Skip to main content
DocVortex logo

DocVortex

Native document parsing. One structure, many outputs.

PyPI Python Downloads Monthly Downloads CI License: MIT

English · 简体中文

Performance · Quick start · Formats · Documentation

Documents in. Possibilities out.

DocVortex is a standalone Python engine for parsing and converting documents. It reads native text and document structure into a unified representation, then exports the result in the formats your workflow needs.

  • Multi-format input — read native-text PDFs, Office files, OpenDocument files, EPUB, HTML, OFD, CSV and TSV.
  • Parse once, export many times — reuse the same result for Markdown, HTML, LaTeX, DOCX, EPUB, PDF and structured JSON.
  • Portable results — save document structure and image assets in a Bundle, then export again without the source file.
  • Composable APIs — use the complete pipeline or integrate analysis, postprocessing and rendering separately.

Native parsing works without an OCR or VLM inference service. Use DocVortex directly through its CLI or Python SDK.

DocVortex does not download models or perform PDF OCR. When OCR is required, it reports pdf_ocr_unsupported and asks you to download and install MinerU to process the document.

DocVortex pipeline: native documents become a unified representation, then Markdown, HTML, LaTeX, DOCX, EPUB, PDF or structured JSON.

PDF performance

Release 0.5.10 uses Rust acceleration: public parsing is 2.93× as fast as 0.4.25 on the tested corpus.

Pipeline 0.4.25 python 0.5.10 rust Speedup Time reduction Peak RSS reduction
DocVortex public parse() 42.36 s 14.48 s 2.93× 65.82% 31.17%
MinerU Flash 44.50 s 14.44 s 3.08× 67.55% 29.84%
MinerU shared PDF processing 76.28 s 12.85 s 5.94× 83.16% 35.65%

Same-machine warm-run comparison on 32 PDFs / 299 pages (31 PDFs for Flash). The existing 0.4.25 measurements are reused. For 0.5.10, each document is warmed once and timed five times in each of three balanced rounds; times sum the medians of the three per-document medians. RSS reductions are medians of per-document process-tree peak RSS reductions. MinerU results include improvements in both projects. The shared pipeline excludes model computation; excluding the extreme dense-table sample, its speedup is 2.87×.

The non-PDF optimization report records the original-to-final comparison, per-format timings, memory checks and remaining hotspots.

Quick start

Requires GIL-enabled CPython 3.11–3.14, pypdfium2 5.13.0 or newer, and ONNX Runtime 1.24.4 or newer (DirectML on Windows x64). On macOS, installation requires Apple Silicon and macOS 14 or newer; Intel Macs are no longer supported as of 0.6.0.

Version 0.6 keeps the Python SDK and CLI and requires its bundled Rust native extension. Releases contain ABI3 wheels for Linux x86_64/aarch64 (glibc 2.28+), Windows x86_64/ARM64, and macOS arm64. Pure Python wheels and source distributions are no longer published. PyPy and free-threaded Python are not supported. Starting with 0.6.1, DocVortex ignores the retired DOCVORTEX_BUILD_NATIVE and DOCVORTEX_COMPUTE_BACKEND variables, regardless of their values. The 0.6.0 package still rejects these variables. Existing 0.5 installations can stay on docvortex<0.6 while migrating.

Install

pip install docvortex

Command line

Convert a text PDF to Markdown:

docvortex convert report.pdf --format markdown --output output/report.md

Replace report.pdf with a local file in any supported input format. Use --format to choose the output; run docvortex convert --help for options. The root-level --log-level option controls loguru output and defaults to info. It must precede the command; DOCVORTEX_LOG_LEVEL=warning can also configure it.

Python

Parse a document once and export it twice:

import docvortex

result = docvortex.parse("report.pdf")
result.export("output/report.md", output_format="markdown")
result.export("output/report.docx", output_format="docx")

The result owns its document structure and assets, so further exports do not reopen or reparse the source. Existing output files are protected by default; use overwrite=True in Python or --overwrite in the CLI to replace them.

