Read Me
Trained on one image of Akari from the digital artbook (3578x6080 resolution). Downscaled to (and trained on) 1024x1024.
It's not really that good since I only had 1 image to work with. It's better to use the style lora of the artbook and prompt her that way without this character lora. Less detail that way but also much less artifacts and a lot more compatible --> Pokemon Legends Arceus Artbook Style + Characters: Adaman, Akari, Arezu, Cogita, Irida, Mai, Melli, Rei, Volo, Ingo, Kamado, Ginter - Strong V1.5 E6 | Illustrious LoRA | Civitai
Trigger: 4k4r1
Helpful (and maybe even needed) tags: black hair, grey clothes, red scarf, white headwear, brown footwear
Optional: akari \(pokemon\), full body, portrait
Use 0.8 strength. 1 can work but has a lot more artifacts on higher Epochs
V1 E15 and V1 E10 have more artifacts but are more consistent. V1 E6 has a lot less artifacts but is less consistent
Please post the stuff you create down below, I'm curious to see what you guys came up with :3
Description
Adjusted settings - much less artifacts and a lot more flexibility