DonutsDelivery SDXL Workflow (Simplified) - Upscale, Facedetailer, Ipadapter — MiniDonut v6

This ComfyUI workflow is for creating SDXL images with face detailing, IPAdapter conditioning, and upscaling; it takes text prompts and image input, then saves image output.

The workflow file includes `Donut Detailer`, `IPAdapterAdvanced`, `IPAdapterUnifiedLoader`, and `CR Upscale Image` nodes.

It also includes `Image Comparer (rgthree)`, `PreviewImage`, and `SaveImage` nodes.

The creator says the prompt in the Face related section is automatically included with the main prompt, so you do not need to enter it twice.

The creator's instructions give `/sams/` as the corrected location for the SAM model.

Node packs listed include `bleh`, `bsz`, `cg-use-everywhere`, `comfy-exp`, `comfyroll`, `ComfyUI_Comfyroll_CustomNodes`, `comfyui_essentials`, `comfyui_ipadapter_plus`, `comfyui_layerstyle`, `comfyui-custom-scripts`, `comfyui-impact-pack`, and `comfyui-impact-subpack`.

Also listed are `comfyui-inspire-pack`, `comfyui-kjnodes`, `comfyui-logic`, `ComfyUI-Long-CLIP`, `derfuu_comfyui_moddednodes`, `donutnodes`, `mikey_nodes`, `RES4LYF`, `rgthree-comfy`, and `was-node-suite-comfyui`.

Model files listed are `4x_NickelbackFS_72000_G.pth`, `bbox/Eyes.pt`, `bbox/lips_v1.pt`, `bbox/pussyV2.pt`, `bbox/yolov11m-face.pt`, `CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors`, `longclip-L.pt`, `sam_vit_b_01ec64.pth`, `SDXL/DDMv4.safetensors`, and `sdxlVAE_sdxlVAE.safetensors`

The creator recommends their SDXL models; the workflow is listed for base model SDXL 1.0.

Before installing, verify which package and release you intend to use for `bleh`, `bsz`, `comfy-exp`, `comfyroll`, `donutnodes`, `comfyui_ipadapter_plus`, and `was-node-suite-comfyui`.

Suggestion · not verified