Copy this folder directly to ComfyUI\user\default\workflows and refresh your browser:
What's included:
TXT to VIDEO
IMG to VIDEO
StartEndFrames
Face to VIDEO
VIDEO EXTENSION
VIDEO to LOOP
Frames interpolations
Upscaler
Video merge
Resources you need:
📂Files :
Recommendation :
24 gb Vram: Q8_0
16 gb Vram: Q5_K_S
<12 gb Vram: Q4_K_S
T2V Quant Model: Wan2.1-T2V-14B-gguf
In models/diffusion_models
I2V Quant Model: Wan2.1-I2V-14B-480P-gguf or Wan2.1-I2V-14B-720P-gguf
In models/diffusion_models
CLIP: umt5_xxl_fp8_e4m3fn_scaled.safetensors
in models/clip
CLIP-VISION: clip_vision_h.safetensors
in models/clip_vision
VAE: wan_2.1_vae.safetensors
in models/vae
ANY upscale model (depreciated):
Realistic : RealESRGAN_x4plus.pth
Anime : RealESRGAN_x4plus_anime_6B.pth
in models/upscale_models
📦Custom Nodes :
Description
Base version
FAQ
Comments (22)
非常感谢！！
我很高兴，请毫不犹豫地提供反馈，以便我可以改进这些工作流程。
Good workflows, been using them for a while. I am having one small issue, I can not get Tensorrt nodes to install. I tried with comfy manager, manually, a fresh install, tried doing my own research and it seems like a fairly common issue with Tensorrt?
I gues you need to install TensorRT itselves. But I haven't checked if it works. https://docs.nvidia.com/deeplearning/tensorrt/latest/installing-tensorrt/installing.html#zip-file-installation
@Roshag I have tried both 12.6 and 12.8, the nodes still fail to import. I have tried running through various forum posts , running multiple commands, upgrading things, downgrading things. From what i can gather the requirements.txt is missing something. I am sure I could get it working with OP's install script, but i like to do things myself.
@uggiugi681 just tested - it works. Cuda Toolkit, TensorRT, some manipulations I can't remember, that include adding TensorRT's libraries to Cuda installation, adding some stuff to PATH and something else.
@Roshag I know it works for some people. But I get errors during install, searching the errors leads to many posts of people with the same issue going back months before WAN was even released. I have found multiple "fixes" and none of them have worked. It is simply not worth the time to troubleshoot.
@uggiugi681 if you are on windows did you try my 2.2 instalation scripts?
@UmeAiRT I did not, I am being stubborn and doing things myself like normal. =)
Great workflows!
I've been generating all of my stuff with yours
I had a few problems with one after an update but managed to fix it, the math for the mxSlider2d link had been deleted, and the crop method was reset to "true". All I did was reset crop method to center and add the expression 1-(a/100) and all perfect! ♥
Hey im having this same issue:
"MathExpression|pysssss
Compex types (LATENT/IMAGE) need to reference their width/height, e.g. a.width"
And I cant seem to find the settings you changed. Any help would be appreciated.
@teeburt undercheckpoint name click on the square (6 dots) that says image generation and the tab will open up and there you will see a lot of node links a , b , c , clip, CLIP TEXT Encode etc in the lower part you will see a tab to write in that says "Expression" greyed out.
There you put this ---> 1-(a/100)
before the update i wasnt getting oom error now i am, is there something im doing wrong
Try launch comfy with --normalvram it doesnt seem to detect my system as normal limit default anymore. Make sure u are using umt5_xxl_fp8_e4m3fn_scaled.safetensors clip and preferably a gguf wan2.1 model
@saxxie33504 Thank you, I figured it out my Invidia fallback return or something? it fixed the issue i was having
Thanks for the workflows. Not very good with Comfy, and I'm getting an error trying GGUF I2V (regular I2V seems to load error free):
Missing Node Types
When loading the graph, the following node types were not found
UnetLoaderGGUFDisTorchMultiGPU
Having installed the detected missing nodes and then trying manual install of https://github.com/pollockjj/ComfyUI-MultiGPU and restarting etc. Everything appears to be fully updated yet fails. Conflict is my best guess just not sure how to debug it further.
Check if you have this node installed https://github.com/city96/ComfyUI-GGUF
@Roshag thanks that seems to have done the trick, sorry for noob hour
Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.
There are no connections to the Tea Cache node from the comfyui-mxtoolkit. Where should I connect it?
Finally the workflow that does the job! Easy to use and quite issues-free. At least compared to all other workflows I have tried. Question - would it be possible to add a feature that would enable last frame - first frame flow for loops creation?
Can anyone pls help how to make videos more than 8s here ?