CivArchive
    LOW (V)RAM WAN2.2 AiO T2V Workflow - v1.0
    NSFW

    🎬 Create an 8-second, 480x480px video from text input – in just 5 minutes on an average Gaming PC with 8GB VRAM & 32 GB RAM! 🎮⏱️

    Welcome to the fast lane of video generation! You're about to make a short video in just ≈5 minutes on an average Gaming PC... which means you have just enough VRAM to generate a video clip, but definitely not enough to run a small town.

    You are going to utilize the WAN2.2-14B-Rapid-AllInOne-GGUF model: a fast, optimized solution for Text-to-Video (T2V) and Image-to-Video (I2V) in a single model. It's built on the powerful WAN 2.2 14B model and optimized for speed and efficiency (FP8 precision) on consumer-grade hardware using the lightweight GGUF format.

    💡 More LoRas and higher resolutions demand more (V)RAM - getting memory allocation errors? Simply lower the number of frames (the "length" value in the resolution node)!

    What You Need
    AllInOne GGUF Models Hugging Face - WAN2.2-14B-Rapid-AllInOne-GGUF
    --> Examples were made with THIS ONE
    Encoder GGUF Hugging Face - umt5-xxl-encoder-gguf

    Description

    FAQ

    Workflows
    Wan Video 2.2 T2V-A14B

    Details

    Downloads
    403
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/11/2025
    Updated
    4/28/2026
    Deleted
    -

    Files

    lowVRAMWAN22AioT2V_v10.zip

    Mirrors

    CivitAI (1 mirrors)