aiNodes is a simple and easy-to-use Python-based AI image / motion picture generator node engine.
The engine comes packed with various features and functionalities, and some of the backend and nodes are heavily inspired by ComfyUI, a web-based node engine repository.
- Full modularity - download node packs on runtime
- RIFE (Real-Time Intermediate Flow Estimation) support
- FILM interpolation integration
- Coloured background drop
- Easy node creation with IDE annotations
To get started with aiNodes, follow the steps below:
Requirements:
- Python 3.10 (https://www.python.org/ftp/python/3.10.0/python-3.10.0-amd64.exe)
- Git (https://github.com/git-for-windows/git/releases/download/v2.40.1.windows.1/Git-2.40.1-64-bit.exe)
- nVidia GPU with CUDA and drivers installed
Windows:
- Download the 1 Click Installer from the releases on the right menu and run it in a folder of your choice
- It will create a virtual environment, and install all dependencies, to start next time, you can use the shortcut on your Desktop.
Linux:
git clone https://github.com/XmYx/ainodes-engine
cd ainodes-engine
python -m venv nodes_env
source nodes_env/bin/activate
pip install -r requirements.txt
cd custom_nodes
git clone https://github.com/XmYx/ainodes_engine_base_nodes
cd ainodes_engine_base_nodes
pip install -r requirements.txt
cd ..
git clone https://github.com/XmYx/ainodes_engine_deforum_nodes
cd ainodes_engine_deforum_nodes
pip install -r requirements.txt
Once the app is up and running, you can start check the File - Example Graphs option to start creating, and you can also access your model folders from the File menu.
Contributions to the Ainodes Engine are welcome and appreciated. If you find any bugs or issues with the app, please feel free to open an issue or submit a pull request.
aiNodes is an open source desktop ai based image / motion generator, editor suite designed to be flexible, and with an Unreal-like execution chain. It natively supports:
- Deforum
- Stable Diffusion 1.5 / 2.0 / 2.1
- Upscalers
- Kandinsky
- ControlNet
- LORAs
- Ti Embeddings
- Hypernetworks
- Background Separation
- Human matting / masking
- Compositing
- Drag and Drop (from discord too)
- Subgraphs
- Graph saving as metadata in the image file
- Graph loading from image metadata
This project came to life thanks to many great backend functions borrowed from ComfyUI, and adapted to work in this unique, live, controllable manner with a strict user declared execution chain, leading to data values possible to be iterated at different points in time in your pipeline.
This project is licensed under the L-GPL License. See the LICENSE file for details.