ComfyUI DeepFuze CUDA installation

DeepFuze is a state-of-the-art deep learning tool that seamlessly integrates with ComfyUI to revolutionize facial transformations, lipsyncing, video generation, voice cloning, face swapping, and lipsync translation. Leveraging advanced algorithms, DeepFuze enables users to combine audio and video with unparalleled realism, ensuring perfectly synchronized facial movements. This innovative solution is ideal for content creators, animators, developers, and anyone seeking to elevate their video editing projects with sophisticated AI-driven features.

[Gundi]

Hallo
auch ich habe alle Installationsschritte Schritt für Schritt ausgeführt.
Leider bekomme ich folgende Fehlermeldung:
DeepFuzeFaceSwap
erwartetes str-, bytes- oder os.PathLike-Objekt, nicht NoneType
Wo könnte mein Fehler sein?
Danke im voraus

[jeandupont]

I think i won't be the only one to be really greatfull for some update news. Are the nodes still working as of today? 2025?

[AdelEmam-dz]

I get the following error:
DeepFuzeFaceSwap
expected str, bytes or os.PathLike object, not NoneType

[Audiopizzamusic]

Hi! Amazing tool. But I'm getting this error:

DeepFuzeAdavance
could not be loaded with cv."

I did absolutely everything as instructed. Installed CUDA 11.8. Added new environment paths. Replaced the files with version 8.9.2.26. Checked that this version of CUDA is used. Installed onnxruntime and onnxruntime gpu.


SOLVED: Turns out I didn't have ffmpeg installed. And it was not listed as a required library in the installation instructions.

[samu]

I've your result and they're amazing! I've followed the instructions to the letter. Everything should be in place. But I get this error:

Warning: Missing Node Types
When loading the graph, the following node types were not found:

DeepFuzeFaceSwap

What am I missing? I've googled it but so far I can't find a solution

Edit: I've been told the node doesn't work anymore?

[plasmar]

RunTimeError: The NVIDA driver on your system is too old (found version 11080) is what I get trying ** for myself just installing latest version cuda drivers and cudnn seems to let comfyui work but am specifically trying to get passed pytorch allocating a bunch vram(I'm not sure if this is a normal behavior) ** nothing this video seems to be a fork of comfyui I think

[ppaayyaamm]

Hi Thank you for good video, I Have RTX 4070Ti that i think suport cuda 12.0 and above, is it not Working in this way, Or Is the way to use cuda from 4070ti for Comfyui ? ( I am not expert ) thank you

[vincentnaples]

heads up, you should probably burn that openAI API KEY

[samu]

I'm getting this error, how could I fix it? Please help, going crazy here :(

could not be loaded with cv.

Also, it's not uding Cuda, but Cpu before it gives the error, although it takes 2 seconds, so I guess it's not really working. I installed Cuda to the letter.