I trained a z-image Lora on my face, im incredibly impressed

The last time i did this was with stable diffusion 1.5. That was already incredible, but there was always this "AI" vibe around it. Now with z-image all those pics feel so real, and the resemblance is not uncanny, its literally me. incredible.

https://preview.redd.it/3xwamymjcq4g1.jpg?width=896&format=pjpg&auto=webp&s=8f6b3f6c955509952fd668e3279b18dcf42faf78

https://preview.redd.it/q2uy3izkcq4g1.jpg?width=1504&format=pjpg&auto=webp&s=c2f8ae8fa094fc1eb2e1c18e1328d0ce23d88da3

https://preview.redd.it/5zf8b0smcq4g1.jpg?width=1296&format=pjpg&auto=webp&s=0e798a6223fca77f6d37e6dd43a0647b7a564350

https://preview.redd.it/s1ekbi9ocq4g1.jpg?width=1504&format=pjpg&auto=webp&s=409c4308b58a2532612f1c1a35e9354d9240e1cd

Edit:

I add a training photo for comparison:

https://preview.redd.it/frqcos4bnq4g1.png?width=1728&format=png&auto=webp&s=972048408377927a0d4d9dcddc88b65adfc718b5



https://redd.it/1pc0psn
@rStableDiffusion

Z Image Turbo ControlNet released by Alibaba on HF

https://redd.it/1pc2enz
@rStableDiffusion

Finally !! ZTurbo ControlNet
https://redd.it/1pc2jis
@rStableDiffusion

GrainScape UltraReal. Z-Image LoRA

https://redd.it/1pbxfut
@rStableDiffusion