Subnet AKA Subnetwork.

The Subnet node is a powerful organizational tool that allows you to collapse a complex network of nodes into a single, manageable block. Think of it as a folder or a function for your node graph. It helps keep your main graph clean and makes it easier to reuse parts of your logic.

When you create a Subnet node, it comes pre-populated with a Subnet Input and a Subnet Output node inside.

How It Works

You can double-click a Subnet node to “dive into” it and edit its internal graph. When you’re inside, a breadcrumb trail will appear at the top-left of the graph editor, allowing you to navigate back out to the parent graph.

1. Input (Optional): If nodes are connected to the Subnet in the parent graph, their data is passed to the Subnet Input node inside.
2. Internal Logic: Inside the subnet, you can build any network of nodes. If the subnet receives data from the parent graph, you can start your logic from the Subnet Input node. Alternatively, you can generate all data from scratch within the subnet itself.
3. Output: The final result of your internal logic should be connected to the Subnet Output node. This result then becomes the output of the main Subnet node in the parent graph.

Creating a Subnet

You can create a Subnet in two ways:

1. From the search menu: Press Tab or right-click in the graph and search for “Subnet”. This creates a new, empty subnet.
2. From a selection: Select two or more nodes and use the shortcut Ctrl+C. This will automatically wrap your selected nodes and their internal connections into a new Subnet, rewiring the external connections for you.
3. Using NodeGraph menu bar: Actions > Create Subnet.

See Also

• Subnet Input
• Subnet Output