Posted for civitai.com and tensor.art only
Each of these DoRAs take me dozens of hours and GPU usage out of pocket. Every review is appreciated and encourages me to do more of these.
-MoC
See about this version for model details 👉
Description
My first DoRA attempt, based off the quality derived from this first attempt, not seeing a reason to stick to standard LoRAs anymore.
Dataset gathered and trained on a single concept, ranging across this model's career.
~11 hour Kohya_ss training in 15975 steps across 25 epochs @base dim (rank): 32.
1248x1248px training resolution base
Used a wide variety of resolutions in training data, however by far the most common bucketed resolutions were @ 1024x1472 & 1216x1216px.