4 Step Generator For Comfy UI (Face Fix, Highres) — V.2 Updated

This Comfy UI workflow is for four-step generation with face fixing and high-resolution processing; it uses SDXL 1.0 and the listed model files and node packs, while no user prompt, image input, or output format is specified.

Referenced files include pancakeMix_v30.safetensors (checkpoint), papercut.safetensors (LoRA), sharpspectrumvaexl_v1.safetensors (VAE), ponyEnhancedVAE_pastelHelperV1.safetensors (VAE), vaemix_v6.safetensors (VAE), and xlVAEC_g95.safetensors (VAE).

Other referenced files are v1-5-pruned-emaonly-fp16.safetensors (checkpoint), bbox/face_yolov8m.pt (object detector), and sam_vit_b_01ec64.pth (segmentation model).

The listing identifies the item as a Workflow from LitchCastle and names SDXL 1.0 as its base model.

The named node packs are comfyui-easy-use, comfyui-impact-pack, comfyui-impact-subpack, and rgthree-comfy.

The listed node-pack versions are comfyui-easy-use 1.3.1, comfyui-impact-pack 8.17.1, comfyui-impact-subpack 1.3.4, and rgthree-comfy 1.0.2506230015.

For setup, match each named file to its listed role: checkpoint, LoRA, VAE, object detector, or segmentation model.

Vorschlag · nicht geprüft

Before use, verify that pancakeMix_v30.safetensors, papercut.safetensors, v1-5-pruned-emaonly-fp16.safetensors, bbox/face_yolov8m.pt, and sam_vit_b_01ec64.pth are the intended files.

Vorschlag · nicht geprüft

No license is listed; check the applicable usage terms before sharing or reusing the workflow.

Vorschlag