Artists Arrange AI Scenes with Compose-Plugin-Comfyui Node

    
        By vramkickedin    
     | 
    
            April 7, 2026 at 4:43 pm        
    
     | 
    
        2 min read

A new custom node for ComfyUI gives artists better control over image composition. Compose-Plugin-Comfyui lets users arrange multiple images on a canvas, resize them, rotate layers, and send the final composition directly to image generation models.

The tool was created by a developer who needed precise spatial control for professional work. It works with image editing models like Flux Klein 9b and 4b, allowing users to guide composition visually rather than relying only on text prompts.

How the layer editor works

Load up to eight images and arrange them on a canvas.
Resize, rotate, and reposition layers like in photo editing software.
Paint directly on layers or use the eraser to blend edges.
Adjust layer opacity for blending effects.
Connect output to VAE Encode or ControlNet workflows.

Artists working on complex compositions can place subjects exactly where they want them before generation. The visual approach helps when text prompts alone cannot describe precise spatial relationships. Someone building a scene with multiple elements could layer a background, add foreground subjects, and erase unwanted edges to create natural transitions.

Developer notes and current limitations

The creator describes this as a 'vibe coded node' and notes they are not a professional programmer. The project began as an experiment to test the spatial capabilities of Flux Klein. Users may encounter some issues during use. The node sometimes requires a manual reload when new images are added, and the developer mentions that oversaturation can appear in final images, though this stems from Flux Klein itself rather than the plugin.