See More

A very common reason is a wrong site baseUrl configuration.\n

Current configured baseUrl = / (default value)\n

We suggest trying baseUrl = \n\n',document.body.prepend(n);var e=document.getElementById("__docusaurus-base-url-issue-banner-suggestion-container"),s=window.location.pathname,o="/"===s.substr(-1)?s:s+"/";e.innerHTML=o}document.addEventListener("DOMContentLoaded",function(){void 0===window.docusaurus&&insertBanner()})

Skip to main content

PDFStract

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

Get StartedView on GitHub

The Complete Data Preparation Pipeline

Extract. Chunk. Embed. — From PDF to vector-ready in one command.

Extract

Convert PDFs with 10+ libraries including Marker, Docling, PyMuPDF4LLM, and more. Each optimized for different document types.

Chunk

Split text with 10+ methods from simple token-based to advanced semantic chunking powered by AI.

Embed

Generate vector embeddings with multiple providers including OpenAI, Sentence Transformers, and local models.

One unified API. Switch between libraries, chunkers, and embedding providers with a single parameter change.

pdfstract convert-chunk-embed document.pdf --library auto --chunker auto --embedding auto