Skip to main content

​
Quickstart

This page gets you from a blank terminal to a working PDF extraction in as few steps as possible. No prior knowledge of MuPDF.NET required.

​
1. Create a project

dotnet new console -n Pdf4LlmDemo
cd Pdf4LlmDemo

​
2. Install PDF4LLM

dotnet add package PDF4LLM

​
3. Add a PDF

Copy any PDF into the project folder and note its filename. If you don’t have one to hand, download a sample: 
curl -o sample.pdf https://www.w3.org/WAI/WCAG21/wcag-2.1.pdf

​
4. Convert to Markdown

Replace the contents of Program.cs with: 
using MuPDF.NET;
using PDF4LLM;

Document doc = new Document("sample.pdf");

string markdown = PdfExtractor.ToMarkdown(doc);

doc.Close();

Console.WriteLine(markdown);
Run it: 
dotnet run
You should see the PDF content printed to the console as Markdown — headings prefixed with #, tables as pipe syntax, bold and italic preserved.

​
5. Save the output

Write the result to a file instead of printing: 
using MuPDF.NET;
using PDF4LLM;
using System.IO;

Document doc      = new Document("sample.pdf");
string   markdown = PdfExtractor.ToMarkdown(doc);
doc.Close();

File.WriteAllText("output.md", markdown, System.Text.Encoding.UTF8);
Console.WriteLine("Saved to output.md");

​
6. Try the other output formats

Switch the extraction method to see different representations of the same document. Plain text — same layout analysis, no Markdown syntax: 
string text = PdfExtractor.ToText(doc);
JSON — full layout structure with bounding boxes and block types: 
string json = PdfExtractor.ToJson(doc);
File.WriteAllText("layout.json", json, System.Text.Encoding.UTF8);

​
7. Extract specific pages

Pass a zero-based list to process only the pages you need: 
string markdown = PdfExtractor.ToMarkdown(
    doc,
    pages: new List<int> { 0, 1, 2 }
);

​
You’re up and running

That’s the core loop: open a Document, call an extractor method, close the document. Everything else — OCR, image extraction, LlamaIndex loading, form fields — builds on this pattern.