Chaining nodes

You can create chaining logic by dragging the connectors between nodes. Chaining makes it easy to use a Large Language Model to generate an out, and then use that generated output in another node, say an image generator node.

Connecting two nodes together only means the order they will be run. You can use content from preceding nodes without having to connect those nodes separately — which makes the chain easier to read.

Node basics

You can drag nodes from the left hand side toolbar onto your canvas. You can delete a node by selecting a node and pressing Backspace or Del. Each node has connectors: input ▸ and output ● (except for start node, which only has the output connector, as you can only need to start a chain once).