ComfyUI Extension: ComfyUI-QwenEdit-Urbanism-by-UGA

Authored by perezjoan

Created

Updated

0 stars

Offline text-driven image editing for urban and architectural workflows using open-source Qwen-Image-Edit model in GGUF format with custom ComfyUI nodes for batch and stochastic image processing.

Custom Nodes (0)

    README

    🏙️ ComfyUI-QwenEdit-Urbanism-by-UGA

    License: MIT ComfyUI Urban Geo Analytics

    Qwen Image Edit for Urbanism brings localized, text-driven image editing to urban and architectural workflows using the open-source Qwen-Image-Edit model in GGUF format. All processing runs fully offline — no API, no cloud — making it ideal for architects, planners, and researchers who need privacy, reproducibility, and full visual control.

    🧠 1. Qwen Image Edit for Urbanism : Comfy Pipelines

    | Version | Description | Tutorial | |----------|--------------|-----------| | v1.0 | Basic Qwen Image Edit workflow for single-image editing. Adapts automatically to input ratio and size. | v1.0 Tutorial → | | v1.1 | Adds image editing from a reference image and advanced sampling capabilities for complex scenes. | v1.1 Tutorial → |

    Download the Workflows

    Right-click any link above and choose “Save link as…” to download the workflow file. Then, open ComfyUI → File → Load → Workflow, and select the downloaded .json file to load it into your workspace. | Version | Workflow File | |---|---| | v1.0 | Download Qwen-Edit-UGA-v1.0.json | | v1.1 | Download Gwen-Edit-UGA-v1.1.json |

    ⚙️ 2. Custom Nodes

    Custom nodes developed for urban image processing and automation within ComfyUI.

    | Node Name | Function | Description | |------------|-----------|--------------| | 🎲 Random Image Selector | Randomly selects one image among several inputs each run. | For stochastic rendering. | | 🔁 Sequential Image Loader | Loads each connected image in order across executions. | For batch processing. |

    🧩 Installation

    1. Download and install ComfyUI
    2. Open the desired workflow and install the required models. Follow the Qwen-Edit-UGA tutorials to download and place the necessary checkpoints, VAE, CLIP, and other model files inside your ComfyUI/models directory.
    3. Add custom nodes (required from v1.2). Navigate to your ComfyUI directory: ComfyUI/custom_nodes/
    4. Clone or download our repository and put the QwenUrbanismNodes folder in your custom_nodes directory https://github.com/perezjoan/ComfyUI-QwenEdit-Urbanism-by-UGA.git
    5. Restart ComfyUI.

    The nodes will appear under image/sequence and image/random categories.

    🧩 Credits

    Developed by Urban Geo Analytics (UGA)
    Based on open-source work by QuantStack, ComfyUI, and Qwen Image Edit contributors.

    🪪 License

    License: MIT