I spent a fair bit of time updating my current model to generate "realistic" images.
While it's still far from being photo-realistic, this new update is definitely on the right track.
I'll still be testing this new checkpoint as a lot of the prompts and weights I'm used to doing are completely different, it'll take some time to figure out what works best.
But I thought I'd mention it so it doesn't look like I haven't been doing anything lol
I'm really happy with these results, while a good few are a bit distorted or look wrong, theres a lot of promise!
I'd love to hear some thoughts though