Home
Creators
Search Recent Missing Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

VACE Wan2.1 Video2Video | ComfyUI Installation Guide (Patreon)

Published:
2025-04-11 04:00:38
Imported:
2026-06
Tags:
haven't archived this post yet. have a subscription? use the importer!

Downloads

  • wanvideo_1_3B_VACE_MDMZ.json

Missing 1 full-res photo, 1 file, as this post has not yet been imported.

Content

Before 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)

Download

  • Video Model: Place inside ComfyUI\models\diffusion_models

    • rename to: "vace_preview_fp32.safetensors"

  • VAE: Place inside ComfyUI\models\vae

  • Text encoder: Place inside ComfyUI\models\text_encoders

🚀Advanced Tutorial: https://www.patreon.com/posts/advanced-video-2-127430452
*Workflows attached

💡 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

Files

Previews only

Comments

Rob Shepps

Keep getting "WanVideoModelLoader Invalid WanVideo model selected" error

shawn huang

Help!!! Please.........WanVideoVACEEncode shape '[41, 66, 8, 120, 8]' is invalid for input of size 21254400

malcolm sutherland

I have got it working, but my results look so bad. they are nowhere near as clean as yours - I have tried a bunch of different settings / samplers and videos / images, but it always looks like soup. any suggestions? (im on a 3060 with 12GB - not getting any memory errors)

MDMZ
>>172801037

did you try playing with the prompts? and don't drop the resolution too low

kally

Can you do a tutorial for video to video but with specific anime character lora? Would be fantastic

Marcus Lam

I am using Comfy workflow on ThinkDiffusion. The issue is that it does not allow me to upload a file larger than 2GB from my computer. The video model file to be renamed to "vace_preview_fp32.safetensors" is 6.7GB. So, I cannot rename it and upload. The only way is to copy the file over to ThinkDiffusion via a URL. But there is no option to rename the file with this option. Any workaround?

MDMZ
>>177789711

you dont have to rename it, just make sure you select the correct model file from the node, the name doesnt matter

J

when I run it it gets to the WanVideo Sampler and then just goes idle

MDMZ
>>182984138

check the logs, it could be running slowly in the background or running errors

Nagarajan S

I'm having issues from the output 😕

editor pc

i cant see the manager button or anything on the right side

MDMZ
>>184865210

You might need to install it manually using the command window, i put instructions on that under the comfyui installation video

editor pc

it gave me generation once did everything but now it keeps stopping at video combine. it heeps stopping . tried it in 3 different device. same issue. please help