Update-20250730：
If you find the workflow isn't working right, go ahead and redownload.
Just finished my first two Wan2.2 LoRAs today—one trained on the high-noise checkpoint, the other on the low-noise—so I’ve updated the workflow for a couple of reasons:
Turns out the Low-noise checkpoint is the real final boss. Some people have already spotted that if you only call your character LoRA during the high-noise stage, the Low-noise pass will basically torch everything you just did.
Think of it this way:
A = LoRA trained only on the high-noise ckpt
B = LoRA trained only on the low-noise ckpt
A+B = both combined
Quality comes out in this order: A+B > B > A
The uploaded zip file contains two workflows: I2V and T2I.
They're tweaked versions of the ComfyUI workflow templates - just minor adjustments, arranged the way I like, and with modified generation parameters (I put my test results in my daily update article).
Also, go with the flow, I added upscaling and noise nodes to T2I. They're simple nodes, not using a repaint upscaling method like A1111, to make sure they can run on my PC.
Description
FAQ
Comments (17)
Both of these work great.
Great work!!
but why i stiil got this error.
KSamplerAdvanced
Given groups=1, weight of size [5120, 36, 1, 2, 2], expected input[1, 32, 17, 90, 60] to have 36 channels, but got 32 channels instead
If you're using the I2V feature, make sure the input image dimensions are divisible by 32.
https://civitai.com/images/90220958
Also, double-check that your checkpoint and VAE are correct.
SamLiu Thx, my problem has been resolved. I was using the incorrect VAE.
Same error for me. And I have all like in the example image T__T
saehara151 Delete the node flow2-wan-video using the node manager. This faulty node (that is not in an active workflow) prevents wan2.2 i2v from working. It may seem weird, but the wonderful way comfyUI is coded allows none active nodes to break active nodes (is there anything ComfyUI gets right?)
blobby99 Thanks, but when I went to the ComfyUi Manager and entered the name of that node, it tells me that I don't have it installed. Maybe there's another node out there that's also causing problems.
My input image is ignored. It creates an output video based soley on the prompt. And yes, I am using the i2v workflow. What could I be doing wrong?
sounds like you're loading a t2v model.
lorenzo881743 you are exactly right. I am dumb. Thank you!
If you see overlapping frames, try bumping the sampling steps to 8.
0-4, 4-8
Do you see "LoRa key not loaded..." errors with lightx?
We’re gonna need a new step-distill LoRA.
Where should I put the missing node “Film Grain”?
Just delete it if you don’t want it.
custom_nodes
Warning- the i2v error "groups=1..." is caused by the node flow2-wan-video. Delete this node using comfyUI manager. And NO, this node breaks your workflow by simply being installed, and not by being part of your workflow. It is insane that comfyUI allows non-active code to break active code.