This is a character focused Anime model with 2.8D look and feel.

Some suggestions in using this model:

(VAE is not baked in)

Recommended VAE: vae-ft-mse-840000-ema-pruned.safetensors

https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main

Basic positive prompt

(highres, highly detailed:1.2), cinematic lighting, vibrant colors

Basic negative prompt

(low quality, worst quality:1.2), (low saturation, grainy, border), (watermark, signature, text:1.1)

LoRA

a) Detail Enhancer https://civarchive.com/models/82098/add-more-details-detail-enhancer-tweaker-lora

b) epi_noiseoffset https://civarchive.com/models/13941/epinoiseoffset

Embeddings you can use:

EasyNegativeV2, bad-hands-5, ng_deepnegative_v1_75t

https://huggingface.co/nolanaatama/embeddings/tree/main

BadDream

https://civarchive.com/models/72437?modelVersionId=77169

clip skip 2

Preferable sampler: DPM++ 2M SDE Karras

Sampling steps: 30

CFG: 7-8 works fine.

Resolution: 512 x 768 // 512 x 912

hires fix Upscaler: 4x-UltraSharp https://huggingface.co/uwg/upscaler/blob/main/ESRGAN/4x-UltraSharp.pth

Upscale by: 2x

hires fix steps: 10

Denoising strength: (0.4 - 0.5)

Recommend using After detailer, default settings are fine, but could also try:

Inpaint denoising stregth: 0.3

This model is for educational and non-commercial purposes, and not to cause harm, injury or violate laws.