BunnyFox Mix

Please carefully

read the prompt in the images provided. Consider and edit as needed.

When generating, I did not use Restore Faces to get portraits. When several characters appeared on the image, you can use Restore Faces or fix it when upscaling. I used samplers Euler a, DDIM, DPM++2M Karras at resolution 576x1024 or 1024x576.

You can use any resolution you like. Keep in mind that as the size of the generated image increases, the amount of distortion will also increase.

If possible, I will rebuild this merge again.

This model combines the following works. The merging process was carried out about 10 times in different sequences and tested for several days.

/model/3950/art-and-eros-aeros-a-tribute-to-beauty v_aEros-NoVAE

/model/4201/realistic-vision-v13 v_1.3

/model/4823/deliberate v_1.1

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