Motion Transfer Video to Video w/ WAN 2.1 SCAIL in ComfyUI — One-Click Installer
WAN 2.1 SCAIL delivers high-quality motion transfer for video-to-video tasks, combining scalable diffusion with efficient motion extraction. The updated vantagewithai workflow enables a streamlined run of WAN 2.1 SCAIL — and I created a one-click installer that sets up everything you need to run it inside ComfyUI. A few small edits to the workflow reduce compute demands while preserving motion fidelity and output quality.
Preloaded Models Within the Installer (Low VRAM)
- umt5-xxl-enc-fp8_e4m3fn.safetensors (clip folder) — Hugging Face (click "Load more files")
- wan_2.1_vae.safetensors (vae folder) — Hugging Face
- Wan21-14B-SCAIL-preview_comfy-Q3_K_S.gguf (unet folder) — Hugging Face
- 2xLexicaRRDBNet_Sharp.pth upscale model — Hugging Face
- Wan21_I2V_14B_lightx2v_cfg_step_distill_lora_rank64.safetensors Lightx2v LoRA — Hugging Face
- clip_vision_h.safetensors (clip_vision folder) — Hugging Face
- vitpose-l-wholebody.onnx (detection folder) — Hugging Face
- yolov10m.onnx (detection folder) — Hugging Face
Note: The standard WAN 2.1 14B SCAIL preview diffusion models (FP16 and FP8) are not bundled with this installer. Download them from the Kijai WanVideo Hugging Face repository and place them in your ComfyUI/models/diffusion_models folder.
Custom Node and Portable Install
- Custom Node with install command is included for the Portable ComfyUI Windows package (manual install)
- ComfyUI Manager: GitHub
- Command to install requirements:
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\ComfyUI-Manager\requirements.txt - Tip: Use ComfyUI Manager's Install Missing Custom Nodes feature to ensure all nodes install cleanly
Speed
- Generate 7-second 480p videos at 16fps (8 steps) in 10–20 minutes on an RTX A4000 (16 GB VRAM, 32 GB RAM)
- The workflow scales with your hardware — 720p and higher frame rates (24fps+) are achievable on more powerful GPUs
System Requirements
- Nvidia RTX 30XX / 40XX / 50XX GPU (FP16 supported)
- CUDA-compatible GPU with 12 GB+ VRAM (16 GB+ recommended)
- Windows OS
- Minimum 40 GB free storage
What's Included
- Portable ComfyUI Windows Installer, fully configured for WAN 2.1 14B SCAIL preview video-to-video motion transfer
- Automated downloads for all required models and nodes
- User-friendly workflow optimized for beginners and advanced users
Usage Notes
- Load either the GGUF model or WAN 2.1 SCAIL FP8 model (both compatible with the node)
- Upload a reference video to extract motion from
- Upload an image of your character in the target setting to transfer motion to
- Enter a prompt describing the movements in the generated video
- Adjust resolution, duration, and frame rate — higher settings require more compute
- Start with modest settings and scale up as your hardware allows
Buy on Patreon
Available at patreon.com/TheLocalLab

