PyMuPDF is a high-performance Python library, built on the MuPDF C engine, for extracting, analysing, converting, rendering and manipulating PDF and other documents, aimed at Python developers building data-extraction, document-processing and AI retrieval pipelines.

What it is

PyMuPDF is a Python binding over MuPDF, described in its README as a lightweight, fast C engine. It exposes both low-level control over document internals and high-level convenience APIs, and it ships with no mandatory external dependencies, so a single pip install pymupdf yields text extraction, rendering, conversion and page manipulation in one package. Input spans PDF and derivatives (PDF, XPS, EPUB, CBZ, MOBI, FB2, SVG, TXT, MD), raster images (PNG, JPEG, BMP, TIFF, GIF and more), Microsoft Office formats and Korean Office formats through the Pro extra, while output covers PDF, SVG and PNG/JPEG images.

The concrete problem it solves is fragmented document handling in the Python ecosystem. Text extraction, rendering, annotation, redaction, merging, splitting and format conversion are normally spread across several packages and external binaries, each with its own build and deployment requirements. PyMuPDF consolidates those tasks behind one library and one install, with pixel-accurate text extraction that carries font, colour and position metadata rather than plain strings.

Key capabilities

• High-level text extraction with font, colour and position metadata, plus table extraction (a listed topic).
• LLM-ready extraction through PyMuPDF4LLM (pip install pymupdf4llm), producing Markdown and JSON output for RAG and AI pipelines.
• Rendering to vector SVG pages and to raster image formats such as PNG and JPEG.
• Document manipulation: read, write, annotate, redact, merge and split.
• Broad input support: PDF, XPS, EPUB, CBZ, MOBI, FB2, SVG, TXT, MD, plus PNG, JPEG, BMP, TIFF and GIF.
• Office and Korean Office input (DOC, DOCX, XLS, XLSX, PPT, PPTX, HWP, HWPX) via the pymupdfpro extra.
• OCR for scanned pages and images through Tesseract, and an extended font collection via pymupdf-fonts.

Who uses it and how

• AI and RAG pipelines that need Markdown or JSON text from PDFs; the README states the project powers AI pipelines worldwide.
• Data-science and analytics teams doing bulk text and table extraction from large document sets, one of its declared topics.
• Digitisation and archival workflows that OCR scanned pages with Tesseract before indexing.
• Back-office conversion jobs that turn Office or HWP documents into PDF at full fidelity using pymupdfpro.
• Embedded deployments in containers and batch services, since the core library needs no external services and no mandatory dependencies.

Getting started

Install with pip install pymupdf. Pre-built wheels cover Windows, macOS and Linux on Python 3.10–3.14; where no wheel exists, pip compiles from source and a C/C++ toolchain is required.

How it compares

The facts provided name no paid products this project replaces, and no comparable Python tools. MuPDF appears only as the C engine PyMuPDF is built on, not as an alternative. On the evidence supplied here, it stands alone in this registry.

When to use it — and when not to

A self-hoster operates no database, storage or SMTP service, but optional features bring their own requirements: OCR needs Tesseract installed separately (brew install tesseract, sudo apt install tesseract-ocr), Office support needs the pymupdfpro extra, and platforms without a pre-built wheel must provide a C/C++ toolchain. The licence is AGPL-3.0, so teams shipping closed-source products that cannot meet AGPL obligations should look elsewhere. Those needing Office conversion without the Pro extra, or OCR without an extra system package, should also plan around those gaps.

pip install pymupdf

# More fonts
pip install pymupdf-fonts

# LLM-ready extraction
pip install pymupdf4llm

# Office support
pip install pymupdfpro

# OCR (Tesseract must be installed separately)
# macOS
brew install tesseract

# Ubuntu / Debian
sudo apt install tesseract-ocr

import pymupdf

doc = pymupdf.open("document.pdf")
for page in doc:
    print(page.get_text())

import pymupdf

doc = pymupdf.open("document.pdf")
page = doc[0]

blocks = page.get_text("dict")["blocks"]
for block in blocks:
    if block["type"] == 0:  # text block
        for line in block["lines"]:
            for span in line["spans"]:
                print(f"{span['text']!r}  font={span['font']}  size={span['size']:.1f}")

import pymupdf

doc = pymupdf.open("spreadsheet.pdf")
page = doc[0]

tables = page.find_tables()
for table in tables:
    print(table.to_markdown())

    # or get as Pandas DataFrame
    df = table.to_pandas()

import pymupdf

doc = pymupdf.open("document.pdf")
page = doc[0]

pixmap = page.get_pixmap(dpi=150)
pixmap.save("page_0.png")

import pymupdf

doc = pymupdf.open("scanned.pdf")
page = doc[0]

# Requires Tesseract installed and on PATH
text = page.get_textpage_ocr(language="eng").extractText()
print(text)

import pymupdf

md_doc = pymupdf.open("example.md")
md_doc.save("example.pdf")

import pymupdf4llm

md = pymupdf4llm.to_markdown("report.pdf")
# Pass directly to your LLM or vector store
print(md)

import pymupdf

doc = pymupdf.open("contract.pdf")
page = doc[0]

# Add a highlight annotation
rect = pymupdf.Rect(72, 100, 400, 120)
page.add_highlight_annot(rect)

# Add a redaction and apply it
page.add_redact_annot(rect)
page.apply_redactions()

doc.save("contract_redacted.pdf")

import pymupdf

merger = pymupdf.open()
for path in ["part1.pdf", "part2.pdf", "part3.pdf"]:
    merger.insert_pdf(pymupdf.open(path))

merger.save("merged.pdf")

import pymupdf.pro

pymupdf.pro.unlock("YOUR-LICENSE-KEY")

doc = pymupdf.open("presentation.pptx")
pdf_bytes = doc.convert_to_pdf()

with open("output.pdf", "wb") as f:
    f.write(pdf_bytes)

import pymupdf4llm
import pymupdf.pro

pymupdf.pro.unlock("YOUR-LICENSE-KEY")

md = pymupdf4llm.to_markdown("document.docx")
print(md)