Skip to main content
net10.0net9.0net8.0.NETStandard2.0

The Syncfusion® DocumentChunking library is a .NET Standard library that helps split large documents into smaller, meaningful chunks. These chunks make it easier to process, search, index, and use document content in AI-based applications.

Published
Oct 13, 2026
11d ago
Total Downloads
373
373

Install

dotnet add package Syncfusion.DocumentChunking.Net.Core --version 35.1.40
ProductVersions
.NET
net10.0net8.0net9.0
.NET Standard
netstandard2.0

README

Syncfusion .NET Document Chunking library

The Syncfusion .NET Document Chunking library is a structure-aware, high-performance .NET library used to extract and split content from Word, PDF, PowerPoint, Excel, and Markdown documents into meaningful, token-bounded chunks. It is designed for retrieval-augmented generation (RAG), semantic search, vector database indexing, document intelligence, and other AI-powered workflows.

The library preserves document structure, metadata, citations, source locations, and parent-child relationships while applying format-specific semantic boundaries.

Using this package with an AI coding agent

Before generating Syncfusion code, ask your AI coding agent to run the Syncfusion setup using the following prompt:

Set up this project for Syncfusion before generating code. Fetch and follow the official instructions at https://ai.syncfusion.com, then report the detected platform, installed skill pack, MCP status, and any required licensing action.

The setup detects your project's platform and loads the appropriate Syncfusion skill pack. No Syncfusion account or MCP key is required to install or read Syncfusion agent skills.

Platform reference: https://ai.syncfusion.com/document-sdk/llms.txt

Key Features
  • Supports Word documents.
  • Supports PDF documents.
  • Supports PowerPoint presentations.
  • Supports Excel workbooks.
  • Supports Markdown documents.
  • Supports both file-path and stream input.
  • Supports configurable maximum token limits.
  • Supports configurable overlap between recursively split chunks.
  • Preserves meaningful structures such as titles, headings, sections, paragraphs, lists, tables, worksheets, ranges, cells, slides, shapes, and notes.
  • Provides format-specific chunking modes for selecting preferred semantic boundaries.
  • Preserves all meaningful content when a chunking mode uses a preferred boundary.
  • Recursively splits oversized structures into smaller semantic units while respecting the configured token limit.
  • Supports parent-child relationships between related chunks.
  • Generates deterministic document and chunk identifiers suitable for vector database upsert and re-indexing workflows.
  • Provides document-level metadata such as file name, file size, last-modified date, author, and format-specific document properties when available.
  • Provides citation details including source file, source path, content type, page number, slide number, worksheet name, cell address, heading path, section identifier, and structural coordinates when available.
  • Supports custom token counters through the ITokenCounter interface.
System Requirements

For complete platform requirements, refer to the Syncfusion Document Processing system requirements.

Getting Started

Install the Syncfusion .NET Document Chunking library from the NuGet Package Manager Console:

Install-Package Syncfusion.DocumentChunking

You can also install it with the .NET CLI:

dotnet add package Syncfusion.DocumentChunking
Chunk a Document from a File Path

Try the following code example to get the generated chunks from Markdown document.

using Syncfusion.DocumentChunking;

ChunkingService chunkingService =
    new ChunkingService();

IChunkingResult result =
    chunkingService.Chunk(
        "../../../Data/TextMarkdown.md",
        new ChunkingOptions
        {
            MaxTokens = 50,
            OverlapTokens = 0,
            IncludeMetadata = true,
            IncludeCitation = true,
            EnableParentChild = true,
            SourceOptions =
                new MarkdownChunkingOptions
                {
                    ChunkingMode =
                        MarkdownChunkingMode.Auto
                }
        });
License

This is a commercial product and requires a paid license for possession or use. Syncfusion's licensed software, including this component, is subject to the terms and conditions of Syncfusion's EULA. You can purchase a license here or start a free 30-day trial here.

About Syncfusion

Founded in 2001 and headquartered in Research Triangle Park, N.C., Syncfusion provides enterprise software components and frameworks for web, mobile, and desktop development, along with document-processing libraries, dashboards, reports, data integration, and related development solutions.

[email protected] www.syncfusion.com Toll Free: 1-888-9 DOTNET

Changes from Previous Version

No structural changes between v35.1.39 and v35.1.40.

Adoption Assessment

A verified publisher signal is available. Assess fit from the package's supported frameworks, license and dependency graph rather than popularity alone.

No SPDX license expression is published. Review the license text and your organisation's policy before redistribution.Compatibility metadata lists 4 target frameworks; confirm the exact target used by your application.The published metadata declares 21 direct dependency entries across framework groups. Inspect the resolved dependency graph for your target framework.

Related Guides & Workflows

Learn how to evaluate, integrate, or audit this package in your .NET project.