Supported formats

Native inputs · 20 formats

Document family File extensions
PDF with native text or scanned pages .pdf
Word & rich text .doc, .docx, .rtf
Presentations .ppt, .pptx
Spreadsheets .xls, .xlsx, .csv, .tsv
OpenDocument .odt, .ods, .odp
E-books .epub
Web documents .html, .htm, .shtml
Web archives .mhtml, .mht
Open Fixed-layout Document .ofd

Outputs · 7 formats

Output --format / output_format
Markdown markdown
HTML html
LaTeX latex
Word document docx
EPUB e-book epub
PDF pdf
Structured JSON structured_content

PPT/PPTX and XLS/XLSX are input formats only. Structured JSON is an export format; the document JSON protocol separately defines the analysis and intermediate representations.

Save now, export later

A Bundle packages the parsed document and its image assets for reuse across processes or machines. Load it whenever you need another output format:

import docvortex

result = docvortex.parse("report.pdf")
result.save_bundle("output/report.bundle")

restored = docvortex.load_bundle("output/report.bundle")
restored.export("output/report.epub", output_format="epub")

The restored result works without the original file. See the usage guide for Bundle contents, asset handling and overwrite rules.

Need only the title, authors or other source properties? docvortex.extract_metadata() reads metadata without parsing the document body.

Choose the right workflow

  • PDF parsing defaults to pdf_structure="auto", which uses available Tagged PDF semantics and reading order, with Flash rules for uncovered content. pdf_structure="off" selects geometric parsing. The CLI accepts --pdf-structure auto|off. Original roles and block associations are recorded in docvortex_pdf_structure extensions. Tagged PDF usage
  • PDF classification under parse_mode="auto" first accepts sufficient structure evidence, then classifies the selected pages as a whole when needed. A txt result uses native parsing. An ocr result raises DocumentError with code pdf_ocr_unsupported and a MinerU download link. parse_mode="txt" forces native parsing. The CLI accepts --parse-mode auto|txt|ocr for PDF input.
  • PDF OCR: parse_mode="ocr" and --parse-mode ocr are retained for compatibility and always report pdf_ocr_unsupported for PDF input. Download and install MinerU for OCR. DocVortex does not install or invoke MinerU automatically.
  • Mixed PDFs: auto mode does not classify or switch backends per page. A selection classified as txt keeps native parsing even if it contains scanned pages. Use MinerU when those pages need OCR.
  • PDF classification: docvortex classify report.pdf returns txt or ocr without inference or networking. Metadata extraction and the low-level PdfModel.predict() native text API also remain offline.
  • PDF export: PDF sources with page geometry default to block layout restoration, with selectable text and HTML-based tables; charts retain region images. Other sources and older results use semantic reflow. Use --pdf-layout original|reflow to select explicitly; fonts, line breaks and drawing instructions are not reproduced losslessly. See PDF output layout.

Documentation

Guide What you will find
Usage Stage APIs, PDF pages, classification, images and Bundles
Agent skill CLI and Python SDK workflows for agents; copy the entire skills/docvortex folder to reuse
Examples Local samples for PDF, Office, EPUB, web documents and OFD with a runnable demo
Metadata Source properties and per-format coverage
JSON protocol Document schemas, extensions and protocol migration
HTML protocol Semantic markers and round trips
Public SDK & migration Supported integration boundaries and the 0.4 upgrade
Rendering ownership DocVortex exports and MinerU-specific renderers

Development

See Repository architecture for directory responsibilities, the Python/Rust boundary, and native development commands. Source builds require the Rust toolchain pinned in rust-toolchain.toml and the platform's C/C++ linker.

From a local checkout:

uv venv
uv pip install -e ".[test,dev]"
uv run --no-project python -m pytest -q
uv run --no-project ruff check src
uv run --no-project ruff format --check src
uv build --wheel

Bug reports and contributions are welcome. When reporting a parsing issue, include a reproducible command and a sample document you can share in GitHub Issues.

License

DocVortex project code is released under the MIT License.

