If you like my work and would like to leave a review, feel free to. Feedback and model usage is appreciated but don't feel forced. I do this for fun first and then probably degeneracy second.
This LoRA is part of a Collection!
Model Information
This is a Character LoRA of Sango from Inuyasha: The Final Act, but it also contains trained images taken from doujins, allowing for some extra styles.
Locally trained with Easy LoRA Training Scripts. This dataset consists of anime screenshots plus doujin images. 280 images (Sango from TFA, some artificially created) + 155 images (6 doujins) for a total of 435 images. Different repeats were used for each.
For advanced information, see this image
Prompt Usage
The first version was banished to the shadow realm, so everything here is for v2 and the current information.
This LoRA is technically capable of producing several styles of this character. I have a bunch of images in the gallery you can reference. The main trigger word is "Sango". You'll always use it regardless, but there are other important ones to take note of.
Sample Prompts (basic appearance + outfits)
◽️"sango, brown eyes, makeup, blunt bangs, brown hair, low-tied long hair, kimono, pink kimono, skirt, green skirt, fingerless gloves"
◽️"sango, brown eyes, makeup, blunt bangs, brown hair, ponytail, bodysuit, black bodysuit, shoulder pads, fingerless gloves"
Doujin Triggers
The following triggers attempt to generate a specific style. Use one at a time, or if you want to experiment, mix them. I linked doujin sources.
◽️ doujin_heizo - (source 1)
◽️ doujin_zakupiyo - (source 1)
◽️ doujin_chikaya - (source 1)
◽️ doujin_sadakoji - (source 1, source 2)
◽️ doujin_nekomaru - (source 1, source 2)
◽️ doujin_kaminobe - (source 1, source 2)
Tips
◽️ For the anime style, use "anime coloring, no_style". I tagged the original dataset with "no_style" in an attempt to not have styles bleed, but not sure of result.
◽️ For a doujin style, try excluding the stuff above.
◽️ If you want, put what you don't want to see in the negatives
◽️ For doujins, make use of "comic, grayscale, monochrome" and the sort.
◽️ Use a weight between 0.8 and 1.
Description
Experimental version of a LoRA. Consists of anime + doujin datasets. Please see description.
Looks like we don't have an active mirror for this one right now.
This archive is kept alive by community contributions, so there are still some gaps, especially for newly removed content.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.