Our take
Activeregular commits and releases- Activity
- 55
- Community
- 47
- Issues
- 43
- Pull requests
- 67
+1 stars in the last 30 days
- Slower to close issues than 99% of the projects we track
sumy is an actively maintained, Apache-2.0 Python toolkit for practical extractive summarization, though its contributor activity is concentrated and issue resolution has been slow historically.
Good fit if
- You need an embeddable Python library or CLI for summarizing plain text and HTML with established algorithms.
- You want to evaluate extractive summaries against reference text, or run the tool in Docker.
- You can work with Python 3.8 or later and the project's Apache-2.0 license.
Look elsewhere if
- You need generative summaries rather than summaries extracted from source sentences.
- Your adoption depends on rapid issue resolution, given the long historical median time to close issues.
- You need a project whose maintenance is not heavily dependent on one contributor.
All health signals
| Last commit | 2026-10-25 (3 days ago) |
|---|---|
| Commits, last 90 days | 51 |
| Releases, last 12 months | 2 (latest v0.13.0, 2026-08-12) |
| Contributors | 31 (top contributor: 81% of commits) |
| Issues closed, last 90 days | 20 (typically closed in 1116 days) |
| Pull requests merged, last 90 days | 16 (typically merged in 0 days) |
| Project age | 13 years |
Checked on 2026-10-29 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
3,701Forks
532Change since Nov 14, 2026: 0 stars·0 forks
Overview
sumy is a Python library and command-line tool for automatic text summarization. It extracts summaries from plain-text documents and HTML pages, helping developers and researchers condense documents without building the parsing and summarization workflow themselves.
The project provides several summarization algorithms and a simple framework for evaluating summaries. It supports Python 3.8 and later, and its documentation describes how to add language support when a language is not already covered.
Key Features
- Summarize HTML pages from a URL or plain text from a file or string.
- Use the command-line utility or integrate summarization into Python applications.
- Choose among algorithms including LexRank, LSA, Luhn, and Edmundson.
- Set summary length by sentence count or percentage in CLI usage.
- Configure language tokenization, stemming, and stop words.
- Evaluate summarization methods against reference summaries with
sumy_eval. - Run the CLI in a Docker container.
Use Cases
- Developers can add extractive summaries of articles or documents to a Python application.
- Researchers can compare summarization algorithms against reference summaries using the evaluation utility.
- Readers can quickly condense a web page from the command line when a short, sentence-based summary is sufficient.
- Teams working with supported languages can adapt the tokenizer and language resources for their text-processing workflows.
What you need
Detected in the repository
- Python >=3.8 (from pyproject.toml)
- A Dockerfile, so it can run in a container
- A test suite and automated checks on GitHub Actions
License in plain words
Apache-2.0permissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the license, the NOTICE file and a note of your changes
- Share your changes: no
- Includes an explicit patent grant from the contributors:
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Install with pip:
pip install sumy
For command-line examples, Python usage, and development instructions, see the README.
Alternatives
- newspaper: Newspaper3k focuses on crawling news articles and extracting article data, with summarization as one feature; sumy focuses on extractive summarization of text and HTML.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| sumy | Python | Apache-2.0 | 3.7k | Active |
| newspaper | Python | MIT | 15.2k | Maintained |
Considerations
This is an extractive summarizer, so it selects sentences from source material rather than generating new text. Language support depends on the available tokenizer and language resources; the project documents how to add a language if needed. The repository is active, with recent commits and releases, but most measured contributor activity comes from one person, and issue closure has been slow historically. The Apache-2.0 license should be reviewed against your organization's distribution and attribution requirements.
Found this useful?
Share it with someone who would like sumy.