Parsr
StaleDescription
Transforms PDF, documents and images into enriched structured data with table recognition, reading order restoration, and Markdown output.
Key Features
- Multi-format document parsing: supports PDF, DOCX, images, and EML files with minimal footprint
- Structured output in JSON, Markdown, CSV/Pandas DataFrame, and TXT formats
- Document structure detection: headings, tables, lists, table of contents, page numbers, headers/footers, and links
- Table recognition and extraction with reading order restoration for accurate content preservation
- Docker-based deployment with GUI for document submission and result visualization
- Python client and Jupyter Notebook integration for programmatic document processing
Use Cases
💡 Automated data entry and document analysis automation for insurance and financial documents
💡 Converting scanned PDFs and images into structured, machine-readable data for downstream processing
💡 Archival and digitization of legacy documents with proper reading order and structure preservation
💡 RAG pipeline preprocessing: extracting clean text and structured data from documents for AI ingestion
Categories
Quick Start
docker pull axarev/parsr
docker run -p 3001:3001 axarev/parsr
# API available at http://localhost:3001
# Install Python client:
pip install parsr-client
# Or use the GUI:
docker run -t -p 8080:80 axarev/parsr-ui-localhost:latest