sumy
Automatic text and HTML page summarization library and CLI tool.
Extract concise summaries from documents and web pages using multiple algorithms.
Sumy is a Python library and command line utility designed to extract automated summaries from HTML pages or plain text documents. It suits compliance and audit professionals needing to quickly parse lengthy regulatory documents or web sources.
What it is
It provides implementations of classic summarization methods like LSA, Luhn, Edmundson, LexRank, and TextRank-style algorithms. It also includes a basic evaluation framework to test generated summaries against reference texts.
Who it's for
Suits data-literate compliance officers and auditors comfortable with Python or CLI tools. It does not suit non-technical users requiring turnkey graphical interfaces.
Key features
Multiple Extraction Algorithms
Supports LSA, Luhn, Edmundson, and LexRank methods out of the box.
HTML and Plain Text
Parses content directly from URLs or local text files.
Summary Evaluation Framework
Compares generated summaries against provided reference summary files.
Multi-language Tokenization
Supports tokenization and stemmers across various natural languages.
Up and running in 3 steps
Our honest verdict
What's great
- Includes a built-in evaluation framework for testing summary quality against references
- Provides a command-line interface for rapid document triage without writing code
- Supports direct URL parsing alongside plain text file inputs
- Offers multiple distinct algorithmic approaches to summarization
Where it falls short
- Requires Python 3.8+ and comfort with the command line or code integration
- Running batch jobs at scale requires setting up your own scripts or containers
- Lacks a native graphical user interface for non-technical auditors
Pricing & license
It is free under its open source license. The only cost is running it yourself on your own hardware or infrastructure.
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
Can sumy parse pages directly from URLs?
Yes, the HTML parser can fetch and parse content directly from web URLs.
Does it support languages other than English?
Yes, multiple languages are supported with easy extensibility for new ones.
How do I evaluate summary quality?
The package includes a sumy_eval command to compare outputs against reference text.
