<h1 id="donut-workflow-v5">🍩 Donut Workflow V5</h1>Krea2 • DonutNodes 3.0.23+Generate, edit, inpaint and outpaint in one workflow—with organized panels, flexible references and previews for every finishing stage.💬 <a target="_blank" rel="ugc" href="https://discord.gg/5djEeQ3wpU">Discord — help, artwork &amp; updates</a> · ☕ <a target="_blank" rel="ugc" href="https://ko-fi.com/donutsdelivery">Support on Ko-fi</a> · 🛠️ <a target="_blank" rel="ugc" href="https://github.com/DonutsDelivery/ComfyUI-DonutNodes">DonutNodes on GitHub</a><h2 id="v5-highlights">✨ V5 highlights</h2><ul><li>Cleaner layout — separate panels for setup, generation, upscaling and face detail.</li><li>Inpainting &amp; outpainting — paint or draw a selection, invert it, adjust the seam, or expand with directional canvas presets. Outpainting stays within your megapixel budget.</li><li>Smart masks — describe what to select with SAM3.1, or use BiRefNet for foreground/background separation. Refine masks manually.</li><li>Independent A/B crops — frame each reference separately, with clipboard paste and optional reference guidance.</li><li>Flexible finishing — first and second upscales, Face Detailer and SeedVR2, with individual result previews.</li><li>Prompt variants — add, duplicate and reorder prompts; keep one active or cycle automatically. Includes autosizing editors and shared wildcard controls.</li><li>Searchable LoRAs — find models by typing, then adjust strengths and block weights.</li></ul><h2 id="quick-start">🚀 Quick start</h2><ol><li>Install/update DonutNodes 3.0.23+. Restart ComfyUI and refresh your browser.</li><li>Open DonutWF_v5.json.</li><li>Use Manager → Install Missing Nodes, then restart.</li><li>Install the models using one of the options below.</li><li>Review your models, LoRAs, prompt and resolution. Leave Editing off for your first text-to-image run, then hit Run.</li></ol>📦 GitHub installation: use Download missing for the supported model catalog.📦 Registry/Manager installation: download the optional <a target="_blank" rel="ugc" href="http://DonutNodes-model-installer.zip">DonutNodes-model-installer.zip</a> from this version’s files. Extract it, follow the included README and run the Windows or Linux/macOS launcher. It downloads catalog models into the appropriate ComfyUI folders and reuses matching existing files. Some downloads may require an access token.The Registry package provides model links and the separate installer link instead of the built-in downloader.<h2 id="editing-masks-and-outpainting">🎨 Editing, masks &amp; outpainting</h2>Enable Editing, add your base image to A, and optionally place a second reference in B. Describe the change in the prompt.Open Paint / outpaint… to edit a selected area or extend the canvas. Use Overlap to let generation reach into the original image and Seam width to soften the transition.For B, choose Prompt selection to mask a named object, or Auto subject for the foreground. Use full-resolution references for better results.Reference guidance is a separate option for borrowing visual elements during generation; it pauses while Editing is enabled.<h2 id="saving-and-moving-workflows">💾 Saving &amp; moving workflows</h2>The supplied preset saves to output/Final with seed-based filenames. Saving supports PNG/WebP workflow metadata.Keep your old workflow when upgrading, then copy your personal settings into V5. Reference images and wildcard files are stored separately—copy those too when moving installations.<h2 id="feedback">🐛 Feedback</h2>Found a bug? Share your workflow with private content removed, ComfyUI/DonutNodes versions, GPU/VRAM and the full error on <a target="_blank" rel="ugc" href="https://discord.gg/5djEeQ3wpU">Discord</a>.High-resolution editing with multiple references and NAG can be VRAM-heavy. Seam quality depends on the image, prompt and overlap settings.

Professional Krea 2 Workflow for high IQ individuals

DonutNodes now runs about 70-100% faster when using lora's (also than standard lora loading) on lesser vram cards (12gb and below). Remember to update. also changes to how lora's are applied when txtfusion layers are uncensored. Depending on what you want to do, it might make sense disabling the LoRA's as fineporn base model is very good at nsfw on it's own when the uncensoring is applied, which it is by default. the loras are applied for style and realism, not capability.

Files with SHA256: 5172df40348a7550142173535c0a5995d14ce617154f0a3f20cf6bc0874541ed