Unfortunately, not the best flexibility due to limited dataset
good range: 0.55 - 0.85
recommended with: 0.75
CLIP: 2
CFG: > 8
Steps: > 35
Samplers: DPM varieties
HiResfix
model trained on: SD 1.5-pruned-emaonly
negative prompts should use: blurry, out of focus, watermark
if you would like to support me or requests:
https://ko-fi.com/s6yx0
added more images to dataset
redone captioning
128 dimm/rank (aka bigger file)
Go ahead and upload yours!
Your query returned no results β please try removing some filters or trying a different term.