Extensions/ComfyUI-Batched-Meta
ComfyUI Extension

ComfyUI-Batched-Meta

A collection of batch-processing utility nodes for ComfyUI. These nodes read structured JSON data files and output parameters one entry at a time, driven by a seed value — making them ideal for automated batch illustration and generation workflows.

By MushroomFleet·Created 6 months ago·Updated 6 months ago· 0
MushroomFleet/ComfyUI-Batched-Meta
Nodes—
On cloudLocal install
Stars0
Updated6 months ago
Readme

ComfyUI-Batched-Meta

A collection of batch-processing utility nodes for ComfyUI. These nodes read structured JSON data files and output parameters one entry at a time, driven by a seed value — making them ideal for automated batch illustration and generation workflows.

Nodes

DJZ Batch Illustrator

Loads a JSON file containing an array of illustration entries and outputs a single entry's parameters (filename, prompt, width, height) based on a seed value. Incrementing the seed steps through the entries sequentially, wrapping around when the end is reached.

Category: DJZ/Batch

Inputs:

| Input | Type | Description | |-------|------|-------------| | json_file | Dropdown | Select a JSON file from the data directory | | seed | INT | Index selector — entry is chosen via seed % total_entries |

Outputs:

| Output | Type | Description | |--------|------|-------------| | filename | STRING | The filename stem for the current entry | | prompt | STRING | The generation prompt for the current entry | | width | INT | Image width in pixels | | height | INT | Image height in pixels |

Installation

Manual Install

  1. Navigate to your ComfyUI custom nodes directory: 
    cd ComfyUI/custom_nodes/
  2. Clone this repository: 
    git clone https://github.com/MushroomFleet/ComfyUI-Batched-Meta.git
  3. Restart ComfyUI.

No additional Python dependencies are required.

Usage

1. Prepare a JSON Data File

Create a JSON file containing an array of entry objects. Each entry must have four fields:

[
    {
        "filename": "chapter_01_cover",
        "prompt": "A lone traveller stands at the edge of a vast misty forest, dramatic lighting, painterly style",
        "width": 768,
        "height": 1024
    },
    {
        "filename": "chapter_02_village",
        "prompt": "A medieval village at dusk, warm lantern light spilling from windows, cobblestone streets",
        "width": 1024,
        "height": 768
    }
]

| Field | Type | Description | |-------|------|-------------| | filename | string | Output filename stem (no extension) | | prompt | string | The image generation prompt | | width | integer | Image width in pixels | | height | integer | Image height in pixels |

An example file (example_batch.json) is included in the batch_illustrator_data/ directory.

2. Place the JSON File

Put your JSON file into the batch_illustrator_data/ folder inside this node pack's directory:

ComfyUI/custom_nodes/ComfyUI-Batched-Meta/batch_illustrator_data/your_file.json

The directory is created automatically on first run if it does not exist.

3. Add the Node to Your Workflow

  1. In ComfyUI, right-click the canvas and search for "DJZ Batch Illustrator" (or navigate to DJZ > Batch).
  2. Select your JSON file from the json_file dropdown.
  3. Connect the outputs to your workflow:
    • filename → connect to a Save Image node's filename prefix, or any string input.
    • prompt → connect to a CLIP Text Encode (positive prompt) or any string input.
    • width / height → connect to an Empty Latent Image node or any integer inputs that control image dimensions.

4. Batch Processing with Seed

The node uses the seed value to select which entry from the JSON file to output:

  • seed = 0 → first entry
  • seed = 1 → second entry
  • seed = N → entry at index N % total_entries (wraps around)

To process the entire batch, set your workflow to queue multiple generations with incrementing seed values. For example, if your JSON has 5 entries, queue 5 generations with seeds 0–4.

Tip: Pair this node with ComfyUI's built-in batch queue or an auto-queue workflow. Each queued generation increments the seed, stepping through every entry in your JSON file automatically.

Example Workflow

[DJZ Batch Illustrator] → prompt →  [CLIP Text Encode] → [KSampler]
                        → width  →  [Empty Latent Image]
                        → height →  [Empty Latent Image]
                        → filename → [Save Image (filename_prefix)]

File Structure

ComfyUI-Batched-Meta/
├── __init__.py                          # Node pack entry point
├── DJZ_batch_illustrator.py             # Batch Illustrator node
├── batch_illustrator_data/              # JSON data directory
│   └── example_batch.json               # Example batch file
├── requirements.txt                     # Dependencies (none)
└── README.md                            # This file

Notes

  • The JSON file is cached in memory after the first load for performance. Restart ComfyUI or change the selected file to reload.
  • The node prints the current entry index and filename to the console on each execution for progress tracking.
  • Entry selection wraps around, so a seed of 5 with a 5-entry file will loop back to the first entry.

Citation

Academic Citation

If you use this codebase in your research or project, please cite:

@software{comfyui_batched_meta,
  title = {ComfyUI-Batched-Meta: Batch-processing utility nodes for ComfyUI},
  author = {Drift Johnson},
  year = {2025},
  url = {https://github.com/MushroomFleet/ComfyUI-Batched-Meta},
  version = {1.0.0}
}

Donate:

Ko-Fi