Custom Workflow for <a target="_blank" rel="ugc" href="https://github.com/Acly/krita-ai-diffusion">Krita's AI Diffusion Plugin</a>. Works directly from the visible canvas. Upscale, Enhance the image, Downscale to same size. Can be used as normal ComfyUI workflow with a bypass, just connect it like you need it. There 2 two groups that work in sequence, so you do not need to use the Krita (imbedded custom workflow) if you want to use a part of this idea in another comfy workflow. <a target="_blank" rel="ugc" href="https://civarchive.com/models/71980/comfyui-manager">ComfyUI</a> Manager recommended to get missing nodes, as they usually not part of the Local installation of Krita's Comfy backend.Following additional nodes are used, (<a target="_blank" rel="ugc" href="https://github.com/ssitu/ComfyUI_UltimateSDUpscale">Upscaler</a>), (<a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">JKNodes</a>), (<a target="_blank" rel="ugc" href="https://github.com/WASasquatch/was-node-suite-comfyui">WASNodes</a>), (<a target="_blank" rel="ugc" href="https://github.com/yolain/ComfyUI-Easy-Use">EasyUse</a>), (<a target="_blank" rel="ugc" href="https://github.com/pythongosssss/ComfyUI-Custom-Scripts">CustomScripts</a>) for Math, [(<a target="_blank" rel="ugc" href="https://github.com/rgthree/rgthree-comfy">RGthree</a>) Downscale], and of course the Krita Required <a target="_blank" rel="ugc" href="https://github.com/Acly/krita-ai-diffusion/wiki/ComfyUI-Setup">custom nodes</a>.What the benefit? Better details, included eye fix without enlarging the image. Also works great on already upscaled ones. At least i have better results as the build in Krita Upscaler feature (it noises a lot), without destroying already present details. At the standard values are discovered by experimentation, and seems to do for my taste. At least you do not need to sharpen afterwards.For multiple runs it can over sharpen or darken the image, as the upscale works chunk by chunk, it recommended to adjust the settings or in post to counteract this. The default 'seam' parameters work fine for the most cases, but it can produce chunk shadows on larger upscales values. Larger Upscale values works better for indicate fine details, while higher steps smoothen details. Overall works best with 30-50% denoise, depends on the picture/prompts. The Result blend very good with the source, and can be masked for fine control.Add a revision V2 with some changes, see changelog. V1 is still good for oneshot 'txt to img', V2 is better suited for 'img to img' as its softer in the defaults.V3 has some needed changes; see changelog. Mirror and description of the Text to Img workflow can be found on <a target="_blank" rel="ugc" href="https://www.deviantart.com/lavadragonx/art/XL-TTI-Base-Workflow-Comfy-UI-1284471821">Deviant Art</a>. Happy Drifting.Note: If you have problems with losing Color, saturation or brightness, use another VAE.

Krita Upscale Detailer (Workflow)

Major Changes in V3:<ul><li>Fixed Tile Logic (The "Tile Multiplier"):<ul><li>Replaced the confusing "Canvas Divide" math with a new "Tile Multiplier" slider.</li><li>0.5 = Micro-Detail (High Texture)</li><li>1.0 = Optimal Standard (Balanced) (1024x base)</li><li>1.5+ = Smoothing/Noise Reduction</li><li>Includes a math snap to 32-pixel grids to prevent border errors and artifacts.</li></ul></li><li>Improved User Interface:<ul><li>Parameters are now strictly sorted and grouped (e.g., "1a. Prompt", "2a. Upscale") for a logical workflow.</li><li>Expose Force Tiled Decode, Force Uniform tiles, Tiled Mode</li><li>Added "Use Default VAE" toggle to easily switch between checkpoint VAEs and external high-quality VAEs (like SharpSpectrum).</li></ul></li><li>Optimized Defaults:<ul><li>Parameters defaults are "smoother" then previous versions; </li><li>Tuned defaults prioritize quality/time (Lower step counts, optimized Denoise strengths).</li><li>Half Tile Seam Fix is now default for invisible borders on detailed images.</li></ul></li><li>Dependency Updates:<ul><li>Important: Updated <code>Was Node (ns)</code> compatibility. Please install the recommended fork as the original is retired.</li></ul></li><li>Combined Text to IMG (with IMG to IMG) Workflow.<ul><li>Added a "XL Base Workflow" with Highers fix for Comfy UI users that has the "KUD" Workflow integrated.</li></ul></li></ul>

Files with SHA256: d851159d89d9b4e5fd4046c6a18f08e65ca25bd0315b2780d639a100f15d9342