Wan 2.2 I2V Beginner Friendly Workflow: v2.0: wan22I2VBeginner_v20.zip

Wan Video 14B i2v 720p Standard Archive
Download

Functionally it's the same as the last version, I just wanted to adjust the way the workflow looked a bit and color code the important nodes, enjoy!

Image to Video Workflow for easy setup with multiple custom loras

- INSTRUCTIONS -

This workflow is mainly for beginners who just want something easy to setup and go, I've found that most custom workflows are incredibly hard for new people to understand (myself included).

I've written all the instructions you'll need to setup and start generating your creations.

Step 1:

Download the zipped .json file from here then once you got it you'll unzip and extract it

You're going to want to locate your Comfy UI session folder on your computer and place the json file directly within the workflows folder, now finding this initially can be tricky so here's the path you wanna use:

This PC/Local Disk (C:)/Users/ [your user name]/AppData/Local/Comfy-Desktop

When you install Comfy, you have the option to do a cloud based session or create your local session that you name, so if you haven't done that, just create a new session and then when you go to this path above, you just click on the one that has the session name, ignore the others for now

Step 2: Drop the workflow in your workflows folder using this path below

C:\Users\your user here\AppData\Local\Comfy-Desktop\your session name here\ComfyUI\user\default\workflows

Once you drop it in, it should look like the image above

Step 3:

Launch Comfy UI and sign into your local session

Go to the workflows tab, it should be under models

  • Then Click and load the workflow, once you do you'll see this:

If Comfy UI gives you errors and says "missing models/nodes etc." Don't panic because that's expected to happen

Step 4:

  • You'll go to the error tab, look for an option that says "download all" and click on it

  • What will happen is that comfy UI will then download any and all the resources that are relevant to the workflow such as; models, checkpoints, custom nodes and anything else it picks up, once it's done downloading you should see the top right download notification will be green, that means that it's completed and you'll need to restart Comfy UI,

  • ALWAYS RESTART WHEN INSTALLING ANYTHING NEW IN COMFY UI

Once you've restarted the program, it should now work

Below is a simple visual guide on how the workflow steps function:

Everything is already wired and ready to go, all you have to do is just plug in the following;

  • The input image

  • Whatever loras you want to try

  • Positive prompt

Some beginner tips that may help;

  • ALWAYS use the most high quality images you can, upscale if necessary

  • For this workflow I would not mess with the Ksampler nodes unless you know what you're doing otherwise you may get all kinds of ugly outputs

  • If you get blurry videos and you haven't messed with the Ksamplers it's most likely because the image you used is low/poor quality or one of the loras you added is causing conflict (in which case just lower or higher the custom lora strengths until it comes out smooth, do NOT adjust the lightning loras strengths), I recently reinstalled everything and I was able to get this workflow up and running with solid gens the same day by using high quality images for my inputs

  • Wan 2.2 base model is limited to 6 second creations, this means that if you try to extend the duration passed 6 seconds it will either reset the starting position of the image or it will slow down the pace of the video, if you're looking to extend the duration you'll need to find a custom workflow that was built with that in mind.

  • Higher width/height dimensions are limited to your PC's capabilities and the workflow, I'm currently running this on an 5060 16GB which can do most things in good quality but after a certain point comfy UI will crash if there's not enough memory, this is partly to do with the fact that this workflow is setup to only use 4 steps and maximize speed, if you want higher visual fidelity I would suggest trying a different workflow with higher steps

  • Can't find your inputs or outputs? Search for this folder "ComfyUI-shared", they recently moved them all here so if you're trying to find your creations open it and it'll contain the folders for your "input" and "output" files

  • Where to put downloaded Lora's This is more so general knowledge but when you download loras, you want to specifically place them in this folder path below:

C:\Users\your username\AppData\Local\Comfy-Desktop\your session name\ComfyUI\models\loras

That's really it, if you have any general questions feel free to ask away in the comments, I'm not an expert by any means but I'll do what I can, best of luck and enjoy!