zlacker

[return to "Releasing weights for FLUX.1 Krea"]
1. qlm+Cf1[view] [source] 2025-07-31 21:23:00
>>vmatsi+(OP)
Images still look AI generated
◧◩
2. dvrp+Qh1[view] [source] 2025-07-31 21:39:58
>>qlm+Cf1
Yeah, there are still imperfections. But, it’s surprising to us how much the quality can be improved without the need of a whole pre-training (re-) run.
◧◩◪
3. Lerc+Lv1[view] [source] 2025-07-31 23:27:00
>>dvrp+Qh1
Is it possible (or do people already do this), to train a classifier to identify the AI look and use it as an adversary to try and maximise both 'quality' and 'not that sort of quality'?
[go to top]