Docling Serve icon

Docling Serve

Convert documents using Docling Serve API

Actions17

Chunk → Chunk From File (Async)

AI-generated

Summary

Process a binary file asynchronously to create text chunks using Docling Serve, employing either a hybrid or hierarchical chunking strategy.

Inputs

  • Chunker Type — Select the chunking strategy to use: hybrid or hierarchical.
  • Input Binary Field (required) — Specify the name of the binary property containing the file to chunk (default is 'data').
  • Additional Options — Configure advanced settings including OCR behavior, table extraction, token limits, image handling, processing pipeline, and related preferences.

Output shape

a JSON object containing the chunked document data after asynchronous processing, paired with the input item index.

The node uploads the binary file as multipart/form-data, submits the chunking request asynchronously, automatically polls for task completion, and returns the final processed chunked document JSON.

Examples

Example 1: Chunk a scanned PDF file asynchronously using the hybrid chunking strategy, enabling OCR and table extraction, specifying the binary input 'data', and retrieving the processed chunks in JSON format after task completion.

Set 'Chunker Type' to 'Hybrid', specify 'Input Binary Field' as 'data', set desired 'Additional Options' such as enabling OCR and table extraction, then execute to receive chunked content asynchronously.

Links

Discussion