PDFStract

The Data Preparation Layer for RAG — Extract. Chunk. Embed.

One unified API. Switch between 10+ extraction libraries, 10+ chunking methods, and multiple embedding providers with a single parameter change. Focus on your RAG outcomes, not library dependencies.

Installation

pip install pdfstract              # Base - pymupdf4llm, markitdown
pip install pdfstract[standard]    # + OCR (pytesseract, unstructured)
pip install pdfstract[advanced]    # + ML-powered (marker, docling, paddleocr)
pip install pdfstract[all]         # Everything

Python API

from pdfstract import PDFStract

pdfstract = PDFStract()

# Extract
text = pdfstract.convert('document.pdf', library='auto')

# Chunk
chunks = pdfstract.chunk(text, chunker='semantic', chunk_size=512)

# Embed
vectors = pdfstract.embed_texts([c['text'] for c in chunks['chunks']])

# Combined pipelines
result = pdfstract.convert_chunk('document.pdf', library='marker', chunker='token')
result = pdfstract.convert_chunk_embed('document.pdf', embedding='sentence-transformers')

Extract Examples

# Auto-select best available library
text = pdfstract.convert('document.pdf', library='auto')

# Use specific library
text = pdfstract.convert('document.pdf', library='marker')
text = pdfstract.convert('document.pdf', library='docling', output_format='json')

# Batch processing
results = pdfstract.batch_convert('./pdfs', library='pymupdf4llm', parallel_workers=4)

# Async
text = await pdfstract.convert_async('document.pdf', library='marker')

Chunk Examples

# Token-based chunking
chunks = pdfstract.chunk(text, chunker='token', chunk_size=512, chunk_overlap=50)

# Semantic chunking
chunks = pdfstract.chunk(text, chunker='semantic', chunk_size=1024)

# Code-aware chunking
chunks = pdfstract.chunk(code_text, chunker='code')

# Access results
for chunk in chunks['chunks']:
    print(f"Chunk {chunk['chunk_id']}: {chunk['token_count']} tokens")

Embed Examples

# Embed multiple texts
vectors = pdfstract.embed_texts(["First text", "Second text"], model='sentence-transformers')

# Embed single text
vector = pdfstract.embed_text("Hello world", model='openai')

# List available providers
providers = pdfstract.list_available_embeddings()

CLI

pdfstract convert document.pdf --library marker
pdfstract convert-chunk document.pdf --chunker semantic
pdfstract convert-chunk-embed document.pdf --embedding sentence-transformers
pdfstract batch ./pdfs --parallel 4

What's Included

Chunkers: token, sentence, semantic, recursive, code, and more

Embeddings: OpenAI, Azure, Google, Ollama, Sentence Transformers

Documentation

📖 pdfstract.com — Full docs, guides, and API reference

GitHub: github.com/aksarav/pdfstract · Issues · MIT License