Skip to content

User Guide

This guide provides comprehensive documentation for the Kreuzberg document intelligence framework, covering core concepts, configuration options, and integration patterns.

Contents

Best Practices

  • Use the async API for better performance in web applications and concurrent extraction
  • Configure OCR language settings to match your document languages for better accuracy
  • For large documents, consider file streaming methods to reduce memory usage
  • When processing many similar documents, reuse configuration objects for consistency

Common Use Cases

Document Analysis:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from kreuzberg import extract_file, ExtractionConfig

async def analyze_document(file_path):
    result = await extract_file(file_path, config=ExtractionConfig())

    # Get basic document content
    text = result.content

    # Access metadata
    title = result.metadata.get("title", "Untitled")
    author = result.metadata.get("authors", ["Unknown"])[0]

    return {"title": title, "author": author, "content": text, "word_count": len(text.split()), "char_count": len(text)}