Syncfusion Document Chunking Library
22 Sep 20261 minute to read
The Syncfusion® Document Chunking Library provides a unified solution for dividing Excel, Word, PDF, PowerPoint, and Markdown documents into meaningful and structured chunks.
The library processes documents based on their native structure and supports format-specific chunking modes. It also allows you to configure the chunk size and overlap and optionally include metadata and citation information in the generated chunks.
The generated chunks can be used for embedding generation, semantic search, vector indexing, hybrid search, document Q&A, grounded AI responses, citation-based retrieval, and enterprise Retrieval-Augmented Generation (RAG) workflows.
Key Capabilities
- Supports Excel, Word, PDF, PowerPoint, and Markdown documents.
- Provides chunking modes based on the structure of each supported document format.
- Supports configuring the maximum token count and token overlap.
- Supports including source metadata in generated chunks.
- Supports including citations that identify the original location of the chunk content.
- Provides structured chunks for search, retrieval, and AI-based document-processing workflows.