⚠️ The I2V Vid extend workflow is highly experimental and requires high end GPUs.
I've created this Wan Video Image2Video workflows based on Kijai's workflow and the Native ComfyUI workflow, added an easy to use interface along with upscaling and frame interpolation.
Key Features:
1st pass - generating vanilla video
2nd pass upscaling using user selected upscaler and frame interpolation to 32FPS
This workflow comes pre loaded with my Wan14B RunPod template:
https://civarchive.com/articles/11960/wan14b-in-1-click-with-workflows-included-runpod-template
Join my Discord server for updates on new LoRAs, tips & tricks, Workflows, RunPod templates and the holy right to be close to an amazing charismatic person like myself:
https://discord.gg/fyha5Pzm
Description
Added an additional workflow with native ComfyUI nodes
FAQ
Comments (16)
Error: It seems that models and clips are mixed and interconnected between SDXL Base, SDXL Refiner, SD1.x, and SD2.x. Please verify.
I think you're commenting in the wrong place
I'm getting the same error with the default comfyui example workflow, haven't seen a solution yet
i hope its related to clip vision model, wher did you get clipvision for WAN?
@myprivacy27091991221 I removed the --fast command line option and that seems to have resolved the issue, give that a try and let me know if it works.
@Punchd Where do you remove that command? I'm still having this issue :(
@Zelky What is the error you're getting?
@Hearmeman I'm getting the error that says it seems that models and clips are mixed between SDXL Base, SDXL Refiner, etc. I think it might be because for some reason my models are the fp8 versions instead of the fp16 ones? I'm having trouble finding the fp16 ones though. Are they the models hosted on Civitai?
FYI: Models etc. can be downloaded here:
Has anyone been able to use sage attention with native wan nodes? It results in a black output.
I can only get sage attention to work with kijai nodes.
Hi, I am getting black screens as the output in this workflow. Other worflows work on these models. What can be wrong? https://i.imgur.com/IDzf4Nm.png
I had a similar issue with Hunyuan and restarting my computer and ComfyUI solved it (strangely). Maybe that helps?
Recently been experimenting and I'm sure what you want to do is put an empty Hunyuan Video Latent into the sampler for the latent and leave the Wan i2v node as conditioner only. Way better (and uncensored) outputs.
Hey, could you please post the JSON for that setup? I'm not exactly sure what nodes to use and what goes where.
Basic question... What is the difference between the native version and the other worlkflow...
Wan_Video_Image2Video-Upscaling_FrameInterpolation
And
Native_ComfyUI_Wan_Video_Image2Video-Upscaling_FrameInterpolation.json