Skip to main content
Ask a yes/no question about the text, images, audio or video you connect and get a boolean to route on. Connect result to an If/Else Switch node to send the run down one of two paths, and read answer and probability when you want to see why the model answered that way.

Inputs

Connect at least one of text, images, audio or video; the node raises an error when there is nothing to look at. With several images, name them image_1, image_2 and so on inside the question: a batch is expanded frame by frame, so the numbers count individual frames across the slots in connection order.

Outputs

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 0339031baf5648609b82526bed97ce8744986878bcbb706dcd3aa79a7370d83f