[SD 1.5][Checkpoint]DDosMix DDosMix_v2
<p>v2: 2023-02-20 I erased the roughly made picture and put in the painstakingly made picture.</p><p>Do img2img SD upscaling.</p><p>img2img SD upscale method: scale 20-25, denoising 0.2-0.3 After selecting SD Upscale at the bottom, tile overlap 64, scale factor2</p><p>caution! Sampler must be DPM++SDE karras.</p><p></p><p>If you want to make it more realistic, use DPM++ SDE Karras. If you want a clean face, use eluer a.</p><p>Have fun using it.</p><p>I used vae kl f8 anime for the example meme, but I recommend using 840000.</p><p>clip skip 2</p><p><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/sd-vae-ft-mse-original/blob/main/vae-ft-mse-840000-ema-pruned.ckpt">https://huggingface.co/stabilityai/sd-vae-ft-mse-original/blob/main/vae-ft-mse-840000-ema-pruned.ckpt</a><a target="_blank" rel="ugc" href="https://huggingface.co/AIARTCHAN/aichan_blend/tree/main/vae">https://huggingface.co/AIARTCHAN/aichan_blend/tree/main/vae</a>
Apply VAE.
You will get better color results.</p><pre><code></code></pre><p><br /></p><p></p><p>v1: I don't like the model with too many prompts and forcing you to use LoRa.</p><p>So it was created.</p><p></p><p>Let's extract high-quality images with 6-word prompts!</p><p></p><p>Check the sampler, scale and step in the example image.</p><p>Looking at the example image, there is a repeating prompt.</p><p></p><p>VAE: vae-ft-mse-840000-ema-pruned</p><p></p><p>Skip Clips: 2</p><p></p><p>It was a very simple test, so I think there will be many better results.</p><p>I didn't even do upscale testing.</p><p></p><p>Have fun while testing!</p><p>!There is one downside. It's fine for dark eyes, but for other eye colors, the pupils often collapse without upscaling.</p><pre><code></code></pre><p><br /></p>
页:
[1]