ModelsLab/SakushiMix (Finished) - V1.2

For this model, I have set out to depict typical anime appearances of the newer generation with a touch of realism, without compromising the character's style. It was also a priority for me to ensure a diverse range of colors.

Furthermore, this model is well-suited for depicting explicit content.

__________________________________________________________________________________________________

Recommended parameters:

((masterpiece)), (best quality),

Negatives: (worst quality, low quality:1.4),

Sampler: Euler a: 20 to 30 steps.

Sampler: DPM++ SDE Karras: 20 to 30 steps.

Sampler: DPM++ 2M Karras: 20 to 30 steps.

Sampler: DPM++ 2M SDE Karras: 20 to 30 steps.

CFG Scale: 7 to 12. Recommend: 8

Resolutions: 512x768, 512x1024 for Portrait!

Resolutions: 768x512, 1024x512, 1536x512 for Landscape!

Hires.fix: RealESRGAN_x4plus_anime_6B, with 20 steps at 0.5 - 0.6 denoising, upscale 2.0.

Clip Skip: 2

Eta noise seed delta: 31337

Textual Inversion

__________________________________________________________________________________________________

Bad-Hand-5

https://huggingface.co/yesyeahvh/bad-hands-5/tree/main

NegfeetV2

https://huggingface.co/qdas/negfeet/blob/main/NegfeetV2.pt

VAE

_________________________________________________________________________________________________

kl-f8-anime2.vae

https://huggingface.co/Norisuke193/kl-f8-anime2/blob/main/kl-f8-anime2.vae.pt

vae-ft-mse-840000-ema-pruned

https://huggingface.co/stabilityai/sd-vae-ft-mse-original/blob/main/vae-ft-mse-840000-ema-pruned.safetensors

__________________________________________________________________________________________________

Base of Danbooru

https://danbooru.donmai.us/wiki_pages/tag_groups