OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Pulpie

Pulpie is a Python library that uses encoder-based models to extract main content from raw HTML, remove boilerplate, and return the retained blocks as clean HTML or Markdown.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Pulpie — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Pulpie when HTML has already been collected and needs main-content extraction before it is passed to another text-processing, search or RAG stage.

Who it is for

  • Python developers preparing webpage content for search, analysis or RAG workflows.
  • Teams processing batches of HTML pages that want an extraction pipeline with CPU and GPU stages.

Documented uses

  • Strip navigation, ads, sidebars and footers from HTML pages.
  • Convert retained page content into clean HTML or Markdown.
  • Batch-process pages before downstream text or retrieval workflows.
Before you choose

Requirements & limits

The repository documents installation through pip and requires Python 3.9 or later. The model downloads from Hugging Face on first use. The repository identifies the license as Apache-2.0.

  • Orange Small is the default 210M-parameter model; Base is 610M and Large is 2.1B.
  • The README reports WebMainBench ROUGE-5 F1 results of 0.862, 0.863 and 0.873 for Small, Base and Large, respectively.
  • Markdown output requires the markdown extra; HTML output is shown in the basic usage.
  • The batch Pipeline overlaps CPU preprocessing with GPU inference and supports one or more GPUs.

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 Pulpie and extract a page

Install the library and use its documented Extractor interface to obtain cleaned content from HTML.

Before you start

  • Python 3.9 or later
  • HTML input
  • An internet connection for the model download on first use

Steps

  1. Install the package with the command below.

    Documented command
    pip install pulpie

  2. Import Extractor from pulpie and create an Extractor() instance.

  3. Pass HTML to extractor.extract(html).

  4. Read the result’s html field for clean HTML; install pulpie[markdown] if Markdown output is needed.

Expected outcome

A result containing extracted page content, including clean HTML and, with the Markdown extra, Markdown output.

Guide limitations

  • The default model downloads from Hugging Face on first use.
  • The guide covers extraction from supplied HTML; it does not cover collecting webpages.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
feyninc/pulpie · Official homepage title · retained in its source language
How we review sourcesSuggest a correction