This is the workflow I usually use as a base for Img2Img, when i want change style, or I want improve the quality of an image. Usually I start from Txs2Img, so this is a second step.<h3 id="version-2.z">VERSION 2.Z</h3>-Improved graphics-Improved functionality and prompt division-Saver with metadata for comfyui-Ability to set 2 checkpoints, 1 for the base, 1 for the refiner-Added separate VAE and enabled it for V-PRED models-Add bookmark -Node colors coordinated with my other workflows, so as not to cause confusion-Added explanatory notes.<h1 id="general-rules:">General rules:</h1>-The template must be the same for both the base and the refiner.-The loras are optional; if you don't have any, or if you have fewer, you can delete the box.-You must (always) select the steps and the cfg based on the template you're using, including the layout. The positive and negative prompts. Other things don't change.-There are custom nodes; you can add them to have my workflow once opened.I haven't been in this wonderful world for long, so I still have a lot to learn. I might update it over time if I find something isn't working well.

Img2Img SDXL + Refiner _ ComfyUI _ Crix _ SDXL NOOBAI ILLUSTRIOUS PONY

Version 2.Z. Workflow refined, improved, and even made more aesthetically pleasing. Works with the entire XL family: SDXL, Illustrious, Pony, NoobAI.

Files with SHA256: 625da2e735423e7c2ac2a6a10f748efe726522e0eddfc83a384ebbcaed1973e1