====== v2 ====== So v1 was trained on synthetic data from a <a target="_blank" rel="ugc" href="https://civarchive.com/models/566599?modelVersionId=631480">SD1.5 model</a>, this one is trained from synthetic data from the v1 Lora, for science. I think it's a bit better at prompt following and there is less bias. But I don't know yet if I'm gonna keep going with the incestious training.

Files with SHA256: 5d5f4fc0575aa1193349960f603286e5a4e303d3e38889253fb70a6ec3d0338f