The Stations (Nodes)
Nodes are the building blocks of your canvas. Each node performs a specific function — from providing inputs to generating media. You build a workflow by placing nodes and connecting them together.
Text Node
The Text Node is a simple text input card. It holds a string of text — a prompt, a script line, a description — that gets passed as an input to connected generative nodes.
It does not process or generate anything on its own. Its role is to supply text that other nodes (like Image or Video nodes) consume as their prompt.
Supports:
Text-to-Image (connected to an Image Node)
Text-to-Video (connected to a Video Node)
Image Node
The Image Node generates or displays a static image. It can operate in two modes depending on what is connected to it:
Text-to-Image: Receives a text prompt and generates an image using the selected model.
Image-to-Image: Receives an existing image and produces a variation, upscale, or transformation of it.
Image-to-Video: Passes the generated image downstream to a Video Node.
Video Node
The Video Node generates video output. It accepts an image or prompt as input and produces a video clip using the selected model.
Text-to-Video: Generates a video directly from a text prompt.
Image-to-Video: Uses an upstream image as the first frame and animates it.
Utility Nodes
Utility nodes help you process and refine media within your workflow. They do not consume generative credits.
Split Node: Separates a composite image (e.g. a 2×2 grid) into individual frames.
Extract Frame Node: Pulls a single frame from a video clip for use as a static image.
Assembly Node: Combines multiple clips into a single output video.
Connecting Nodes
To connect two nodes, click the output handle on the right side of one node and drag it to the input handle on the left side of another. The data type of the output must match the input — for example, an image output can connect to an image input.
Last updated