MDMZ
patreon
WAN 2.1 ComfyUI: Installation Guide | MDMZ
🕑 Added 2025-03-05 02:27:55 +0000 UTCBefore proceeding with installation, ensure you have:
ComfyUI installed (Tutorial)
A compatible GPU (NVIDIA with at least 8GB, other GPUs might work but it's not guaranteed)
Text-to-video
Text encoder → Place inside ComfyUI/models/text_encoders/
VAE → Place inside ComfyUI/models/vae/
Video Models → Place inside ComfyUI/models/diffusion_models/
Workflow: text_to_video_wan.json
*if the link doesn't work, download the workflow from the attachments below
Image-to-video
Clip vision → Place inside ComfyUI/models/clip_vision/
Video Models → Place inside ComfyUI/models/diffusion_models/
Workflow: image_to_video_wan_example.json
*if the link doesn't work, download the workflow from the attachments below
✏️ How to Prompt for Wan 2.1: https://www.patreon.com/posts/wan-2-1-best-123662398
📊 Benchmark: Here
By following this guide, you should be able to set up and run both Text-to-Video and Image-to-Video in ComfyUI with WAN seamlessly. Happy experimenting!
💡 Need Help?
Consider upgrading your Patreon membership! Members get access to private support channels, where I’ll personally assist as much as I can to help you solve any problems.
🔗 https://discord.gg/b2jg5JgYgz
Comments
Sal Demetriou
Why is this completely different from your youtube tutorial video?
MDMZ
Which url is expired ?
X
Expired URL, and I can't download the workspace. I'm new to this sort of thing, so I'd really appreciate if you'd be able to do this for me. Please and thank you.
MDMZ
did you try installing those nodes from the manager there ?
Devon Hutton
Hey there, I'm using thinkdiffusion and just copied the image to video workflow over to my comfyai machine and when doing so, I get "Missing node types: SaveWEBM and WanImageToVideo"... any ideas? I believe I have all the necessary files saved to my machine on there
Rushikesh Desai
Unable to Import workflow
rania
yeah same I changed many setting yet I could not manage to fix it
just trollin
its not following my picture
MDMZ
They are lighter on your GPU, but doesn't mean they work no matter what, settings and hardware still matter
Pasha Neruda
Hello! In the video you said that fp8 models are more stable. But accordingly to your benchmark, there are more failed works in fp8 models. Something changed or where should i start? Thanks in advance! (RTX 4080 16gb)
