buzzes