pdf-oxide

What it is and what it does

pdf-oxide is a Python binding to a Rust-based PDF toolkit that handles text extraction, image extraction, markdown and HTML conversion, form field manipulation, and PDF creation. It ships as precompiled wheels for common platforms (Linux, macOS, Windows on x86_64 and ARM) with no system dependencies or build step required. The library exposes a page-based API where extraction is lazy—properties like `page.text` and `page.markdown()` compute results only when accessed.

The package targets developers building document pipelines, RAG systems, or LLM integrations who need reliable PDF parsing without the overhead of heavier alternatives. It supports Python 3.9 through 3.14, has zero runtime dependencies, and is licensed under MIT or Apache-2.0, making it suitable for commercial and closed-source projects. The fact sheet indicates active maintenance (latest release 17 days old, 955 GitHub stars) and no known security vulnerabilities.

Use it for:

• Extract text from PDFs for indexing into a vector database or RAG pipeline.
• Convert PDF documents to Markdown for processing by language models.
• Batch extract images from scanned documents or reports.
• Fill PDF form fields programmatically and save the result.
• Search for keywords across a PDF and retrieve their page and bounding-box locations.
• Generate PDFs from Markdown, HTML, or images as part of a document generation workflow.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Extracts text, images, and metadata from PDFs and converts them to Markdown or HTML, powered by a Rust core compiled to native wheels with no system dependencies.

Yes, if you need fast, reliable PDF text or image extraction without licensing restrictions. The permissive dual license, zero runtime dependencies, and precompiled wheels make it a low-friction alternative to AGPL-licensed competitors. Active maintenance and no known vulnerabilities support stability. Install friction is medium but manageable; the main constraint is Python 3.9+. Suitable for production use in commercial and closed-source projects.

Install

pdf-oxide on PyPI

pip install pdf-oxide

uv add pdf-oxide

poetry add pdf-oxide

Installing pdf-oxide

Before you install

Medium friction: precompiled wheels for Linux, macOS, and Windows (x86_64, aarch64, arm64) mean no build step or Rust toolchain required. Requires Python 3.9+. Active maintenance with latest release 17 days old.

License in practice

Dual MIT / Apache-2.0 license permits unrestricted commercial and closed-source use, unlike some competing PDF libraries. No copyleft obligations.

Quickstart

pip install pdf_oxide

from pdf_oxide import PdfDocument

with PdfDocument("paper.pdf") as doc:
    print(len(doc))  # page count
    for page in doc:
        text = page.text
        md = page.markdown(detect_headings=True)

Requires Python 3.9 or later.

Verify before relying

• Whether the 0.8ms mean and 100% pass rate claims are independently verified or from the maintainer's own benchmark.
• Whether image extraction returns raw bytes or requires additional dependencies to save to disk.
• Whether form field editing and PDF creation are fully stable or still experimental in this version.

Package facts

Evidence: pdf_oxide-0.3.77-cp38-abi3-macosx_10_12_x86_64.whl; pdf_oxide-0.3.77-cp38-abi3-macosx_11_0_arm64.whl; pdf_oxide-0.3.77-cp38-abi3-manylinux_2_28_aarch64.whl; pdf_oxide-0.3.77-cp38-abi3-manylinux_2_28_x86_64.whl; pdf_oxide-0.3.77-cp38-abi3-musllinux_1_2_aarch64.whl; pdf_oxide-0.3.77-cp38-abi3-musllinux_1_2_x86_64.whl; pdf_oxide-0.3.77-cp38-abi3-win_amd64.whl; pdf_oxide-0.3.77-cp38-abi3-win_arm64.whl

Keywords: pdf, text-extraction, pdf-parser, pymupdf-alternative, rag, llm, ocr, markdown, document-parser, pdf-to-text, pdf-to-markdown, data-extraction

Tags

More Python Modules packages