✨ WAN2.2 — Image to video — Simple Workflow
A clean, all-in-one WAN image-to-video workflow built entirely with the UmeAiRT Toolkit for ComfyUI.
Only 12 nodes. No spaghetti wires. Just load your model, write your prompt, and hit generate.
⚠️ IMPORTANT — Nodes 2.0 Required
This workflow is built for the Nodes 2.0 (Vue) interface of ComfyUI. If you don't enable it, the workflow may have display problems.
How to activate Nodes 2.0:
Open ComfyUI
Go to Settings (⚙️ icon, bottom-left)
Find "Use Nodes V2 (Vue)" and toggle it ON
Refresh the page
Load the workflow
If you prefer the classic interface, check out my Legacy version of this workflow instead (link).
🎯 Features
Text-to-Image generation
Automatic download of models in auto version
Built-in SeedVR2 upscaler — high-quality tiled upscaling (toggleable on/off) Slower than a classic upscaler, but significantly better quality
Full metadata embedding — your images are saved with all generation parameters, ready for online publishing and remixing
3 LoRA slots — with individual on/off toggles and strength control and you can connect as many other lora modules to each other for as many LoRA as you want.
📦 Custom Node Required
Only one custom node to install:
Install via ComfyUI Manager (search "UmeAiRT") or use the UmeAiRT Auto-Installer.
The Toolkit packages everything internally — upscaler, face detailer, metadata saver. No other custom nodes needed.
📂 Files you need (in manual version)
For base version
I2V Model : wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors and wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors
In models/diffusion_models
CLIP: umt5_xxl_fp8_e4m3fn_scaled.safetensors
in models/clip
For GGUF version
I2V Quant Model : wan2.2_i2v_high_noise_14B_QX.gguf and wan2.2_i2v_low_noise_14B_QX.gguf
In models/unet
Quant CLIP: umt5-xxl-encoder-QX.gguf
in models/clip
For speed version
lightning LoRA : Wan2.2-Lightning_I2V-A14B-4steps-lora_HIGH_fp16.safetensors and Wan2.2-Lightning_I2V-A14B-4steps-lora_LOW_fp16.safetensors
For AIO version
I2V FAST AIO Model : wan2.2-i2v-rapid-aio.safetensors
In models/checkpoint
VAE: wan_2.1_vae.safetensors
in models/vae
ANY upscale model:
Realistic : RealESRGAN_x4plus.pth
Anime : RealESRGAN_x4plus_anime_6B.pth
in models/upscale_models
Description
New LoRA nodes to chose high/low noise for each LoRA.
FAQ
Comments (14)
Hey Mate,
Thanks for you great Workflow.
One Question:
Is there a way to increase the FPS?
i see it's 16fps native and maybe i'm to stupid to find the option for customizing the fps :P
There's an interpolation node in post-production. Otherwise, I just modified the parameter node to allow manual FPS adjustment.
I found a new bug, It doesn't save the metadata or workflow with the video output, even with the output toggled on to save metadata. it's losing it somewhere in the flow.
where to put the high noise and the low noise, or is combined
I don't understand your question—are you talking about the model or the LoRAs?
In the manual workflow, the loader already provides an example of how to select the high/low noise levels. As for the LoRAs, you can choose exactly what you want to apply each one to.
@UmeAiRT I was talking about the order, but it worked by putting the high noises on the first tab, and the low ones on the second tab
@UmeAiRT pd epic awesome workflow, you can get 5 seconds videos in record time, not like default wan workflow.
@gonfinal12 I've incorporated quite a few optimizations to improve rendering and speed, even though it's not perfect yet
This workflow is friendly for new guy, Thankyou. Is it possible to use FLF2V?
@UmeAiRT amazing, this is what I am looking for
I see that there are various models in the manual version, when I download them, where should I put them? In the "WAN LORA 3x" sections? Also, how much ram and vram should I have to run these workflows?
I'm not saying your custom node is an obvious malware but in the current state it presents several security risks.
HF_TOKEN can be sent to modelscope via fallback.
You authorize any http domain for download which mean that anything outside of civitai, modelscope our hugginface could be used during an attack.
You use aria2c.exe in here, not saying that's a malware but you should use its python library to download and not embed an .exe.
Hey there, i saw on clip_vision is nothing installed but i did everything like you said. is this intended?