<h1 id="resolution-normalizer-(comfyui-node)"><strong>Resolution Normalizer (ComfyUI Node)</strong></h1><p><strong>Resolution Normalizer</strong> is a utility node for ComfyUI that rescales image resolutions while preserving the aspect ratio.</p><p>The node supports multiple normalization strategies, allowing you to normalize an image resolution by its <strong>minimum side</strong>, <strong>maximum side</strong>, or <strong>the sum of width and height</strong>.</p><p>This makes it useful for preparing images for Stable Diffusion pipelines, resizing reference images, or normalizing resolutions across workflows.</p><p><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3afb6af5-3bc5-4241-9d55-4c3759d0c7e1/original=true/3afb6af5-3bc5-4241-9d55-4c3759d0c7e1.jpeg" /></p><hr /><h2 id="features"><strong>Features</strong></h2><ul><li><p>Preserve the original <strong>aspect ratio</strong></p></li><li><p>Normalize resolution using different strategies</p></li><li><p>Automatically <strong>upscale or downscale</strong></p></li><li><p>Display the resulting resolution as a formatted string (debug)</p></li></ul><hr /><h2 id="installation"><strong>Installation</strong></h2><ol><li><p>Download the zip file.</p></li><li><p>Extract it into the ComfyUI <code>custom_nodes</code> directory:</p></li><li><p>Restart ComfyUI.</p></li></ol><p>The <strong>Resolution Normalizer</strong> node will appear in the node menu.</p><hr /><h2 id="normalization-modes"><strong>Normalization Modes</strong></h2><p></p><h3 id="min-mode"><strong>Min Mode</strong></h3><p>Ensures the <strong>smallest side</strong> of the image equals the target value.</p><p>Example:</p><p>Input image: 1920 × 1080 Target value: 480 Result: 853 × 480</p><p></p><h3 id="max-mode"><strong>Max Mode</strong></h3><p>Ensures the <strong>largest side</strong> of the image equals the target value.</p><p>Example:</p><p>Input image: 1920 × 1080 Target value: 1024 Result: 1024 × 576</p><p></p><h3 id="sum-mode"><strong>Sum Mode</strong></h3><p>Ensures the <strong>sum of width and height</strong> equals the target value.</p><p>Example:</p><p>Input image: 1920 × 1080 Target value: 2000 Result: 1280 × 720</p><p></p><hr /><h2 id="example-use-cases"><strong>Example Use Cases</strong></h2><h3 id="preparing-images-for-stable-diffusion"><strong>Preparing images for Stable Diffusion</strong></h3><p>Resize images so the shortest side matches a given resolution.</p><p></p><h3 id="limiting-maximum-image-size"><strong>Limiting maximum image size</strong></h3><p>Resize images so the longest side matches a given resolution.</p><p></p><h3 id="sum-of-sides"><strong>Sum of sides</strong></h3><p>Resize images so the sum of the sides of the output image is equal to a given value.</p><p>This is useful to keep the computational cost constant.</p><p></p><hr /><h2 id="example-workflow"><strong>Example Workflow</strong></h2><p>Load Image ➔ Resolution Normalizer ➔ Empty Latent Image</p><hr /><p></p>

Files with SHA256: d259af5738ad96d36480849174c031f1f799c92a66884398d2d567f44f139387