First - original workflow https://civarchive.com/models/1829052/wan-22-for-loop-video-with-scenario (All credits to thesharque588) I only made little (since version 1.5 it's not so small anymore=) ) upgrade.
Second - how this work https://civarchive.com/articles/19108
Third - I'm not an expert. I'm a noob. Most of the time, I don't even know what I'm doing. It works for me, but that doesn't mean I can help you. Sorry.
What I did:
Color correction deleted. Sorry, but this is crap
Add 2 Clean VRam
Add 3 UnloadModel
Added node "Force/set Clip Device" (1.5 update)
Added Sage Attention (2.0 update)
Added more nodes for Loras. 4 Loras nodes for each clip (2.5 update)
Replace node
WanImageToVideowithPainterI2V(2.7 update)
The result was: stable work on the RTX 3060 +32gb ram with a length of 65 and above. Before, everything froze and didn't work.
Explanation for 1.5 update:
Don't use it if you have an older CPU without AVX2 instructions! (click on node "Force/set Clip Device" and press Ctrl+B to disable)
If you have a super-powerful CPU, you can turn it on and never turn it off. But if you have a super computer, you probably don't need this workflow =)
Explanation for 2.0 update:
I added Sage to this workflow. This is completely optional (Ctrl+B to disable), but I recommend figuring out how to set it up. It's a huge speed boost!
Note in patch 1.5 after dozens of tests. "Force/set Clip Device" - With this node, generation time will increase slightly, but your PC will feel much better. I recommend leaving this node running to preserve your PC's health =) If you're willing to do anything to improve generation speed, then turn it off.
Explanation for 2.5 update:
Added more nodes for Loras. Now you can enable up to 4 Loras nodes for each clip (2 main and 2 additional). All of this is optional.
Explanation for 2.7 update:
Replace node WanImageToVideo with PainterI2V (https://github.com/princepainter/ComfyUI-PainterI2V).
Small changes + help update.
Easy install additional nodes for ComfyUI Portable:
ComfyUI Manager - https://www.comfyonline.app/comfyui-nodes/ComfyUI-Manager
How install Triton + Sage Attention (I used this guide and it work!):
Good Luck Everyone! Share the results of your generations!
Description
Replace node WanImageToVideo with PainterI2V
FAQ
Comments (11)
Where do I add frame interpolation? I'm confused by the iterations
@Rifler1 - I feel like I've tried many things - but still.. I cannot get it to go beyond 5 seconds.
I input prompt.. I make sure I hit "enter".. but no matter what.. even changing the segment number to 2..3...4.. nothing.. just 5 secs.
Any ideas where I am missing something?
I don't fully understand the problem, but:
1. TotalStepCount node (e.g. 3)
2. PainterI2V node (WanImageToVideo) - length > 81 (65 = 4 sec, 81 = 5 sec, etc.)
3. String Selector node - every line (the end of the line is where you pressed enter) prompt to each step
Total - TSC/3 + length/81 = 15 sec video
A bunch of models is missing, where can i find them?
SVI models soooo good. And work perfect with this workflow!
I recommend - https://civitai.com/models/2053259/wan-22-enhanced-nsfw-or-svi-or-camera-prompt-adherence-lightning-edition-i2v-and-t2v-fp8-gguf?modelVersionId=2625965
Total noob here. All the nodes seem to be working except that last node? Final Beep? Where do I get this? I tried using comfyui's manager, but it wont find it.
Final Beep? Can you be more precise?
@Rifler1 Sorry the node says Final Beep. Its the last node in the workflow. Comfyui says I'm missing 3 nodes, (Notif-PlaySound, ReActorFaceSwapOpt, and ReActorOptions). I've search the comfyui manager and google, but i cant find these. I know I'm doing something wrong, but I'm not sure what. Thanks in advance for any info you can provide.
@KBZShadow Is this really in this workflow? There's nothing wrong with it. I experimented with a reactor a long time ago (unsuccessfully), but deleted it completely. Notif-PlaySound never existed.
@Rifler1 Sorry bro! I figured it out. It was because I was using the new comfyui node manager instead of the legacy one. The new manager doesnt find the required nodes. Noob mistake. Thank you again!
This workflow can't handle more than 5 steps. If you try 6 or more steps, an error occurs. This is likely because the lora switcher only has 5 iterations for five different loras. I think in most cases, additional loras aren't needed; the main lora is sufficient (but if you need it, just add new loras and add switch step). I'd like to see a version of this workflow with unlimited steps, so i just turn off lora switcher and disabled all additional loras and put connector direct to sampler. After that, I easily got a long video with ten steps. I also added an interpolation node. It's amazing to be able to create a two-minute video with this workflow. Now I'm thinking about how to fix the color-loss problem, since after six steps or more (even with SVI Pro) the colors become slightly degraded from the source image.