Guide to generating good images with this model

Use the VAE I included with the model. To set up VAE, you can refer to this guide.

Use EasyNegative (NOT EasyNegativeV2, it broke the hand fix I applied) for negative prompts (highly recommended), you can add other embeddings and negative prompts alongside it as well.

IMPORTANT: The recommended sampler is "DPM++ SDE Karras" for KayWaii v2.1 and up.

Hi-res fix is a must if you want to generate high-quality and high-resolution images. For the upscaler, I highly recommend SwinIR 4x or 4x-AnimeSharp.

Important: I spotted an issue that in a rare case, the VAE broke the upscaled output. You can set the VAE to "None" if the issue occurs.

Set Clip skip to 2, ENSD (eta noise seed delta) to 31337 and eta (noise multiplier) for ancestral samplers to 0.667.

FAQ

Q: What's up with the frequent updates?

A: AingDiffusion and KayWaii are models I use daily, not something I merge just to gain popularity or for the sake of download count. I made constant efforts to improve the model whenever possible and wanted to share the improvements as quickly as possible.

Q: I can't generate good images with your model.

A: The first thing to remember is that every little change matters in the world of Stable Diffusion. Try adjusting your prompt, using different sampling methods, adding or reducing sampling steps, or adjusting the CFG scale.

Q: What's the difference between AingDiffusion and KayWaii?

A: The difference between AingDiffusion and KayWaii lies in their goals and styles. AingDiffusion aims to create an anime-style output with a touch of pastel aesthetics, as I appreciate how pastel art is colored. Back then, the art style of the pastel model available didn't suit my preferences. On the other hand, KayWaii was initially intended to generate cute girls.

Here's a comparison:

Keep experimenting and have fun with the models! 😄