Parsr
Transforms PDFs, documents, and images into structured, enriched JSON, Markdown, or CSV data.
Turn unstructured PDF, document, and image formats into clean, structured data sets.
Parsr is a minimal-footprint toolchain that ingests PDFs, documents, images, and emails to extract structured and label-enriched information. It suits developers and data analysts who need to convert messy document layouts into formats like JSON, Markdown, or Pandas DataFrames.
What it is
The tool performs document cleaning, hierarchy regeneration for words and paragraphs, and detects headings, tables, lists, table of contents, and page numbers. It exposes an API that you can interact with via a Python client, Jupyter notebooks, or a dedicated GUI.
Who it's for
It suits data scientists and developers building automated document processing pipelines. It does not suit compliance professionals seeking an out-of-the-box GRC platform, as it requires integration work and technical setup.
Key features
Multi-Format Ingestion
Processes PDFs, images, docx files, and emails into structured outputs.
Hierarchy Regeneration
Detects words, lines, paragraphs, headings, tables, and lists automatically.
Multiple Output Formats
Generates readily available data in JSON, Markdown, CSV, or TXT.
Python & GUI Support
Includes a Python client, Jupyter notebook demo, and a Docker-based GUI viewer.
Up and running in 3 steps
Our honest verdict
What's great
- Converts complex layouts into machine-readable JSON, Markdown, or CSV
- Includes a Python client and Jupyter notebook demo for quick testing
- Provides a Docker-based GUI for visualizing extraction results
- Identifies document structures like tables of contents, headers, and footers
Where it falls short
- The project is no longer maintained and receives no security patches
- Requires setting up and running Docker containers to use the API and GUI
- Depends on heavy external tools like Tesseract, ImageMagick, and QPDF
Pricing & license
Parsr is free to use under its license. The only cost is running and hosting the required infrastructure yourself.
Open remote AI-training roles
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
Use it, or have us set it up
FAQ
Is Parsr still actively maintained?
No, the project is officially no longer maintained, and security patches are not being applied.
What output formats does Parsr support?
It outputs structured data in JSON, Markdown, CSV, Pandas DataFrames, or TXT formats.
How do I run the user interface?
You can pull and run the axarev/parsr-ui-localhost Docker image alongside the API container.
