VRGDG_CustomNodePack
Random Nodes
Nodes (151)
VRGDG Custom Node Pack
167 ComfyUI nodes for image and video processing, face repair, audio, local LLMs, LoRA training, and creative workflow tools.
Build face-repair workflows, process long videos, prepare music and speech, create prompts, train LoRAs, and add practical image, text, JSON, and batch utilities to your ComfyUI workflows.
What's included
| Node group | Nodes | Features | | --- | ---: | --- | | Face Fix & Video Enhance | 20 | Face detection and tracking, repair crops, enhancement anchors, video restoration, and image paste-back | | General Utilities | 70 | Text and JSON tools, image switches, film grain, sharpening, color matching, LUTs, image/video comparison, and audio splitting | | Local LLMs | 6 | Qwen prompt generation, VLM and GGUF tools, local LLM connections, and llama.cpp diagnostics | | Long Video | 12 | Overlapping video batches, audio windows, long-video loading, and LongShot planning tools | | LoRA Training | 14 | LTX, Krea, and Z-Image training tools, dataset creation, installers, and preview grids | | LTX & MiniMax H3 | 16 | Guidance and sigma controls, first/last-frame guides, looping, ingredient grids, connected chunks, still images, and continuation trimming | | Music & Audio | 19 | YuE2, SheetSage2, MiniMax Music 3 helpers, VoxCPM2 speech, and audio loading/saving | | Workflow UIs | 10 | Video Editor, Prompt Creator, Start Image Storyboard, and Node Canvas prototype |
Installation
ComfyUI Manager
- Open ComfyUI Manager and search for VRGDG Custom Node Pack or
vrgdg-customnodepack. - Install the pack and its dependencies.
- Restart ComfyUI and refresh the browser with Ctrl+F5.
If the pack is not listed yet, use the manual installation below.
Manual installation
From your ComfyUI/custom_nodes directory, run:
git clone https://github.com/vrgamegirl19/VRGDG_CustomNodePack.git
python -m pip install -r VRGDG_CustomNodePack/requirements.txt
Use the Python environment that runs ComfyUI, then restart ComfyUI and refresh your browser.
Windows portable
Run these commands from your ComfyUI_windows_portable directory:
git -C .\ComfyUI\custom_nodes clone https://github.com/vrgamegirl19/VRGDG_CustomNodePack.git
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\VRGDG_CustomNodePack\requirements.txt
Restart ComfyUI and refresh your browser with Ctrl+F5.
Getting started
Right-click the ComfyUI canvas and search for the node you want to use. Node categories include VRGDG and VRGameDevGirl.
Start with one of the included workflows:
- Face Fix
- Match Grain β Image
- Match Grain β Video
- MiniMax Music 3 Advanced Tuning
- VoxCPM2 Voice Clone / TTS
- YuE2 Starter
- LoRA Training Workflows
Download a workflow JSON and drag it onto the ComfyUI canvas. Install any additional nodes named by that workflow and select your local model files before running it.
Models and additional setup
Each tool has its own model and runtime requirements. Generation model weights are separate downloads; the pack includes the bundled face-detection assets.
- LTX looping: requires ComfyUI-LTXVideo.
- Long-video loading and VHS workflows: require ComfyUI-VideoHelperSuite.
- Local LLMs: configure a supported model or local server. GGUF tools require
llama-cpp-python; installation may need a compatible wheel or local build tools. - LoRA training: configure the external training tools, datasets, and model paths using the relevant installer and settings nodes.
- YuE2 and SheetSage2: follow the setup guide for their generation and transcription runtimes.
- Browser automation: follow the browser automation guide.
- LUTs: place your
.cubefiles inoptional_nodes/general_utilities/LUTS/. - Builder integrations: some UI actions and workflows also use nodes or routes from VRGameDevGirl's AI Video Builder.
Updating
From the installed VRGDG_CustomNodePack folder, run:
git pull
python -m pip install -r requirements.txt
Use ComfyUI's Python environment. For Windows portable, run these commands from ComfyUI_windows_portable:
git -C .\ComfyUI\custom_nodes\VRGDG_CustomNodePack pull
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\VRGDG_CustomNodePack\requirements.txt
Restart ComfyUI and refresh the browser after updating.
Troubleshooting
- Nodes are missing: check the ComfyUI startup log for the pack name and missing dependencies. Install requirements using the same Python that starts ComfyUI.
- UI buttons or panels are missing: restart ComfyUI, then refresh with Ctrl+F5.
- A workflow reports missing nodes: install the additional node packs required by that workflow.
- Duplicate nodes or extensions: keep one installed copy of each pack. Older versions of
comfyui-vrgamedevgirlmay already include these node identifiers. - Generation or training cannot start: check the selected model files, external runtime paths, and the setup instructions for that tool.
For help or bug reports, open an issue with the affected node name, the error traceback, your ComfyUI and Python versions, and a minimal workflow when possible. Remove API keys and other secrets before sharing logs or workflows.
License
The node code and associated documentation retain the VRGameDevGirl Community and Commercial License. Frontend assets retain the terms in web/LICENSE. Repository scaffolding is covered by the root Apache 2.0 license. Third-party components and model weights retain their respective licenses.