---
id: docling-parse
version: "7.13.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# docling-parse — Simple package to extract text with coordinates from programmatic PDFs

License: permissive · Maintenance: active · Downloads: 4.6M/mo

## What it is and what it does

Docling Parse is a Python wrapper around a C++ PDF parser that extracts structured text, geometric coordinates, and images from programmatic PDFs. It splits parsing into two phases: a fixed `DecodeConfig` applied at document open time (controlling sanitization and glyph handling) and a per-page `ContentConfig` that determines what to compute and materialize (character cells, word cells, line cells, shapes, bitmaps). This separation allows cheap initial loading and selective enrichment on demand—if you request richer output later, the page is re-decoded automatically.

The package supports both sequential parsing (one PDF at a time) and parallel multi-threaded parsing with backpressure control. It includes a CLI for single-file processing and integrates with the broader Docling PDF conversion ecosystem. The library is actively maintained, supports Python 3.10–3.14 across major platforms, and provides performance benchmarks against other PDF packages.

Use it for:

- Extract word-level bounding boxes and text from PDFs for document layout analysis or OCR validation.
- Batch-process multiple PDFs in parallel with configurable thread pools and result backpressure.
- Render pages as images with overlaid cell boundaries (character, word, or line level) for debugging or visualization.
- Selectively materialize only the content levels needed per page to optimize memory and CPU in large-scale workflows.
- Integrate PDF parsing into document conversion pipelines that require both text and spatial metadata.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Extracts text, coordinates, and bitmap images from programmatic PDFs with support for character, word, and line-level granularity, offering both sequential and multi-threaded parsing modes.

Yes. Docling Parse is actively maintained, permissively licensed, and offers a well-designed API for structured PDF extraction with multi-threaded support. Install friction is moderate due to compiled components, but pre-built wheels cover all major platforms and Python versions. Suitable for production document processing workflows.

## Install

pip install docling-parse
uv add docling-parse
poetry add docling-parse

## Installing docling-parse

Before you install: Medium install friction due to compiled C++ components with pre-built wheels for Python 3.10–3.14 across macOS, Linux, and Windows. Active maintenance with a release on 2026-08-14 and 326 repository stars.

License in practice: MIT license permits unrestricted use, modification, and distribution in both open-source and commercial projects.

Quickstart:

pip install docling-parse

from docling_parse.pdf_parser import DoclingPdfParser, DecodeConfig, ContentConfig, ContentLevel

parser = DoclingPdfParser(loglevel="fatal")
pdf_doc = parser.load(
    path_or_stream="file.pdf",
    decode_config=DecodeConfig(do_sanitization=True),
    content_config=ContentConfig(
        word_cells_content_level=ContentLevel.COMPUTE_AND_MATERIALIZE,
    ),
)

for page_no, page in pdf_doc.iterate_pages():
    for word in page.iterate_cells():
        print(word.rect, word.text)

Requires Python >=3.10; compiled wheels depend on system C++ runtime libraries.

Verify before relying:

- Whether the package handles encrypted or password-protected PDFs beyond what the CLI suggests.
- Performance characteristics on very large PDFs or batch workloads compared to alternatives.
- Memory footprint when materializing all cell levels for high-page-count documents.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 4.6M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

pdf text extraction with coordinates, pdf parsing python, extract text from pdf, pdf document parser, programmatic pdf processing, pdf image extraction, batch pdf parsing, pdf-parsing, document-extraction, multi-threaded

[View on SkillFed](https://skillfed.io/packages/docling-parse) · [View on PyPI](https://pypi.org/project/docling-parse/)