2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00
2025-12-16 17:24:18 +01:00
2025-12-29 18:43:40 +01:00
2025-12-17 22:12:57 +01:00
2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00
2025-12-29 18:43:40 +01:00

PROJECT-MAD-NODES

Version License Python

Advanced control for LoRA scheduling and visual prompt management in ComfyUI.

This collection focuses on distinct quality-of-life improvements: precisely controlling LoRA strength over time using visual curves and managing your prompt inspiration library directly within the node interface.


Installation

  1. Open ComfyUI Manager.
  2. Click "Install Custom Nodes".
  3. Search for PROJECT-MAD-NODES.
  4. Click Install and restart ComfyUI.

Method 2: Manual Installation

Clone this repository into your custom_nodes folder:

cd ComfyUI/custom_nodes/
git clone https://github.com/PROJECTMAD/PROJECT-MAD-NODES.git

Key Features (Nodes)

1. Multi Scheduled LoRA Loader

Stop guessing numbers. Start drawing curves.

Control Model and CLIP strength over the course of the generation steps. Instead of static values, create fade-ins, fade-outs, or complex strength patterns using a built-in graphical editor.

Multi Scheduled LoRA Loader Interface

  • Visual Curve Editor: Draw your schedule visually in a popup window. No manual coordinate typing required.
  • Independent Control: Set different curves for Model and CLIP strength for every LoRA.
  • Batch Management: Handle multiple LoRAs in a single list view with a unified timeline.
  • Grid Snapping & Resampling: Tools to help you align keyframes and smooth out your curves automatically.
  • State Management: Undo/Redo stack, import functionality, and visibility toggles.

Note: Requires connecting to the Set CLIP Hooks node (official ComfyUI feature). See the Usage Guide for configuration details.

Your personal asset browser inside ComfyUI.

Drag, drop, and reuse. This node acts as a visual container for your reference images, automatically extracting prompts to help you switch styles instantly.

Visual Prompt Gallery Interface

  • Drag & Drop Import: Simply drag images from your OS directly onto the node (saved locally to input/visual_gallery).
  • Auto-Metadata Extraction: Extracts Positive/Negative prompts from EXIF, ComfyUI PNG info, and CivitAI metadata.
  • Gallery Mode: Resizable floating window with thumbnail views (Square, Portrait, Landscape).
  • Offline Capable: Uses exifreader (cached after first load), respecting your privacy and offline workflows.

Quick Start (Workflows)

Nothing here yet.


Requirements

  • ComfyUI: Latest version recommended.
  • Internet: Only required once for the initial load of the gallery JS library (it works offline afterwards).

Credits

  • ComfyUI Team for building the incredible foundation that made this humble collection of nodes possible.
S
Description
No description provided
Readme MIT
33 MiB
Languages
JavaScript 72.8%
Python 15.9%
CSS 11.3%