Skip to content

Add LandingAI Agentic Document Extraction (ADE) node #1253

Description

@kgarg2468

Feature

Add landingai_ade, a data-flow processor node wrapping LandingAI's Agentic Document Extraction (ADE) API. Documents flowing through a pipeline (PDFs, images, office files) are parsed into Markdown text and extracted tables.

Surface

  • Consumes the tags lane (binary document stream) → produces text (Markdown) + table lanes.
  • Official landingai-ade Python SDK; API-key auth via VISION_AGENT_API_KEY; US/EU region select; model dpt-2-latest.

Archetype

Data-flow processor (classType: ["data"]), mirroring the existing reducto / llamaparse document-parser nodes. Net-new, additive, zero core changes.

Why it fits

ADE is a strong document-parsing/OCR option alongside reducto and llamaparse, with high-quality table/layout extraction.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions