<p>Qwen Image 2.1 GGUF Conversion.</p><p></p><p><span style="color:rgb(250, 82, 82)"><strong>Loader nodes</strong></span>: <a target="_blank" rel="ugc" href="https://github.com/molbal/ComfyUI-GGUF">https://github.com/molbal/ComfyUI-GGUF</a> / <a target="_blank" rel="ugc" href="https://registry.comfy.org/publishers/molbal/nodes/comfyui-gguf-reboot">https://registry.comfy.org/publishers/molbal/nodes/comfyui-gguf-reboot</a></p><p></p><p></p><h2 id="workflow-settings">Workflow settings</h2><h3 id="sampler-settings">Sampler settings</h3><p>Both templates sample with <code>steps</code> 25, <code>cfg</code> 1, the <code>euler</code> sampler, and the <code>simple</code> scheduler. The text to image workflow keeps the seed fixed, and the image edit workflow randomizes it.<a target="_blank" rel="ugc" href="https://docs.comfy.org/tutorials/image/qwen/qwen-image-2-1#resolution">​</a></p><h3 id="resolution">Resolution</h3><p><strong>Text to image</strong>: the <strong>Resolution Selector</strong> node sets the aspect ratio (1:1, 16:9, 9:16, 3:2, 2:3, 4:3, 3:4, 21:9) and a megapixel target, where 1.0 MP is about 1024x1024. The template defaults to 1:1 at 1 MP. Qwen-Image-2.1 generates natively at 2K, so set the target to about 4.0 MP for a 2048x2048 square output.</p><p><strong>Image edit</strong>: with <code>custom_size</code> off, which is the template default, the output follows <code>image_1</code>’s aspect ratio, scaled to about <code>resolution</code> x <code>resolution</code> pixels (default 1024, rounded to multiples of 32). That is not the original pixel size of <code>image_1</code>. Set <code>resolution</code> to 0 to keep each reference image at its own size, rounded to a multiple of 32. Turn <code>custom_size</code> on to use the <strong>Resolution Selector</strong> canvas instead, and keep it close to the resized <code>image_1</code> size, otherwise the edit can shift.​</p><h3 id="prompting-for-image-edit">Prompting for image edit</h3><p>Reference images are spliced into the text encoder in slot order, and the prompt addresses each one by index. The template prompt is a good example of the pattern:</p><pre><code>Keep the character and pose in &lt;image1&gt; unchanged, put this light </code></pre><p></p><p></p>

Qwen Image 2.1 GGUF

qwenImage21GGUF_v10_3230961.json

Files with SHA256: df5c7d91624f57333f672d11e0cb953c715d8ca0b879ec80e9f414349f917198