anydoc

Fast Rust library that converts documents (Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF) into clean GitHub-Flavored Markdown. Includes bindings for Node.js and Python.

Built by Firecrawl to turn any office document into LLM-ready Markdown in single-digit milliseconds, with one consistent output no matter which format goes in. It powers Firecrawl Parse, so if you'd rather not run it yourself, the hosted API gives you the same conversion plus our OCR models for the scanned pages anydoc can't read on its own.

Features

• One output for every format. Each format parses into a shared document model and renders through a single Markdown serializer, so escaping, tables, heading anchors, and footnotes behave identically whether the input was a .doc from 2003 or a .pptx from yesterday.
• Full document structure. Headings with anchors, bold/italic/strikethrough, inline code and code blocks, links and internal cross-references, bulleted/numbered/nested/task lists with the source's own numbering, tables with merged cells and header rows, block quotes, footnotes and endnotes, and speaker notes.
• Embedded assets. Images and embedded objects render as their alt text in the Markdown, and the raw bytes stay available on the document model, tagged with their media type. Images with an external URL become ordinary Markdown images.
• Content-based format detection. The format is read from the bytes themselves (PDF header, RTF open group, OLE stream names, ZIP package mimetype), so mislabeled files still convert correctly.
• Fast. Pure Rust, no ML models, no external services. Median conversion time is under 5ms per document.
• Bindings that stay out of the way. Node.js conversion runs on the libuv thread pool and never blocks the event loop; Python releases the GIL so other threads keep running. TypeScript types and Python stubs ship with the packages.
• PDF support built in. Text-based PDFs convert locally through pdf-inspector, no OCR service required.

Supported formats

Format	Extensions
Word	.doc, .docx, .docm
PowerPoint	.ppt, .pps, .pot, .pptx, .pptm, .ppsx, .ppsm
Excel	.xls, .xlsx, .xlsm, .xlsb
OpenDocument	.odt, .ods, .odp
Rich Text Format	.rtf
EPUB	.epub
CSV	.csv
PDF	.pdf

PDFs take a shortcut: pdf-inspector emits Markdown directly, so use to_markdown / to_markdown_bytes for them rather than to_document. Scanned and image-only PDFs need OCR, so anydoc returns an unsupported error for them; route those to Firecrawl Parse, which OCRs them and returns the same Markdown.

Benchmark

anydoc is measured against six other converters on 100 real-world documents spanning fourteen formats. Scores run from 0 to 100, higher is better; speed is the median time to convert one document.

tool	formats	median ms	docs judged	score	completeness	structure	formatting	cleanliness
anydoc	14/14	4.7	94	80	88	78	77	79
libreoffice	12/14	1129.5	87	40	59	43	43	24
unstructured	8/14	572.9	58	65	76	62	52	67
markitdown	6/14	134.8	33	65	80	67	61	53
pandoc	5/14	102.1	34	57	75	57	58	39
docling	4/14	513.6	21	57	63	59	57	52
mammoth	1/14	52.5	8	70	85	68	74	55

Per format, like for like:

format	anydoc	libreoffice	unstructured	markitdown	pandoc	docling	mammoth
doc	88	58	68	-	-	-	-
docm	82	49	-	-	-	-	-
docx	86	53	56	72	68	68	70
epub	74	-	74	77	53	-	-
odp	87	22	-	-	-	-	-
ods	82	42	-	-	-	-	-
odt	80	52	70	-	61	-	-
ppt	80	25	-	-	-	-	-
pptx	76	22	-	59	-	50	-
rtf	89	58	48	-	46	-	-
xls	77	40	68	64	-	-	-
xlsm	70	30	-	-	-	-	-
xlsx	70	31	69	55	-	51	-

How quality was scored: an LLM judge (Claude Sonnet 5) compares two tools' outputs blind against ground truth: the document's first six pages, rendered to images by LibreOffice. Each output is scored on completeness, structure, formatting, and cleanliness. Every pair is judged twice with the outputs swapped to cancel position bias, for 479 verdicts in total. Each tool's score averages its per-format scores over the formats it supports, so a corpus heavy in one format can't skew it. It also means each row averages a different set of formats (mammoth's 70 is docx alone, while anydoc's 80 spans all fourteen), so the per-format table is the fair comparison.

