Nodes/comfy_Pond_Nodes/🐳GIF加载
ComfyUI Node

🐳GIF加载

The Load GIF node ComfyUI somehow still doesn't ship

By Pondowner857·Created about a year ago·Updated 21 days ago· 45
🐳GIF加载
    • images
    • frame_count
    • width
    • height
    gif

    ComfyUI will happily load a single PNG and a whole video file, but a plain GIF? It just... won't. 🐳GIF加载 (Load GIF) is the tiny adapter that fills that hole: pick a .gif from your input folder, and it hands you every frame as an IMAGE batch, plus the frame count and dimensions as integers.

    That unlocks a whole category of workflows that are annoyingly manual otherwise. Restyle a sticker or a reaction GIF frame-by-frame with img2img. Upscale every frame of a short looping animation. Pull a GIF apart, edit a frame, and reassemble it with the pack's ImagesToGIF + SaveGIF nodes. If you've ever screenshotted a GIF frame by frame like it's 2008, you know why this exists.

    How it works

    The node lists .gif files from ComfyUI's input folder (the widget lets you upload one straight from disk, standard ComfyUI behavior). Under the hood it opens the file with Pillow and walks through the frames with seek() until it hits the end - each frame converted to an RGB tensor - then concatenates them into one batch. Simple, correct, and it handles GIFs with hundreds of frames fine, though the resulting batch can get memory-heavy.

    Inputs and outputs

    Input is one thing:

    • gif - a file picker showing .gif files in ComfyUI/input.

    Outputs are the useful part:

    • images - the frame batch (IMAGE). This is what you feed into img2img, an upscaler, or frame-processing nodes.
    • frame_count - how many frames there are (INT). Handy for math - "process every 3rd frame," "repeat this by N."
    • width, height - the frame dimensions (INTs), so you can wire them into resize or latent nodes without a separate size-getter.

    A few honest notes: the node converts every frame to RGB, so any transparency in the original GIF becomes white. And if your GIF is huge or has a thousand frames, the batch it returns is huge - that's physics, but know it before you wire it into something that processes the whole batch eagerly.

    Install

    Standard Pond Nodes install:

    ComfyUI Manager → search "Pond Nodes" (comfy_Pond_Nodes)
    

    or:

    cd ComfyUI/custom_nodes
    git clone https://github.com/Pondowner857/comfy_Pond_Nodes
    cd comfy_Pond_Nodes
    pip install -r requirements.txt
    

    Restart after. This node only needs Pillow and numpy - the rest of that requirements.txt is for the pack's heavier nodes, and nothing here needs a model download. Pack-level warning for the record: the README documents a console-spam conflict with comfyui_HiDream-Sampler if both are installed.

    Category🐳Pond/GIF

    Inputs (1)

    NameTypeDefaultDescription
    gifCOMBO1 options:

    Outputs (4)

    NameTypeDescription
    imagesIMAGE
    frame_countINT
    widthINT
    heightINT