SDXL LoRA trained with Civit-AI trainer for Pony V6/Pony merges, trained on the style of atmospheric anime artist Wlop, known for their painterly and watercolor style. While models were done before for this artist, none were done for Pony Diffusion specifically.
Wlop subjects are usually games, fantasy, women in intricate or surreal settings. Generally SFW, though now can produce artistic nudes.
Trigger words 'wlop' and 'by wlop'
Uses Danbooru styled tagging
(v3 update. Now much better adherence. Works well on stock pony diffusion with DPM++ 2M)
Sample Prompt: <lora:Wlop_v3_Style_PonyXL:0.8> wlop, by wlop, score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime,
Description
FAQ
Comments (5)
More stylized than WLOP's actual art. The lighting needs to be better, I think this needs more training. It's a good first step, though, I see where it's trying to go. This is a very cool style to try and capture.
That's fair! Thank you for the feedback. I did get a few different results with the samplers so it could be how I was prompting it too.
Could be, I'll check it out. Thank you!
I re-trained and uploaded a new version! I think it has now much more detail and adherence to the style. Thank you for your suggestion before!
Thing is dataset needs more polishing and cleaning and also of course, playing with Dim/Rank. Not bad however.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.