Description

<p align="center"> <a href="https://github.com/pymupdf/pymupdf4llm"> <img loading="lazy" alt="PyMuPDF logo" src="https://pymupdf.readthedocs.io/en/latest/_static/sidebar-logo-light.svg" width="150px" height='auto' /> </a> </p> # PyMuPDF4LLM [](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/) [](https://artifex.com/licensing/gnu-agpl-v3) [](https://pepy.tech/projects/pymupdf4llm) [](https://discord.gg/7pH3gqcRtg) **PyMuPDF4LLM** is a lightweight extension for **PyMuPDF** that turns documents into clean, structured data with minimal setup. It includes layout analysis *without* any GPU requirement. **PyMuPDF4LLM** makes it easy to extract document content in the format you need for **LLM** & **RAG** environments. It supports structured data extraction to **Markdown**, **JSON** and **TXT** , as well as [LlamaIndex](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/index.html#with-llamaindex) and [LangChain](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/index.html#with-langchain) integration. ## Features - Parsing of [multiple document formats](https://pymupdf.readthedocs.io/en/latest/about.html#feature-matrix). - Export structured data as Markdown, JSON and plain text output formats. - Support for multi-column pages. - Support for image and vector graphics extraction. - Layout analysis for better semantic understanding of document structure. - Support for page chunking output. - Integration with popular AI frameworks. ## Installation ```bash $ pip install -U pymupdf4llm ``` > This command will automatically install or upgrade [PyMuPDF](https://github.com/pymupdf/PyMuPDF) as required. ## Execution ### Markdown ```python import pymupdf4llm # The remainder of the script is unchanged md_text = pymupdf4llm.to_markdown("input.pdf") # now work with the output data, e.g. store as a UTF8-encoded file import pathlib pathlib.Path("output.md").write_text(md_text) ``` ### JSON ```python import pymupdf4llm json_text = pymupdf4llm.to_json("input.pdf") # now work with the output data, e.g. store as a UTF8-encoded file import pathlib pathlib.Path("output.json").write_text(json_text) ``` ### Plain Text ```python import pymupdf4llm plain_text = pymupdf4llm.to_text("input.pdf") # now work with the output data, e.g. store as a UTF8-encoded file import pathlib pathlib.Path("output.txt").write_text(plain_text) ``` ## Documentation Check out the [PyMuPDF4LLM documentation](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm), for details on installation, features, sample code and the [full API](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/api.html). ## Examples Find our [examples on GitHub](https://github.com/pymupdf/pymupdf4llm/tree/main/examples). ## Integrations For your AI application development, check out our [integrations](https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/index.html#integrations) with popular frameworks. ## Support You can get support for PyMuPDF4LLM via a number of options: - [GitHub Issue Board](https://github.com/pymupdf/pymupdf4llm/issues) - [Discord](https://discord.gg/7pH3gqcRtg) - [MuPDF Forum](https://forum.mupdf.com)