Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord

To support me you can buy me a coffe at https://ko-fi.com/antonioriolo

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• CFG Scale : 5 to 10

• LORA weight for txt2img: reccomended 0.4-0.9.

• Keys: back_view, photo from behind, looking_at_viewer

Very  simple to use, just use the keywords, i advice to use togheter back_view, photo from behind, looking_at_viewer with a weight like 1.3 should work most of the times.

In the training i added looking_at_viewer in captioning for half of the images since the others were looking in front of them and i wanted to split the 2 concepts, but it seem i had to add some captioning also to the others to split better. 

Now it seem it's only capable of drawing images of people looking at the viewer, even putting looking_at_viewer in negative prompt didn't work for me. Maybe in the future ill come up with a new version that can handle also this.

Works with both realistic and anime images.

Enjoy!

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians