Kenything版本v3 (ID: 89746)

This model is a mix of mainly popular anime models.

KenythingV5 is the last version of this model. From now on I'm going to work on a new one.

Also you can check out AIBooru, a website on which you can share your AI-generated works with others.

======================================

Best VAE: vae-ft-mse-840000-ema-pruned or kl-f8-anime2 (highres fix or sd upscaling is desirable to use)
Best upscaler: 4x-UltraSharp

Textual Inversions I recommend to use: EasyNegativeV2 and EasyNegative | bad-artist | ng_deepnegative_v1_75t | bad-hands-5

======================================

Settings I prefer to use:

Sampler: DPM++ 2M Karras | DPM++ SDE Karras | Euler a

Steps: 20-25

Hires fix: 0 steps, upscale by 2, 4x-UltraSharp upscaler, denoise 0,5

VAE: kl-f8-anime2

描述:

Slightly changed style. Better lighting, but more sensitive to prompt.

训练词语: