This is a Model 2.0/2.1+ embedding with 4 vectors. To use the embedding in AUTOMATIC1111, add it to your embeddings directory in your stable-diffusion-webui folder, then invoke the embedding from your prompt.
f.e. a digital illustration of a man with dreadlocks and a hoodie looking down at something in the distance with a pink light, art by flonixsdviewv3
or
a detailed digital painting of a man and a window behind him with a moon in the background and a pink and purple background, art by flonixsdviewv3
This embedding was trained on a large variety of Sideview images to make cool sideportrait illustration. To get better results, prompt for a specific background, for example "with a pink background and a pink circle".
Please leave a heart.
Description
Sometimes V3 gets better results, V4 is better for women
To Get other Step resolution for different styles visit the huggingface site: huggingface.co/datasets/Flonixcorn/SVEmbed
V3 optimal settings:
a man with a mask on his face and a city in the background with blue lines and a orange background with a circle, art by flonixsdviewv3
Steps: 20, Sampler: Euler a, CFG scale: 11, Seed: 3127970809, Size: 768x768, Model hash: e1542d5a
I recommmend going up in steps of thousand and using the 2400 step model if you want only style and going up from there the more you want side portraits
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.