HomeProjectsArticlesPK BoardAbout
Explore Projects

OpenDataLoader PDF

Active
GitHub Java Apache-2.0

Description

opendataloader-pdf is an open-source PDF parser and extractor supporting layout analysis, table extraction, OCR and structured output, fit for RAG and document-agent pipelines.

Key Features

  • PDF parsing — text, images, tables and form fields
  • Layout analysis — auto-detects headers, paragraphs, figures, tables, lists
  • Table extraction — reconstructs tables as structured data
  • OCR integration — invokes OCR on scanned PDFs automatically
  • Multi-format output — JSON, Markdown, HTML, JSONL
  • RAG-friendly — emits chunks ready for vector stores

Use Cases

💡 Batch-parsing contracts, reports and papers for RAG apps
💡 Turning scanned PDFs into searchable documents
💡 Serving as the unified parser for document-agent pipelines
💡 Running structured analysis on PDFs (financial, invoices, research)

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (28.5k stars)
  • • Permissive open-source license (Apache-2.0)
  • • Responsive to issues, low backlog

Tags

pdf ocr extraction document-ai open-source

Categories

📚 RAG Tools ⚡ Agent Tools

Quick Start

pip install opendataloader-pdf

opendataloader-pdf extract ./contract.pdf --format markdown
Visit GitHub Visit Website

Related Projects

Crawl4AI

23.0k · Python
Stale B

Crawl4AI is a web crawling toolkit for LLM and agent systems, offering structured extraction, site traversal, cleanup, and crawl controls for external knowledge acquisition.

crawlingwebextraction +1
  • · LLM-friendly content extraction — clean HTML into Markdown or structured data for LLM consumption
  • · Batch async crawling — concurrent multi-URL fetching with rate limiting and anti-bot handling
  • · Deep site traversal — recursive subpage discovery with automatic sitemap construction
View full details → View on GitHub

PageIndex

35.2k · Python
Active A+

Open-source vectorless, reasoning-based RAG that builds a hierarchical tree index from long documents and uses LLMs to reason over it, achieving 98.7% on FinanceBench without vector DBs or chunking.

ragreasoningretrieval +3
  • · Vectorless retrieval — no vector DB or chunking needed; uses document structure and LLM reasoning
  • · Hierarchical tree index — automatically generates TOC-style tree structure from long documents
  • · Reasoning-driven — every retrieval result is traceable to explicit page and section references
View full details → View on GitHub

PaddleOCR

87.7k · Python
Active A+

PaddleOCR is Baidu's open-source multilingual OCR and document-AI toolkit supporting 80+ languages and PP-Structure layout analysis.

ocrpaddlepaddledocument-ai +2
  • · Multilingual recognition — 80+ languages out of the box including Chinese (Simplified/Traditional), English, Arabic
  • · PP-Structure layout analysis — text blocks, tables, figures, formulas auto-classified and located
  • · PP-OCRv4 high accuracy — state-of-the-art detection + recognition end-to-end accuracy
View full details → View on GitHub