This LoRA generates images based on the style of Itaru Hinoue's artwork for Key's 2000s-era visual novels, specifically AIR.
I've tested it with a few models, including AOM2 and Anything 4.5, and it works well.
Like my MOON LoRA, this one is fantastic for getting a 90s or early 2000s aesthetic into images. You can control the strength and merge it with other models to get the perfect mix of styles.
Notes
Compared to my other LoRAs, it seems to have a more overcooked result. I would like to release an improved version in the future once I've worked out what the cause of the issue is. However, I strongly recommend using a weight of 0.8.
There are no specific tags to invoke the style, like my other LoRAs. However, this includes tagging to invoke specific characters.
Description
This was trained on 301 images with 5 repeats for 10 epochs, at a batch size of 6.
Characters from the VN are tagged: i.e. kamio misuzu, kamio haruko, toono minagi, kirishima hijiri, kirishima kano, michiru, uraha, ryuuya, kunisaki yukito but the tags have a minimal effect.