<div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/ItPtuRM4NwU?si=wLCWvXka7p2HwUXX" start="0"></iframe></div><span style="background-color:rgb(11, 15, 25);color:rgb(179, 188, 201);font-family:&quot;Source Sans Pro&quot;, ui-sans-serif, system-ui, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;, &quot;Segoe UI Symbol&quot;, &quot;Noto Color Emoji&quot;;font-size:16.8px">PiD reformulates the latent-to-pixel decoder as a conditional pixel-space diffusion model, unifying decoding and upsampling into a single generative module. It denoises directly in high-resolution pixel space and produces a super-resolved image in one pass. This repository hosts the released decoder checkpoints, plus the encoder/decoder ("VAE") weights they depend on.WARNING: the distilled Flux Klein 4b and 9b workflows are experimental, they dont always provide the best results due to the distillation forcing too much detail in the pixel diffusion process. it can sometimes create artifacting as well. both BASE models for Klein handle PiD VERY well though! To install the ComfyUI-PiD custom node and utilize its automated model downloader, follow these steps specifically tailored for the ComfyUI Windows Portable environment.<h3 id="1.-custom-node-installation">1. Custom Node Installation</h3>Do not just drag and drop files. For a portable installation, you must use Git to ensure the node's internal structure and submodules remain intact.<ol><li>Navigate to your <code>ComfyUI_windows_portable</code> root folder.</li><li>Open a terminal (CMD) in this folder.</li><li>Navigate to your <code>custom_nodes</code> directory:DOS<pre><code>cd ComfyUI\custom_nodes
</code></pre></li><li>Clone the repository:DOS<pre><code>git clone https://github.com/Merserk/ComfyUI-PiD
</code></pre></li><li>Return to the root folder:DOS<pre><code>cd ..\..
</code></pre></li><li>Install the required dependencies using the embedded Python:DOS<pre><code>.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\ComfyUI-PiD\requirements.txt
</code></pre></li></ol><h2 id="link-to-z-image-base-model-files:">LINK TO Z-Image Base Model Files:</h2><a target="_blank" rel="ugc" href="https://civarchive.com/models/2342907/rebels-z-image-base">https://civarchive.com/models/2342907/rebels-z-image-base</a><h3 id="2.-using-the-auto-download-feature">2. Using the Auto-Download Feature</h3>The PiD nodes are designed to pull models automatically upon the first execution of a prompt.<ol><li>Restart ComfyUI completely.</li><li>Load a workflow that utilizes the PiD Decode or PiD Prepare nodes.</li><li>Ensure the <code>auto_download</code> parameter is set to <code>True</code> in the node properties.</li><li>Queue your prompt.</li><li>Monitor the Terminal: Watch the CMD window. You will see HTTP requests appearing as the node validates and begins downloading the required weights from Hugging Face.</li></ol><h3 id="3.-troubleshooting-&quot;stalls&quot;">3. Troubleshooting "Stalls"</h3>Because there is no progress bar, a silent CMD window can look like a stall. However, if the download actually hangs (no new <code>INFO</code> lines for 5+ minutes):<ol><li>Kill the Process: Close the CMD window completely to force the download attempt to terminate.</li><li>Restart ComfyUI: Open the terminal and restart ComfyUI to clear the failed request from your GPU/Network cache.</li><li>Disable Auto-Download: Before queuing again, go into the <code>PiD Decode</code> node settings and set <code>auto_download</code> to <code>False</code>.</li><li>Verify/Retry: With <code>auto_download</code> set to <code>False</code>, the node will skip the network request and attempt to load existing files. If you suspect the download partially finished before stalling, check <code>ComfyUI\custom_nodes\ComfyUI-PiD\vendor\PiD\checkpoints</code> to see if a folder has been created. If it is empty or incomplete, remove that partial folder before trying the download again.</li></ol>Pro-Tip: If the auto-downloader consistently fails, it is usually due to network timeouts on large files. If this happens, you must perform a manual download once to place the <code>.pth</code> file into the <code>vendor</code> directory as described in the repository's documentation. Once the file exists locally, you can keep <code>auto_download</code> set to <code>False</code> permanently.

Rebels PiD (low vram)

utilizes PiD custom nodes to generate with zimage in pixel space

rebelsPidLowVram_zImageBase.zip

Z-Image-Base

Files with SHA256: 58b2912027316951ffde8cfe7752788921da5268076476ac3cd5a6ff8bc42070