pixaroma/ComfyUI-Pixaroma

What is ComfyUI‑Pixaroma?

ComfyUI‑Pixaroma is a collection of ready‑made nodes that extend the open‑source visual workflow editor ComfyUI. ComfyUI lets you build image‑generation pipelines (load a model, run a sampler, apply post‑processing, etc.) by connecting blocks on a canvas. Pixaroma adds a large toolbox of everyday‑use utilities – loading and resizing images, cropping, in‑painting, 3‑D scene building, audio‑reactive video, LoRA handling, text joining, and more – all wrapped in user‑friendly UI panels.

The repo ships the node definitions, a small set of example workflows, and a help system that lives inside ComfyUI. It is aimed at artists and developers who want a smoother, more “design‑software‑like” experience when working with diffusion models.

Key Features (as described in the README)

Category	Nodes / Tools	What they do
Image handling	Load Image / Load Image Mini / Load Images from Folder	Drop‑in replacements for ComfyUI’s native loader with built‑in resizing, resampling, aspect‑ratio presets, and live preview.
	Image Crop / Image Uncrop / Resize Crop	Visual crop UI, preserve transparency, export crop‑info for later re‑insertion, and a simple crop‑to‑fill node.
	Image Resize / Longest Side	One‑node resizing with many modes (scale, fit, pad, crop‑to‑fill) and live aspect‑ratio preview.
	Image Compare	Side‑by‑side or overlay comparison with a slider to spot differences.
Painting & Editing	Paint Pixaroma	Layer‑based paint editor with custom brushes, smudge, and eraser – useful for masks or detail work.
	Inpaint Crop / Inpaint Stitch	Guided in‑painting workflow: paint a mask, auto‑crop a model‑friendly region, run any inpaint model, then stitch the result back.
Composition	Image Composer	Layered canvas for arranging multiple images, per‑layer blur, scaling/rotation, and AI‑driven background removal.
3‑D Support	3D Builder	Simple 3‑D scene editor (shapes, imported models, lighting, camera) to generate reference depth maps or ControlNet inputs.
Audio‑Reactive Video	AudioReact Pixaroma	Turns a still image + audio track into a short video with 15 motion modes (pulse, shake, glitch, etc.) and a stack of visual effects; outputs directly to MP4.
Parameter Sweeps	XY Plot Pixaroma	Grid‑style comparison of any node parameter (numeric, sampler, LoRA, prompt) without extra wiring; seeds stay locked for fair comparison.
LoRA Management	LoRA Loader Pixaroma	Stack unlimited LoRAs in one node, toggle on/off, set strengths, view trigger words, optional Civitai lookup (API key stored locally).
Text Utilities	Text Join Pixaroma	Concatenate up to four text lines, each can be typed or wired, with custom separators and line‑renaming.
	Pause Text Pixaroma	Breakpoint that shows generated text, lets you edit or regenerate before the rest of the workflow runs.
Workflow Helpers	Pixaroma Workflows panel, Pixaroma Help, Pause Image, Save MP4, Notes & Labels, Resolution, Switches	UI panels for searching, renaming, duplicating workflows; inline help for every node; checkpoints that let you preview or skip heavy steps.

All nodes expose the usual ComfyUI inputs/outputs (e.g., IMAGE, MASK, WIDTH, HEIGHT, FILENAME) and work in both the classic and the newer node interface.

Who Might Use It?

Artists & creators who already use ComfyUI for diffusion‑based image generation and want a more Photoshop‑like toolbox without leaving the canvas.
Researchers or hobbyists building pipelines that involve multiple steps (load‑image → crop → inpaint → upscale → compare) and would benefit from visual editors for those steps.
Content producers looking to turn static AI‑generated images into short audio‑reactive videos quickly.
Anyone managing many LoRAs who needs a compact way to toggle and weight them while keeping trigger‑word metadata handy.

How to Get Started (per the README)

Install – Follow the “📥 Install” link in the README (typically git clone the repo into ComfyUI’s custom_nodes folder and restart ComfyUI).
Refresh UI – After installation, hard‑refresh the ComfyUI browser tab (Ctrl+Shift+R / Cmd+Shift+R) to load the new node icons.
Explore – Open the orange W (Workflows) or ? (Help) panels to browse example workflows and read node documentation.
Try an example – Download one of the provided JSON workflow files (e.g., Image Composer or AudioReact) and load it via the ComfyUI UI to see the node in action.
Ask for help – Use the #pixaroma‑nodes channel on the linked Discord or open an issue on the GitLab mirror if you hit bugs.

Why It Matters

ComfyUI is a powerful but low‑level graph editor; building even simple pipelines can involve many tiny nodes (load → resize → mask → upscale, etc.). Pixaroma bundles many of those routine steps into single, feature‑rich nodes and adds visual editors (crop, paint, 3‑D, audio‑react) that reduce the need to switch between external tools. This speeds up prototyping, makes workflows more reproducible, and lowers the barrier for non‑technical creators to experiment with diffusion models, LoRAs, and related AI‑generated media.

TL;DR

ComfyUI‑Pixaroma is a plug‑in pack for the ComfyUI diffusion‑workflow editor that supplies a suite of user‑friendly nodes for image loading, editing, composition, 3‑D scene creation, audio‑reactive video, LoRA management, and workflow utilities. It’s a genuine, AI‑focused open‑source project aimed at making complex image‑generation pipelines easier to build and use.