CivArchive
    Wan Video T2V - Upscaling & Frame Interpolation - SLG/LoRA/TeaCache/SageAtt
    NSFW

    I've created this Wan Video Text2Video workflow using ComfyUI's native Wan nodes.

    Key Features:

    • 1st pass - generating vanilla video

    • 2nd pass upscaling using user selected upscaler and frame interpolation to 32/60FPS

    This workflow comes pre loaded with my Wan2.1 RunPod template:
    https://civarchive.com/articles/11960/wan14b-in-1-click-with-workflows-included-runpod-template

    Description

    • Added a Power LoRA Loader

    • Added Skip Layer Guidance

    • Added Wan Video TeaCache node

    FAQ

    Comments (18)

    CyberoMar 23, 2025
    CivitAI

    Everything prefect works, you're the best! Thanks!

    jefharrisMar 28, 2025
    CivitAI

    Works great! Nice and clean.

    zczcgMar 29, 2025
    CivitAI

    A question:

    Failed to validate prompt for output 250:

    * KSampler 230:

    - Required input is missing: negative

    If I wanna execute it, i must connect the negative prompt to #228 clip text encoder. Does I do right?

    HearmemanAI
    Author
    Mar 29, 2025

    Which workflow is this?

    zczcgMar 29, 2025

    The lastest version: SLG/LoRA/TeaCache/SageAtt I dl it and having this problem

    HearmemanAI
    Author
    Mar 29, 2025

    @zczcg Did you download it from CivitAI? Or in my RunPod template?

    zczcgMar 29, 2025

    @Hearmeman  I download it from civitai,YES

    1MoreByteMar 29, 2025
    CivitAI

    SimpleMath.execute() got an unexpected keyword argument 'c Do you know how to fix this error for the height node?

    TanKalosi609Mar 31, 2025
    CivitAI

    Any chance you could implement the new Endframe workflow into this? I've been exclusively using your workflows for WAN, but I'm dying to try it with a specified Endframe. Couldn't make it work myself in your template, unfortunately. Thanks!

    sixchinApr 12, 2025
    CivitAI

    Great workflow, produces 5s video really fast compared to other workflows. I am having an issue where 512x512 videos can be produced back to back at a consistent speed. However, when I try to produce videos at other resolutions like 480x832, the first video is generated at a reasonable time, but everything after is 3x the time.

    thaddeuskJun 1, 2025· 2 reactions

    Higher resolution videos require more VRAM, and it sounds likely that you're running out a VRAM, causing it to store some in your system RAM, which makes it massively slower. Maybe try using the fp8 model or a smaller quant GGUF?

    saybaka69Apr 15, 2025
    CivitAI

    If I run this workflow as-is, I get a KSampler error - Flux.forward() missing 1 required positional argument: 'y'. What do I need to add or change to fix this?

    ForbiddenPlanetApr 18, 2025· 1 reaction
    CivitAI

    After loading this up I got a huge list of errors and now ComfyUI will no longer start

    ForbiddenPlanetApr 18, 2025

    After restarting my PC comfy loads, now I get an error for missing node "INT" which I can't find anywhere?

    lolologistApr 28, 2025

    Update your custom nodes.

    jarppa203Jun 2, 2025

    usually its bcoz you need install custom nodes(yet again something else). comfyui will boot if not do anything. it might not boot if install. workflow not do anything harm.

    wan need comfyui to be updated. nodes u need install. so better install custom node manager or what eva it is. makes it easier. couple workflows and you have so many nodes you not even know what they do.

    even one defauld node would do same. peepol like yet again new same nodes. ufff

    honryindianMay 27, 2025· 3 reactions
    CivitAI

    Your workflow is simpler than others, but works better than them. Amazing work!

    HearmemanAI
    Author
    Jun 1, 2025

    Many thanks!

    Workflows
    Wan Video

    Details

    Downloads
    3,489
    Platform
    CivitAI
    Platform Status
    Available
    Created
    3/23/2025
    Updated
    8/11/2026
    Deleted
    -

    Files

    wanVideoT2VUpscaling_slgLoraTeacache.zip