2026-03-09 09:36:28 +01:00
2026-03-08 23:36:17 +01:00
2026-03-08 23:36:17 +01:00
2026-03-08 23:36:17 +01:00
2026-03-08 23:36:17 +01:00
2026-03-08 23:36:17 +01:00

CWK Checkpoints Preset Manager

A ComfyUI custom node that lets you save and load generation presets (sampler, scheduler, CFG, steps, clip skip, resolution, RNG) per checkpoint model — so every model always loads with its ideal settings automatically.


Features

  • 📂 Model Browser Panel — visual grid browser with thumbnails, favorites, NSFW blur, and CivitAI metadata
  • 💾 Per-model presets — save sampler, scheduler, CFG, steps, clip skip, resolution and RNG per checkpoint
  • 🎨 Custom canvas node — fully drawn node UI with inline editable preset fields, no standard widgets
  • 🔄 CivitAI integration — fetch thumbnails, model info, base model tags, and check for updates
  • 🖥️ RNG control — CPU/GPU noise generation applied directly to the model (compatible with smZNodes if installed)
  • ✂️ Clip skip — applied automatically to the CLIP model at load time, no extra node needed
  • 📐 Persistent panel — browser panel remembers its size and position across sessions

Installation

Search for CWK Checkpoints Preset Manager in the ComfyUI Manager node list.

Manual

cd ComfyUI/custom_nodes
git clone https://github.com/cowneko/CWK_Checkpoints_Preset_Manager.git

Then restart ComfyUI.


Requirements


Usage

1. Add the node

Right-click the canvas → Add Node → CWK / CWK Model Preset Manager

2. Load a model

Click 📂 Load Model to open the Model Browser. Select a model and click Load Model.

3. Edit a preset

In the node, click the ◀ ▶ arrows to step values, or click the center value to type a number directly. Dropdown rows (Sampler, Scheduler, RNG) open a native selector.

4. Save a preset

Click 💾 Update Preset to save the current values as the preset for the loaded model.

5. Reset to preset

Click ↩ Reset to restore all values to the saved preset for the current model.


Outputs

Output Type Description
MODEL MODEL Loaded model with RNG setting applied
CLIP CLIP CLIP model with clip_skip applied
VAE VAE VAE from checkpoint
sampler_name SAMPLER Preset or override sampler
scheduler SCHEDULER Preset or override scheduler
cfg FLOAT Preset or override CFG scale
steps INT Preset or override step count
width INT Preset or override width
height INT Preset or override height

clip_skip and rng are applied internally and do not appear as output pins.


Model Browser

  • Fetch Thumbnails/Infos — fetches metadata and thumbnails from CivitAI for all models
  • ↺ Rebuild Cache — re-fetches all metadata, ignoring the local cache
  • 🔍 Check Updates — checks CivitAI for newer versions of your models
  • Right-click a card — access model info, pick/set thumbnail, refresh CivitAI data, or delete the model
  • ⭐ Favorites — star models and filter to favorites only
  • Base Model filter — filter grid by architecture (SDXL, Flux, Pony, etc.)

File Structure

CWK_Checkpoints_Preset_Manager/
├── __init__.py
├── nodes.py                  # ComfyUI node definition
├── server.py                 # Aiohttp routes for API endpoints
├── checkpoint_presets.json   # Per-model presets (auto-generated, not tracked)
├── web/
│   ├── cwk_preset_manager.js # Main node canvas extension
│   ├── cwk_panel.js          # Model browser panel
│   ├── cwk_styles.js         # CSS injection
│   ├── cwk_context_menu.js   # Right-click context menu
│   └── cwk_model_info.js     # Model info overlay
└── README.md

License

MIT — see LICENSE

S
Description
No description provided
Readme
167 KiB
Languages
JavaScript 62.1%
Python 37.9%