My 2-stage (<a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0">base</a> + <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0">refiner</a>) workflows for SDXL 1.0 in <a target="_blank" rel="ugc" href="https://github.com/comfyanonymous/ComfyUI">ComfyUI</a>, with separate prompts for text encoders. If you don't need LoRA support, separate seeds, CLIP controls, or hires fix - you can just grab basic v1.0 workflow. Works with bare ComfyUI (no custom nodes needed).SD+XL workflows are variants that can use previous generations SD models to provide initial scene concept.Links to SDXL 1.0 model files:<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors">sd_xl_base_1.0.safetensors</a> or <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0_0.9vae.safetensors">sd_xl_base_1.0_0.9vae.safetensors</a>* (6.94 GB),</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/resolve/main/sd_xl_refiner_1.0.safetensors">sd_xl_refiner_1.0.safetensors</a> or <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/resolve/main/sd_xl_refiner_1.0_0.9vae.safetensors">sd_xl_refiner_1.0_0.9vae.safetensors</a>* (6.08 GB),</li><li><a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_offset_example-lora_1.0.safetensors">sd_xl_offset_example-lora_1.0.safetensors</a> (49.6 MB).</li></ul>* versions with fixed <a target="_blank" rel="ugc" href="https://github.com/huggingface/diffusers/issues/4310">issue #4310</a>.Link to ControlNet 1.1 model files (used in SD+XL v1.2):<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/lllyasviel/ControlNet-v1-1/tree/main">ControlNet 1.1</a>.</li></ul>Link to my SD 1.5 variant used in SD+XL workflow:<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/22511?modelVersionId=41974">MoonRide Mix 10</a> (you can replace it with any other SD variant you like).</li></ul>Parameters to play with:<ul><li>text in prompts,</li><li>width and height (but stick to combinations used during SDXL training - listed in notes section),</li><li>noise seeds,</li><li>steps ("end_at_step" is switching point between base and refiner - should be about 50% to 95% of total steps, depending on selected scheduler and samplers - look at preview image from base to check what is being passed to refiner stage),</li><li>scheduler,</li><li>sampler names,</li><li>cfg (lower - more creative, but also less defined),</li><li>CLIP Skip (-1 will treat prompts more literally, -3 will generalise),</li><li>LoRA strengths (0 means disabled).</li></ul>Cheers, and have fun!

MoonRide workflows for SDXL 1.0 in ComfyUI

Changes since v1.2:<ul><li>added img2img hires fix (using SDXL refiner).</li></ul>

Files with SHA256: bfeebfa9a73c55cd7fa68cd113a8cc651693bd0aa3db4b2e622a0d68c3117a26