Nodes/ComfyUI-Addoor/🌻 Batch Image Load From Directory
ComfyUI Node

🌻 Batch Image Load From Directory

Feed a whole folder into ComfyUI

By Eagle-CN·Created 2 years ago·Updated about a year ago· 75
🌻 Batch Image Load From Directory
    • Images
    • Image_Paths
    • Image_Names_suffix
    • Image_Names
    • Count
    Directory
    Load_Cap100
    Skip_Frame0
    seed0

    Point it at a folder, get every image in it as a batch. This is the node you want when you're not making one picture - you're processing a stack of them. Bulk upscaling, running a hundred photos through the same img2img, captioning a dataset, applying one edit to a whole shoot. Core ComfyUI's Load Image handles one file; this handles the directory.

    How it works

    You give it a Directory and it loads the images inside, handing them downstream as a list along with their paths and names. Because the outputs are lists, whatever you wire them into runs once per image - that's how ComfyUI iterates a batch. It reads the common formats (jpg, jpeg, png, bmp, gif, webp), so you don't have to pre-convert anything.

    The other three inputs are about which images and how many:

    • Load_Cap - the max number to load (default 100). A guard rail so you don't accidentally pull ten thousand files into memory at once.
    • Skip_Frame - skip the first N. Combined with Load_Cap, this is your paging mechanism: process 0–100, then set Skip_Frame to 100 for the next chunk.
    • seed - present for randomized selection order; leave it be if you want deterministic behavior.

    The outputs

    Five of them, and they're more useful than they first look:

    • Images - the image batch itself.
    • Image_Paths - full paths, list form.
    • Image_Names and Image_Names_suffix - the filenames, with and without extension.
    • Count - how many it actually loaded.

    Those name and path outputs are the sleeper feature. When you're batch-processing and saving results, you almost always want the output named after its input - this is how you carry the original filename through so cat_01.png comes out as cat_01_upscaled.png instead of ComfyUI_00042.png.

    The one thing that trips people up

    Directory is a path on the machine running ComfyUI, not your local desktop if you're on a remote or cloud instance. Get that wrong and you'll get an empty load or a "directory not found." Use an absolute path you know exists on the server, and check Count on the first run to confirm it actually saw your files.

    Installing ComfyUI-Addoor

    ComfyUI Manager: Install Custom Nodes → search "ComfyUI-Addoor" → install → restart. Manual:

    cd ComfyUI/custom_nodes
    git clone https://github.com/Eagle-CN/ComfyUI-Addoor.git
    cd ComfyUI-Addoor
    pip install -r requirements.txt
    

    Restart and find it under 🌻 Addoor / Batch. For big folders, lean on Load_Cap and Skip_Frame rather than loading everything - a giant batch will eat VRAM and RAM fast.

    Category🌻 Addoor/Batch

    Inputs (4)

    NameTypeDefaultDescription
    DirectorySTRING
    Load_CapINT1001–1000
    Skip_FrameINT00–100
    seedINT00–18446744073709550000

    Outputs (5)

    NameTypeDescription
    ImagesIMAGE
    Image_PathsSTRING
    Image_Names_suffixSTRING
    Image_NamesSTRING
    CountINT