Update README.md

README.md

@@ -5,22 +5,47 @@ license: creativeml-openrail-m
 
 ===
 
+Please do mind that I'm not very active on HuggingFace.
+
+Maintaining a stable diffusion model is very resource-burning. Please consider to support me via Ko-fi.
+
+AingDiffusion will ALWAYS BE FREE.
+
+EXP models will be updated here to reduce confusion: https://civitai.com/models/52780.
+
+===
 AingDiffusion (read: Ah-eeng Diffusion) is a merge of a bunch of anime models. This model is capable of generating high-quality anime images.
 
 The word "aing" came from informal Sundanese; it means "I" or "My". The name represents that this model basically produces images that are relevant to my taste.
 
-# Guide to generate good images with this model
-From my research, this model is compatible with NovelAI's animevae and pastel-mix-vae. I highly recommend pastel-mix-vae because it produces more vibrant color than animevae.
+Guide to generate good images with this model
+
+- Use the VAE I included with the model. To setup 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**: Recommended sampler is "DPM++ SDE Karras" for AingDiffusion v7.1 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?
 
-**UPDATE [25/04/2023]**: I found an alternative VAE that you can use, ClearVAE Variant. It generates clearer images without upscaler like pastel-mix-vae did. Hi-res fix is still a good thing to do, though.
+- A: AingDiffusion and KayWaii are models I use daily, not something I merge just for gaining popularity or for the sake of download count. I make constant efforts to improve the model whenever possible and wanted to share the improvements as quickly as possible.
 
-Important: 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. All the samples I uploaded are using SwinIR 4x.
+- Q: I can't generate good images with your model.
 
-I also recommend using EasyNegative for negative prompts.
+- 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.
 
-Set Clip skip to 2, ENSD (eta noise seed delta) to 31337 and eta (noise multiplier) for ancestral samplers to 0.667.
+- Q: What's the different between AingDiffusion and KayWaii?
 
-# Samples generated using the model (More at https://civitai.com/models/34553)
+- 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 arts 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, but it has now evolved into an anime model with thick linearts.
 
-![Sample 1](sample-1.png)
-![Sample 2](sample-2.png)
+Here's a comparison:
+![Sample](AingKayComp.png)
+Keep experimenting and have fun with the models! 😄