A convenient workflow for editing images in models such as Qwen Image 2.1 and Krea 2. This workflow includes three steps: creating images in Krea 2, editing in Krea 2 using the Krea 2 Identity Edit - V2 lora, and editing images using Qwen Image 2.1 models.
Description
Required Custom Nodes (Install via ComfyUI Manager):
- ComfyUI-Easy-Use
- ComfyUI-Custom-Scripts
- ComfyUI ArtVenture
- ComfyUI-post-processing-nodes
- Pixaroma
- ComfyUI-OreX
- rgthree-comfy (Required for pipeline switches!)
- ComfyUI-VAE-Utils (Required for Tiled VAE decode)
Required Models & Exact Folder Structure:
ComfyUI/
└── models/
├── diffusion_models/
│ ├── [Qwen Image 2.1]
│ │ └── qwenvy21_v10.safetensors (int8, 6.76 GB)
│ └── [Krea 2 / Krea 2 T2I]
│ ├── cineroNGKrea2_v4cINT8.safetensors (int8, 11.95 GB, ⚠️ NSFW)
│ ├── kromaKrea2_opd.safetensors (int8, 12.57 GB)
│ ├── solordzZITZIBKrea2_solordzZerorestexp_fp8.safetensors (fp8, 12.56 GB)
│ └── loxsUtopicWorldKrea2_v40Quants_fp8.safetensors (fp8, 11.94 GB, ⚠️ NSFW)
├── text_encoders/
│ ├── qwen3vl_8b_fp8_scaled.safetensors (fp8, 10.6 GB)
│ └── qwen3VLInstruct4bHeretic_v10.safetensors (5.24 GB)
├── vae/
│ ├── texture_fix_vae_for_qwen_image_2_1_bf16.safetensors (bf16, 676 MB)
│ └── qwen_image_vae.safetensors (254 MB)
└── loras/
├── Turbo/
│ └── Qwen-Image-2.1-viggle-turbo-v0.3-6step-lora-r256.safetensors (1.36 GB)
└── Krea 2/
├── krea2_identity_edit_v2.safetensors (1.83 GB)
└── krea2_general_edit_v2.safetensors (1.83 GB)