AI Generate

This block uses an LLM to generate content based on a prompt and edge data.

Models Connection

You should first create a Connection that points to an LLM. Currently, we suggest you use openAI for ChatGPT.

Then select this connection from the Models Connection dropdown.

Select from the available models. This list is specific to the Models Connection.

Before configuring the prompt, you should ensure the incoming edge has data by running the flow or stepping to it.

This ensures that you can test the prompt properly against actual data.

The main part is configuring the prompt. Press the Prompt & Settings button to open the dialog.

Below is an example prompt.

You can test your prompt with the Test button.

When the flow runs, you can see the result by clicking on the outgoing edge.