Metadata

Release files for docvortex 0.6.23

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distributions (wheels)

Table of built distributions (wheels) for docvortex 0.6.23
File Actions
docvortex-0.6.23-cp311-abi3-win_arm64.whl CPython 3.11 abi3 Windows ARM64 Details for docvortex-0.6.23-cp311-abi3-win_arm64.whl
docvortex-0.6.23-cp311-abi3-win_amd64.whl CPython 3.11 abi3 Windows x86-64 Details for docvortex-0.6.23-cp311-abi3-win_amd64.whl
docvortex-0.6.23-cp311-abi3-manylinux_2_28_x86_64.whl CPython 3.11 abi3 Linux glibc 2.28+ x86-64 Details for docvortex-0.6.23-cp311-abi3-manylinux_2_28_x86_64.whl
docvortex-0.6.23-cp311-abi3-manylinux_2_28_aarch64.whl CPython 3.11 abi3 Linux glibc 2.28+ ARM64 Details for docvortex-0.6.23-cp311-abi3-manylinux_2_28_aarch64.whl
docvortex-0.6.23-cp311-abi3-macosx_14_0_arm64.whl CPython 3.11 abi3 macOS 14.0+ ARM64 Details for docvortex-0.6.23-cp311-abi3-macosx_14_0_arm64.whl

Total release size: 26.8 MB

Release files / docvortex-0.6.23-cp311-abi3-win_arm64.whl

Download URL docvortex-0.6.23-cp311-abi3-win_arm64.whl
Size 5.2 MB
Tags CPython 3.11 Windows ARM64 abi3
SHA-256 checksum
How to use checksums
3a350c029a4a091ed16466333392111dad2ceece7532267e569efc7b110494ea
BLAKE2b-256 checksum
How to use checksums
198542c04eeba54c9ed0117e0fd235cfe4dbfaa3f149b7e4bc5893a34c09363a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

Release files / docvortex-0.6.23-cp311-abi3-win_amd64.whl

Download URL docvortex-0.6.23-cp311-abi3-win_amd64.whl
Size 5.4 MB
Tags CPython 3.11 Windows x86-64 abi3
SHA-256 checksum
How to use checksums
81569ac662da18900440acdb7c594cf4e132718a84c1ed3760324e2649dd9190
BLAKE2b-256 checksum
How to use checksums
4215e31c265bdfcb57d8695b35c240c24ce4ddd8f502e0f1af15b0d740224ab7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

Release files / docvortex-0.6.23-cp311-abi3-manylinux_2_28_x86_64.whl

Download URL docvortex-0.6.23-cp311-abi3-manylinux_2_28_x86_64.whl
Size 5.5 MB
Tags CPython 3.11 Linux glibc 2.28+ x86-64 abi3
SHA-256 checksum
How to use checksums
438604fe0f9006429c55cddc5832c3e1a227f94f92012c2ea84fa3e721d5aa2f
BLAKE2b-256 checksum
How to use checksums
e8e2702f54dd5657a7449e9b1edba5b59d8b86d26c0d137ddfc3b52d0aba8a8a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

Release files / docvortex-0.6.23-cp311-abi3-manylinux_2_28_aarch64.whl

Download URL docvortex-0.6.23-cp311-abi3-manylinux_2_28_aarch64.whl
Size 5.4 MB
Tags CPython 3.11 Linux glibc 2.28+ ARM64 abi3
SHA-256 checksum
How to use checksums
5a41b00ea21e1c1f6e5cbca3bf3686906d95b3ddd08c1a22f44afff4507aa3aa
BLAKE2b-256 checksum
How to use checksums
1f04df0d65eb9cd9ae43b28a9026cdce7072b5ca4636b66bd60e5fdd2ff38cc0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

Release files / docvortex-0.6.23-cp311-abi3-macosx_14_0_arm64.whl

Download URL docvortex-0.6.23-cp311-abi3-macosx_14_0_arm64.whl
Size 5.3 MB
Tags CPython 3.11 abi3 macOS 14.0+ ARM64
SHA-256 checksum
How to use checksums
432f8a0737b7ef6a5e466ba1d3cb24e791cfc98dc801b41cf4f7433d13efe257
BLAKE2b-256 checksum
How to use checksums
317428e14a509c425538b4497b14dda1a49e9c6dffce3a0f149225bec0c8f4fb
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

Security for docvortex 0.6.23

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 5 files in this release have been uploaded with provenance metadata.

Select from 5 files

Attestation for docvortex-0.6.23-cp311-abi3-win_arm64.whl

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 16, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.6.23 This release

5 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page