Speed is one warm conversion per document on a Ryzen 9 9950X3D (Windows 11, 64 GB DDR5-6400). anydoc and the Python libraries are timed with process spawn excluded; the CLI tools include it, since that is how they are used. The harness lives in bench/; the corpus is not redistributable and is not in the repo.

Best fit: pipelines that receive a mixed bag of office documents and need one consistent, structured Markdown output. In this comparison, anydoc was the only tool to cover all fourteen formats, scored highest on every judged format except EPUB, and converted documents an order of magnitude faster than the next-fastest tool.

Quick start

Node.js

npm install @firecrawl/anydoc

import { toDocument, toMarkdown, toMarkdownBytes } from '@firecrawl/anydoc';

// From a file path:
const markdown = await toMarkdown('report.docx');

// From bytes, with the format detected from the content:
const fromBytes = await toMarkdownBytes(bytes);

// Or name it, which signature-less formats (CSV) need:
const fromCsv = await toMarkdownBytes(bytes, 'csv');

// Or stop at the document model, which also carries embedded assets:
const document = await toDocument(bytes);

Full API reference: node/README.md

Python

pip install firecrawl-anydoc

import anydoc

# From a file path:
markdown = anydoc.to_markdown("report.docx")

# From bytes, with the format detected from the content:
markdown = anydoc.to_markdown_bytes(data)

# Or name it, which signature-less formats (CSV) need:
markdown = anydoc.to_markdown_bytes(data, "csv")

# Or stop at the document model, which also carries embedded assets:
document = anydoc.to_document(data)

Full API reference: python/README.md

Rust

cargo add anydoc

// From a file path:
let markdown = anydoc::to_markdown("report.docx")?;

// From bytes, with the format detected from the content:
let markdown = anydoc::to_markdown_bytes(&bytes, None)?;

// Or name it, which signature-less formats (CSV) need:
let markdown = anydoc::to_markdown_bytes(&bytes, anydoc::Format::Csv)?;

// Or stop at the document model, which also carries embedded assets:
let document = anydoc::to_document(&bytes, None)?;

CLI

A convert CLI ships in examples/, in all three languages:

cargo run --release --example convert -- file.docx [-f csv] [-o out.md] [--assets dir]
node examples/convert.mjs file.docx [-f csv] [-o out.md] [--assets dir]
python examples/convert.py file.docx [-f csv] [-o out.md] [--assets dir]

Format detection

The format is read from the file content, using the marker its specification designates: the PDF header, the RTF open group, OLE stream names, the ZIP package mimetype and content types. CSV has no such marker, so the extension or an explicit format names it instead.

Format::from_bytes(&bytes); // Some(Format::Docx), or None when nothing matches
Format::from_extension("pptm"); // Some(Format::Pptx)
Format::from_path(Path::new("report.odt")); // Some(Format::Odt)

The same three functions exist in Node (formatFromBytes, ...) and Python (anydoc.format_from_bytes, ...).

How it works

document bytes
  │
  ├─► format detection      → content markers, not the extension
  │
  ├─► format parser          → one per format (doc, docx, ppt, pptx, xls,
  │                            xlsx, odt/ods/odp, rtf, epub, csv)
  │         │
  │         └─► Document     → shared model: blocks, inlines, tables,
  │                            footnotes, assets
  │               │
  │               └─► GFM serializer → Markdown
  │
  └─► PDF → pdf-inspector    → Markdown directly

Because every format funnels through the same document model and serializer, output quirks get fixed once. A table-escaping fix for docx is automatically a table-escaping fix for rtf, odt, and everything else.

Development

cargo test
cd node && npm install && npm run build && npm test
cd python && pip install maturin && maturin develop && python -m unittest discover -s tests

A committed fixture corpus under tests/fixtures/ is snapshot-tested, tests/robustness.rs mutation-tests every fixture, and fuzz/ carries cargo-fuzz targets per format. The speed and quality benchmark lives in bench/.

Releases are tagged v<version>, which publishes the crate, the npm package, and the PyPI wheels from .github/workflows/release.yml. The version lives in three places, bumped together for a release:

• Cargo.toml: the crate
• node/package.json: the npm package
• python/Cargo.toml: the wheel (python/pyproject.toml reads it)

License

MIT