Workflow Goal:
Provide a clean reference implementation for WAN based image-to-video generation.
This workflow focuses on simplicity, reproducibility, and education rather than including every available feature. It reflects a tested baseline for producing high-quality videos from a starting frame. Treat it as a starting point and adapt the prompt and LoRA stack for your preferred artistic direction.
IMPORTANT NOTE:
Validated using an NVIDIA RTX 2080 (8 GB VRAM) with 64 GB System RAM.
Low VRAM does not mean low hardware requirements.
WAN 2.2 Image-to-Video is computationally intensive. Even when GPU VRAM is managed efficiently, higher resolutions and longer videos can require substantial system RAM for memory paging and intermediate data.
⚠️ Hardware & Resolution Notes
While the workflow is designed to be 8 GB VRAM friendly, the available 64 GB of system RAM also helped accommodate memory-intensive operations.
If your system has:
8 GB VRAM + 16 GB RAM, expect slower generation speeds or possible out-of-memory errors.
8 GB VRAM + 32 GB RAM, results should be closer but still depend on your operating system and background memory usage.
8 GB VRAM + 64 GB RAM (or more), you are more likely to reproduce the results achieved during
----------------------------------------------------
Version 1.0
Created as a simple reference workflow for users getting started with WAN Video generation using quantized models (GGUF).
Purpose
✓ Educational
✓ Minimal
✓ Easy to extend
Verified
✓ Image-to-Video
Verified Settings
✓ CFG : 1.0
✓ Steps : 4 (2 High + 2 Low)
✓ 81 Frames (5 seconds @ 16 fps)
✓ Sampler : Euler
✓ Scheduler : sgm_uniform
These settings were used to validate this workflow and are recommended as the baseline configuration.
Verified Resolution:
✓ 480p
Hardware Tested:
✓ RTX 2080 (8 GB)
✓ 64 GB System RAM
✓ ComfyUI 0.27.0
