AI nodes - Default

AI nodes let a workflow call a language model directly. You can generate or transform text without leaving the workflow. AI nodes work with data the workflow already has. To research the live web for net-new information, use the Deep Research node instead.

Add a node in the Workflows builder with Add Node.Many fields accept data from the trigger or an earlier node. Type {{ to open the variable picker.

AI Prompt

Sends a prompt to a language model and returns its response.

Field	What it does	How to use it
Model	Which language model to use.	Pick a model from the list, for example an OpenAI, Anthropic, Google, or xAI model.
Dynamic prompt	The prompt to send.	Write the prompt. Type `{{` to insert workflow data, for example the lead’s company name or an enrichment result.

For example, prompt the model to summarize a lead’s enriched company description into one sentence. Then use that sentence in a Slack notification or CRM field. If the request fails, for example with an invalid model or an empty prompt, the node does not stop the workflow. Instead, it returns a failed result. A later Multi-Branch node can react to the failure. The whole run does not error out. The response text is available to later nodes, along with whether the request succeeded. If it failed, an error message is also available to log or act on.

​AI Prompt

AI Prompt