Simple Wan 2.1 Low vram Comfy UI Workflow (GGUF) 4gb Vram + 16gb ram

UPDATE:

-This uses the new Teacache and TorchCompilers I am gettig about 40-50% speed increase for my renders I will show more results soon.

-Quality of life: Organised with notes

-Controller Coming soon

This is a workflow i made in comfy ui it can run

1. Wan2.1 T2V

2. Wan2.1 I2V

On just a RTX3050 Laptop edition 4gb Vram + 16gb ram

I am a beginner but here is what I did:

Used the GGUF custom nodes and models

Nodes: (or use comfyui manager to install custom nodes)

https://github.com/calcuis/gguf

https://github.com/kijai/ComfyUI-WanVideoWrapper

https://github.com/FlyingFireCo/tiled_ksampler

https://github.com/kijai/ComfyUI-KJNodes

https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite

https://github.com/rgthree/rgthree-comfy

Only for controller or Stack Lora Loader

only models

Models:https://huggingface.co/calcuis/wan-gguf

I recommend using the Q_5_K_M versions its fast but stil accurate from my testing.

I am still getting usuable results with Q_2_K

install the vae,clip,clip vision,Text encoder from here

https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/tree/main/split_files

You should also use tools like Xformer,sageatten,triton or settings to speed it up a bit.

Specs:

It uses Tiled Ksampler tiled vae decoder everything tiled and about 480p quality but then using an upscaller i can get a pretty good result.

Now to the results:

V1.0

Generation for 53 frames at 480x848 plus an upscaller to 1080p at 25 steps took 5.32 hours

V1.5

Generation for 27 frames at 480x848 plus an upscaller to 1080p at 30 steps took 15minutes

If you want me to I Will make some proper tests and results

Notes:

If you want to do text use a hunyuan empty latent video