The Extract Information node (AI Extract Information) pulls specific, structured fields out of unstructured text. Each field can be a single value or a list.
When to use it
Use it to turn free text into clean, named data: extract a name, email, and price from a message, or a list of topics from an article. It is the reliable bridge for splitting one text blob into several named fields for a downstream multi-field consumer.
Configuration
- Content: the source text to extract from. Insert upstream data with variable tags.
- Extraction fields: add one field per value you want. Each has a name, a description of what to extract, and an Is list toggle for multiple values.
Using the output
Each extraction field becomes its own named output:
[[email] from extractInformation-1] [[topics] from extractInformation-1]
See Data Flow and Variable Handling.
Extract Information vs structured output
The AI Model Node can also return named fields directly when you turn on its structured output option, which avoids a second node. Use Extract Information when you want a dedicated step to pull fields out of text that another node already produced.
Local, Cloud, and credits
Runs in both Local vs Cloud execution modes. Costs 2 credits per run in Cloud mode; Local mode uses your own key.