Pytextrank: TextRank for Python and spaCy

Understand the project

Practical overview

Use PyTextRank when a Python text-processing workflow needs graph-based phrase ranking or extractive summarization within a spaCy pipeline.

Python developers building NLP applications with spaCy
Researchers and practitioners who need ranked key phrases or extractive summaries from text

Before you choose

The public repository identifies an MIT license. The README documents installation through pip or Conda.

PyTextRank is a library to integrate into a Python application, rather than a hosted analysis service.
The documented setup uses spaCy and a separately installed language model.
The README says local builds are generally unnecessary except for contributors.
The project lists TextRank, PositionRank, Biased TextRank and TopicRank as supported algorithm families.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install PyTextRank and use the documented English-model example to inspect ranked phrases in a text document.

Install PyTextRank with the command below.
Documented command
```
python3 -m pip install pytextrank
```
Install the example language model with the command below.
Documented command
```
python3 -m spacy download en_core_web_sm
```
In Python, import spacy and pytextrank, then load en_core_web_sm with spacy.load("en_core_web_sm").
Add the component with nlp.add_pipe("textrank"), process text with nlp(text), and inspect doc._.phrases.

The pipeline exposes phrases with text, rank, count and chunk information.

The example uses the English en_core_web_sm model; the README says model choice depends on language and scale.
This guide follows the documented example and is not a report of an execution test.

Evidence and transparency

Source record details

Source availability: Active
Sources checked: 11 October 2026
Original website title: DerwenAI/pytextrank · Official homepage title · retained in its source language