Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
I present you one of the priestesses of mount Mt. Muldoon: Afura Mann this is the LOCON version of the LORA . This is likely my last El Hazard LORA. As I said before I do Like Miz, Diva and Alielle, but the general...
Model source
Source description
I present you one of the priestesses of mount Mt. Muldoon: Afura Mann this is the LOCON version of the LORA . This is likely my last El Hazard LORA. As I said before I do Like Miz, Diva and Alielle, but the general poses and some of the laziness when doing full body images in the series makes it almost impossible to do a first or second generation lora and i seem to need to pad the dataset with images generated from those first LORA. In this particular cases ended up tweaking the dataset several times, then complementing it with images from the manga and then even further having to pad it with generated images from the earlier LORAs. Compounded to that, her only images outside of her wearing her signature clothes are extremely low quality bathing scenes and that whenever her ponytail is shown, it is in fight scenes distorting her hair.
Should work fine with: afura_mann,
Only one trained outfit her classic:
green_sleeveless_bodysuit_brown_capelet_pauldrons_green_headband_wristband_arm_strap_gem_belt: Her classic outfit.
Sources
1 sourceModel artifacts
1 artifactSource excerpts
1 excerptSource context: 248 downloads · Type LoCon · Base model SD 1.5
Anyway this is trained mostly on canon, padded with images created with previous versions. Therefore no dataset for you, as always instead i will add the tag summary.
