v1.5 - Lora update - New Wan 2.2 Lightx2v Moe.
v1.4 - Lynx GGUF option & minor edits.
v1.3 - Added Lynx face consistency model.
v1.2 - Color Match bypass fix and minor edits. Added new Wan 2.2 Reward loras to experiment with. Helps with overall quality, motion, color but can have weird outputs with it also.
v1.1 - Added Color Match to experiment with.
This is a modified version of this workflow.
Using the lynx face consistency model, pusa, lightxv2-MOE for wan 2.2/lightning loras its twice faster than example WanVideoWrapper workflow. Great face consistency with various NSFW loras loaded.
Downside is it is using more vram than the example WanVideoWrapper workflow, I'm not sure why but Pusa scheduler must be the reason, then again double inference speed must also be from it.
Its a looping workflow; by default you can create up to 5 videos with the nodes already present. 2x upscale & interpolation at the end.
Description
Color match bypass fix.
FAQ
Comments (7)
There seems to be something wrong with the For Loop End node, everytime i try to set total video loops to above 1, it throws this error:
!!! Exception during processing !!! Node 125.0.0.3.0.0.351 says it needs input value1, but there is no input to that node at all Traceback (most recent call last): File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 472, in execute execution_list.make_input_strong_link(unique_id, i) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^ File "C:\ComfyUI_windows_portable\ComfyUI\comfy_execution\graph.py", line 119, in make_input_strong_link raise NodeInputError(f"Node {to_node_id} says it needs input {to_input}, but there is no input to that node at all") comfy_execution.graph.NodeInputError: Node 125.0.0.3.0.0.351 says it needs input value1, but there is no input to that node at allindicating there is no value1 input the the ForLoopEnd node (im sure i got all the input right, double checked them already), and the video loop generation would be interrupted, could you tell me how to fix it?
I haven't touched that part of the workflow from the original apart from disabling value2 set and get nodes(which wasn't working for me) and manually connecting it. It's been working for me fine, I never tried 1 loop tho, I tried 3 and 5 loops. You can try the original workflow I've linked, also make sure you have all the custom nodes.
hi bro, any updates on the deep fryer issue? dont give up
Sorry Idk how to fix it, it probably has something to do with the last frame being lower quality than first input image and it keeps degrading on each loop.
I'll probably update the workflow soon but not for the matter at hand, I'll add the new lynx model for significant improvement on face consistency, still testing it.
Upscaling after color-matching reduces the frying but not perfect.
I haven't used this workflow, but in my own I use the "Image Noise Augmentation" node from kjnodes to inject noise at .130 strength on the last frame output. It does make the transition more noticeable, but the noise improves the quality and facial consistency of the next video extension. I do this after color match and upscale.
@dogboy22 Thanks. I'll try it.
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.