🎯 Compact Wan Workflow — Simplify Your Setup 🚀 — v1.0

⚠️ Important Warning

When copying nodes into ComfyUI, parameters can sometimes shift (known bugs).

👉 To avoid issues:

• Use only unpacked versions , as in this project.

• All nodes are separated and thoroughly tested — safely copyable!

• Ensure all components are connected in the correct order (especially Wan Setup ).

Node Parameters

• ?input - optional input

• _output - hidden output (often used for debugging)

• [input/output] - input/output within a single iteration

🎨 Color logic for nodes

• Yellow — Useful utilities

• Purple — Pipeline settings and configuration

• Cyan — Conditioning nodes

• Green — Samplers

• Red — Coders/decoders (VAE)

• Purple-blue — Conditional blocks (branching logic)

• Blue — "Everything in one" — powerful compact nodes

• Black — Specific to Wan Animate, but essentially the same utilities ✨

🛠️ Connected Custom Nodes

• ComfyUI-GGUF

• ComfyUI-wanBlockswap

• ComfyUI-MagCache

• rgthree-comfy

• ComfyUI-KJNodes

• ComfyUI-Easy-Use

• comfyui_controlnet_aux

• ComfyUI-VideoHelperSuite

• ComfyUI-Frame-Interpolation

• ComfyUI-segment-anything-2

• ComfyUI-SAM2

📦 Models

1. Wan_2.2_ComfyUI_Repackaged

2. wan2-gguf (Calcuis Repackaged)

3. WanVideo_comfy (Kijai Repackaged)

🔍 Optimization Recommendations

Below is a table of parameters that can reduce VRAM consumption or speed up video generation:

| Node | Parameter | Impact | Performance | Feature |

| -------------------- | ---------------- | ------ | --------------- | ----------------------- |

| Wan Setup->Load Wan | GGUF | strong | enable | reduces VRAM, speeds up |

| Wan Setup->Load Clip | GGUF | medium | enable | reduces VRAM, speeds up |

| Wan Optimizer | Sage Attention | strong | auto | reduces VRAM, speeds up |

| Wan Optimizer | FP16 | low | enable | reduces VRAM |

| Wan Optimizer | MagCache | medium | enable | speeds up |

| Wan Optimizer | Compile | medium | enable | reduces VRAM, speeds up |

| Wan Optimizer | Block swap | strong | higher = better | reduces VRAM |

| Image Normalize | is_scale | strong | enable | reduces VRAM, speeds up |

| Image Normalize | megapixels | strong | lower = better | reduces VRAM, speeds up |

| Decode | VAE Tiled Decode | strong | enable | reduces VRAM |

Best settings for low VRAM (6–8 GB):

• Wan Setup->Load Wan - GGUF - Q4_K_M

• Wan Setup->Load Clip - GGUF - Q4_K_M

• Wan Optimizer - Sage Attention - auto

• Wan Optimizer - FP16

• Wan Optimizer - Block swap - 40, if working with high-resolution videos or Wan Animate.

• Image Normalize - megapixels - 0.21 for Wan 2.2 14B

• Decode - VAE Tiled Decode

💾 Tip: When RAM is insufficient — set up virtual memory or use Mem Reduct .

💻 My Test Configuration

• 🎮 GPU: RTX 3060 Laptop (6GB)

• 🧠 RAM: 24 GB + 32 GB swap

• ⚡ CPU: Intel i5-11300H

• 💻 Laptop: Asus TUF Dash F15

Runs stably even under these conditions 🏆.

📜 License

I don't know why it's here, but use at your own risk ̄_(ツ)_/ ̄

👤 Author

Created by NeuroContent ✨

• CivitAI 🧠

• Instagram 🎨

• Pinterest 📌

• Donate ☕

TODO

• Add nodes for T2I

• Add clip_vision

• Support S2V

• Support Lucy Edit

• Support Vace Fun

• Add EasyCache and LazyCache to Wan Optimizer

• Add support for generating long videos through the Wan Context Windows node

Changelog

1.1

• Fix for the new version of ComfyUI

• Standard and tile-based encode/decode are now combined into two compact nodes

• Support for clip_vision has been added. You can now enable/disable clip_vision in setup. A new utility node for optional use of clip_vision has been added. It is now connected to each conditioning node, resulting in additional conditioning nodes.

• The reduce_vram parameter has been removed from Optimizer to improve node usage experience

• EasyCache and LazyCache optimizers have been added

• Fix for the negative parameter in Wan Full Setup

Checkpoint · Hugging Face · VMTamashii/rife49