iformat.io Logo iformat.io
iformat.io Logo iformat.io

Convert Any File to Markdown

Convert PDFs, Word docs, Excel spreadsheets, PowerPoint decks, HTML pages, and EPUB ebooks to clean Markdown — for free, in your browser, with no signup. The output is ready to paste into GitHub, Notion, Obsidian, a static site generator, or the context window of an LLM. All six converters run on Microsoft's open-source MarkItDown library, which preserves structure (headings, tables, lists) instead of dumping raw plain text.

Why Markdown

LLM-friendly

Markdown is the format ChatGPT, Claude, and Gemini were trained to write. Feeding Markdown into a model uses fewer tokens than PDF binary or HTML with inline styles, so you fit more content into a context window and pay less per API call.

Git-friendly

Diffs on a Markdown file show real content changes, one line at a time. Diffs on a Word doc or PDF show binary noise. That's why every modern docs-as-code workflow (Docusaurus, mkdocs, Hugo, Astro) is Markdown-native.

Portable

A .md file opens in every text editor, renders in GitHub and GitLab natively, imports into Notion / Obsidian / Logseq / Bear, and converts to HTML or PDF in one command via pandoc. PDF and DOCX are ends; Markdown is a starting point.

Long-term readable

A Markdown file written today will still open in 30 years. A .docx from 1998 already needs a compatibility layer to open cleanly. If you're archiving anything important, Markdown is the safer bet.

Frequently asked questions

Do the files get uploaded to your server?

Yes — the conversion runs on our server, not in your browser (MarkItDown is a Python library). Uploaded files are held in memory only, never written to disk, and discarded the instant the Markdown response is sent back. Nothing is logged or persisted.

Are there rate limits?

Yes — 5 conversions per hour per IP address for the free tier, and files must be under 20 MB. That's enough for typical individual use. If you need higher throughput, run MarkItDown yourself via `pip install markitdown` and the CLI.

Why MarkItDown and not another library?

MarkItDown is Microsoft's open-source project (github.com/microsoft/markitdown) built specifically for LLM ingestion pipelines. It handles a wider range of formats than pandoc for our use cases, preserves table structure better than pdfplumber, and is actively maintained.

Can I convert images or audio to Markdown?

Not yet — those formats need an LLM (vision or transcription) which costs per-request. On the roadmap: image-to-Markdown (with OCR + optional vision descriptions) and audio-to-Markdown (with Whisper transcription). Follow the blog for updates.

Can I run this locally instead?

Yes. Install with `pip install markitdown` and use the CLI: `markitdown file.pdf > output.md`. Our web tool is a hosted convenience for one-off conversions and non-technical users; the library is open source and MIT-licensed.

Powered by Microsoft MarkItDown · MIT-licensed

Reviewed by Editorial team Updated