Repolode
Submit
The-Vibe-Company

The-Vibe-Company/megaparse

A document parser that converts PDF, Word, PowerPoint, Excel and CSV files into LLM-ready text while aiming to keep content intact.

View on GitHubWebsite
Stars
7,415
Forks
417
Open issues
34
Last commit
6 days ago

What is it?

MegaParse is a Python library for parsing documents into a format suited to LLM ingestion. It handles text files, PDFs, PowerPoint presentations, Word documents, Excel spreadsheets and CSV files, and aims to preserve content during conversion.

The parser recognizes structural elements such as tables, table of contents, headers, footers and images. A separate MegaParse Vision mode uses a multimodal model (such as GPT-4o or Claude) to read pages as images, and the project also includes an API server that can be run locally.

Using the library requires Python 3.11 or newer. Depending on the file types, you also need system tools such as poppler and tesseract, plus an OpenAI or Anthropic API key for the vision mode.

Key features

  • 01Parses PDF, Word, PowerPoint, Excel, CSV and text files.
  • 02Preserves tables, table of contents, headers, footers and images.
  • 03Offers a vision mode that uses multimodal LLMs to read pages.
  • 04Can be run as a local API server with documented endpoints.

Use cases

  • Preparing PDF or Office documents as input for retrieval-augmented generation pipelines.
  • Extracting structured text and tables from presentations and reports for LLM processing.
  • Building a document ingestion service that exposes parsing through an HTTP API.

Quick start

  1. 1

    Install MegaParse with pip (Python 3.11 or newer)

    pip install megaparse
  2. 2

    Add your OpenAI or Anthropic API key to the .env file

  3. 3

    Install poppler and tesseract on your computer (needed for images and PDFs)

  4. 4

    Parse a document with MegaParse in Python

Things to know

  • Requires an OpenAI or Anthropic API key for its parsing features.
  • Needs system tools poppler and tesseract installed separately, and libmagic on macOS.
  • Requires Python 3.11 or newer.

Topics

#document-parsing#pdf#llm#ocr#python#rag#docx#parser#powerpoint

FAQ

What is megaparse?
A document parser that converts PDF, Word, PowerPoint, Excel and CSV files into LLM-ready text while aiming to keep content intact. MegaParse is a Python library for parsing documents into a format suited to LLM ingestion. It handles text files, PDFs, PowerPoint presentations, Word documents, Excel spreadsheets and CSV files, and aims to preserve content during conversion.
Is megaparse free to use? What license does it have?
megaparse is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is megaparse still maintained?
Actively maintained. The last commit to megaparse was on Oct 15, 2026.
What can I use megaparse for?
Preparing PDF or Office documents as input for retrieval-augmented generation pipelines. Extracting structured text and tables from presentations and reports for LLM processing. Building a document ingestion service that exposes parsing through an HTTP API.
How do I install megaparse?
Install MegaParse with pip (Python 3.11 or newer): pip install megaparse Add your OpenAI or Anthropic API key to the .env file Install poppler and tesseract on your computer (needed for images and PDFs) Parse a document with MegaParse in Python
Where can I get megaparse?
The source code and installation instructions are on GitHub: https://github.com/The-Vibe-Company/megaparse

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects