Extract PDF Node (LLM Extraction)

Extract PDF Node (LLM Extraction)

Overview

A new Extract PDF Node has been introduced under the Rubi AI category to enable structured data extraction from PDF documents using Large Language Models (LLMs). The feature supports configurable prompt-based extraction pipelines for generating structured reusable outputs from PDF files.

Step 1 – Extract PDF Node under Rubi AI

The Extract PDF Node is available under the Rubi AI category and can be added directly to the pipeline canvas.


Step 2 – Choose Business Context

Users can select the required business context for document extraction such as Finance, Procurement, HR, Operations, IT, or Custom extraction scenarios.


Step 3 – Choose Extraction Prompt

Users can select predefined extraction prompts for commonly used document extraction use cases.


Step 4 – Add Custom Prompt

Users can also create custom extraction prompts to define specific extraction requirements.


Step 5 – Configure Custom Prompt

Custom prompts support user-defined prompt names and extraction instructions for tailored document processing.


Step 6 – Model & Execution Settings

Users can configure the LLM model and execution settings before processing the extraction request.


Feature Enhancements

  • Added Extract PDF Node under the Rubi AI category.

  • Supports input from PDF Reader and Split PDF nodes.

  • Allows multiple PDF nodes to be connected as input.

  • Supports predefined and custom prompt-based extraction pipelines.

  • Structured extracted results are available in the Explore section.

  • Users can rerun extraction with modified prompts without recreating the node.

  • Generated output can be passed to downstream pipeline nodes.

Benefits

  • Automated extraction of structured data from PDFs.

  • Reduced manual document processing effort.

  • Flexible AI-powered extraction pipelines.

  • Reusable structured outputs for analytics and automation.

  • Improved document intelligence capabilities.