Download
1 variant available
2,1160 1 2 3 4 5 6 7 8 9,0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
79.7K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
69.8K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
(429)
Jun 24, 2024
Pony
This is a much larger LoRA with many characters included, with the intention of something closer to a full-finetune in the tagging and training method.
Trained on Pony Diffusion v6
Expanded dataset for many more characters
Much longer training run (5,940 steps at batch size of 6)
Added tagging for quality and aesthetic (masterpiece, very aesthetic etc)
Show more

6.4K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
11M0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9M

License:
CreativeML Open RAIL++-MDelicious in Dungeon - ダンジョン飯 - Collection
Collection of character LoRA's intended to create non-commercial fan works of the Delicious in Dungeon (ダンジョン飯) series.
See versions for each character LoRA and recommended settings.
Characters:
The characters and their outfits may blend (particularly staff, armor) in these combined models
Recommended prompt structure:
Positive prompt:
score_9, score_8_up, score_7_up, score_6_up, source anime, {{tags}}, dungeon meshi, masterpiece, best quality, very aesthetic, absurdresNegative prompt:
score_4, score_5, score_6, monochrome, (censor), 3d, realistic, render, CGI, text, watermark, cartoon, sketch, greyscale

