Skip to main content

Parallel

A resizable container into which you drag several AI Agent nodes: they then run simultaneously on the document instead of one after another, and their respective annotations are merged on output — without deduplication or arbitration (for that, use a Supervisor).

How it works​

Once an agent is dragged inside the container, its own individual input/output ports are hidden: the container carries the group's single input and output. Dragging an agent back out of the container detaches it and restores its normal ports.

Ports​

  • Input (left, content type).
  • Output (right, annotation type).

Configuration​

None, aside from resizing the container itself.