Pucci Lavmin from Rain Code.

Prompt style

This lora was trained in narrative, rather than tag, style. It should do fine with either for additional elements but you'll probably want to stick with variations on the following prompts for Pucci herself.

Prompts

Standard outfit: "Pucci Lavmin is wearing her Rain Code light uniform. Pucci Lavmin wearing a navy blue double-breasted jacket with gold buttons and a matching skirt. She has a medal on the left side of her chest. Her skirt has black lace along the hem and she is also wearing black and purple boots with red thigh high socks that have black ruffle trimming. Her short blonde hair is parted to the side and styled with a pink bow on the back and a bunny hair clip on her left side. She has a pair of red headphones."

Raincoat: "Pucci Lavmin wearing a beige double breasted cape coat with red buttons and a red belt, red thigh high socks, and black boots. She has short, layered blond hair tied back with a pink bow and a bunny ear hair clip on the left side of her head. She has red headphones."

Accessories

Pucci's headphones are around her neck by default. However, you can simply prompt to have her wear them on her head.

Styles

There's not a huge amount of data for Pucci out there to train on so the built-in styles of this lora are fairly limited. Sadly, they don't work as well in this lora as in the original Flux version. Instead the lora tends to more strongly inherit the predominant style of the model it's used with.

Game style (Attempts to mimic the shaded in-game graphics): "Rain Code game style."

Anime style (Specifically the art style of the game's artist, Rui Komatsuzaki. This works better in the Flux version than the Illustrious or SDXL version): "The image is in the anime art style of Rui Komatsuzaki. The image's style highlights large expressive eyes and exaggerated hairstyles. The image has a vivid and diverse color palette. There is a high level of detail in the image which include intricate patterns on clothing, accessories, and background elements. The image blends cute and charming elements with darker, more sinister themes. The image has dynamic and dramatic poses. The image features a striking blend of vibrant colors and highly expressive character designs."

Flat Rui Komatsuzaki style (AKA a more Danganronpa look): "Danganronpa sprite style."

Standard: If you don't specify any style in the prompt. The previous options are weak enough in this lora that I'd advise trying without any specified style at all to see if that might be the better option.

Some help from Chiaki

Pucci's design has a lot in common with another character that Rui Komatsuzaki portrayed - Chiaki Nanami. As such mixing my SDXL Chiaki Nanami lora in with this one within a prompt, at around strength 0.3, can help to cover elements not present in the Pucci training data. It's not a perfect fix. In particular there's usually some conflict between Pucci's rabbit hair clip and Chiaki's galaga hair clip which results in something closer to a white blob hair clip. Still, I found it works well enough to be worth mentioning. You can use it with something like "<lora:Pucci_Lavmin_SDXL:1>,<lora:Chiaki_Nanami_SDXL:0.3>, This is an anime screencap from Danganronpa 3." You should find the Chiaki Nanami lora linked to in the resources section.

Additional notes

This one was really constrained by lack of training data but I think it came out fairly well. One minor issue is that her hairclip can be a little blurred or distorted. It tends to come out better in close up shots than if she's further in the distance.

描述:

训练词语: