Document-to-Markdown Done Right

Convert 14 office formats to Markdown in 5ms with anydoc. Open-source Rust library for Word, PPT, Excel, PDF and more.

In today’s increasingly widespread adoption of AI applications, feeding internal enterprise Word reports, PPT proposals, Excel spreadsheets, and PDF contracts into large language models for summarization, extraction, and analysis has become a routine operation for many teams.

However, each format is like a distinct “species,” requiring different toolchains and conversion logic, and the quality of the resulting Markdown varies widely.

Even worse, after you finally find a tool that can convert Word documents, it often doesn’t support PPT; when you find one that handles Excel, it struggles with merged cells in tables; and even for the same format, files generated by different versions can produce inconsistent conversion results.

Is there a “universal remedy”?

The anydoc library, written in Rust by the Firecrawl team, is designed precisely for this goal.

This open-source project has already garnered over 13,300 stars on GitHub. It supports unified conversion of 14 office formats into GitHub-Flavored Markdown, with a median conversion speed of under 5 milliseconds. More importantly, the Markdown structure produced from all formats is completely consistent.