# PDF RAG Chunker & Citation Pack Convert a PDF into heading-aware RAG chunks with page numbers, bounding boxes, and citation metadata > Canonical page: https://elysiatools.com/en/tools/pdf-rag-chunker-citation-pack - **Category:** AI Tools - **Keywords:** pdf, rag, citation, chunking, vector database ## Overview Upload a PDF to generate retrieval-friendly chunks with page references, heading paths, and bounding boxes. The output is a JSON pack that works well for vector stores, answer citation, and PDF-grounded chat systems. ## Inputs - **PDF File** (file) - **Chunk Mode** (select) - **Maximum Characters Per Chunk** (number) - **Use Struct Tree** (checkbox) - **Sanitize Sensitive Data** (checkbox) - **Include Table Nodes** (checkbox) ## When to use - Preparing PDF documents for ingestion into vector databases for semantic search. - Building PDF-grounded AI chat systems that require precise source citations and bounding box highlights. - Extracting structured, heading-aware text chunks from complex reports while preserving document hierarchy. ## How it works - Upload your target PDF file into the tool. - Select your preferred chunking mode, such as heading-aware or element-per-chunk, and set the maximum character limit. - Toggle advanced options like structural tree usage, sensitive data sanitization, or table inclusion based on your needs. - Download the generated JSON pack containing the text chunks, page references, and bounding box metadata ready for your RAG pipeline. ## Use cases - Ingesting financial reports and earnings calls into a vector store for an AI financial analyst assistant. - Processing legal contracts to build a semantic search tool that links directly back to specific clauses in the original document. - Chunking technical manuals and product documentation to power an accurate, citation-backed customer support chatbot. ## Frequently asked questions ### What format does this tool output? The tool outputs a structured JSON file containing the text chunks along with their corresponding metadata, such as page numbers, heading paths, and bounding boxes. ### What is the difference between heading-aware and element-per-chunk modes? Heading-aware mode groups content under its respective section titles up to the maximum character limit, while element-per-chunk treats every individual paragraph, list, or table as a separate, isolated chunk. ### Can I control the size of the generated chunks? Yes, you can set a maximum character limit per chunk, ranging from 200 to 4000 characters, to optimize retrieval performance for your specific vector store. ### Does the tool extract tables from the PDF? Yes, as long as the 'Include Table Nodes' option is enabled, the tool will extract tables and include them in the generated RAG chunks. ### What are bounding boxes used for in the output? Bounding boxes provide the exact spatial coordinates of the text on the original PDF page, allowing frontend applications to visually highlight the cited source text for users. ## Related tools - [PDF to Clean Text for LLM](https://elysiatools.com/en/tools/pdf-to-clean-text-for-llm): Extract clean text from PDFs with OpenDataLoader for summarization, translation, embedding, and other LLM workflows - [Markdown to PDF Theme Pack](https://elysiatools.com/en/tools/markdown-to-pdf-theme-pack): Convert Markdown to PDF with dark, light, or print-ready themes - [PDF/A Convert](https://elysiatools.com/en/tools/pdf-a-convert): Convert PDF documents to a self-declared PDF/A archival profile with zero external dependencies - [PDF to Excel](https://elysiatools.com/en/tools/pdf-to-excel): Extract tabular data from PDF files and convert to Excel spreadsheets with customizable parsing options - [PDF to Markdown Converter](https://elysiatools.com/en/tools/pdf-to-markdown): Convert PDF documents to Markdown format with text extraction and formatting preservation - [PDF to Text Advanced](https://elysiatools.com/en/tools/pdf-to-text-advanced): Advanced PDF to text converter with page selection, formatting options, and metadata extraction - [OCR PDF to Structured JSON Bridge](https://elysiatools.com/en/tools/ocr-pdf-to-structured-json-bridge): Extract the PDF text layer with geometry (lines by y-position, tables by column gaps, headings by font size, colon key-value pairs), then fill a user-supplied JSON Schema field by field — labels matched by normalized keys, values coerced to declared types and validated with ajv. - [PDF Header/Footer Noise Remover](https://elysiatools.com/en/tools/pdf-header-footer-noise-remover): Compare extraction with and without repeated page furniture to spot header/footer noise before using PDF text in RAG, summarization, or editing workflows ## Samples - [PDF Samples](https://elysiatools.com/en/samples/pdf-samples): Generated PDF samples from tools dated 2026-02-01 to 2026-02-10 - [Markdown Slide Deck Samples](https://elysiatools.com/en/samples/md-slide-deck-to-pdf): Remark/Marp style Markdown slide decks for testing PDF export layouts - [Time Zone Workflow Scheduler ICS Samples](https://elysiatools.com/en/samples/time-zone-workflow-scheduler-ics-samples): ICS files generated in the same structure returned by the Time Zone Workflow Scheduler, with multiple VEVENT meeting candidates exported from overlap windows - [ASS Subtitle Samples](https://elysiatools.com/en/samples/ass-samples): Sample ASS subtitle files from simple to complex for style-aware subtitle parsing, translation, conversion, and localization QA ## Related content - [Document OCR Extraction](https://elysiatools.com/en/hubs/document-ocr-extraction): Extract OCR text from scans and convert document pages or images into Markdown, JSON, tables, captions, and retrieval-ready chunks with quality checks. - [PDF LLM and RAG Preparation](https://elysiatools.com/en/hubs/pdf-llm-rag-prep): Turn PDFs into clean, safe, citation-ready inputs for LLM summarization, embedding, search, and RAG pipelines. - [RAG Chunking and Retrieval Prep](https://elysiatools.com/en/hubs/rag-chunking-retrieval-prep): Clean PDFs and Word files, remove extraction noise, scan hidden prompt risks, split text, score chunk quality, and package citation-ready RAG inputs.