Skip to content

MDFlux v0.1.0

Latest
Latest

Choose a tag to compare

View all tags
@ibrahimqureshae ibrahimqureshae released this 22 Jun 02:36
· 1 commit to main since this release
v0.1.0
d3013d0

Turn any document into clean, AI-ready Markdown. Runs on your machine, nothing uploaded, and it reads the scanned PDFs other tools miss.

This is the first public build of MDFlux, a portable Windows app built on Microsoft's MarkItDown.

What it does

  • Converts PDF, DOCX, PPTX, XLSX, EPUB, HTML, CSV, JSON, XML, images, and audio into clean Markdown.
  • Reads scanned, image-only PDFs with built-in OCR (plain text extractors return zero characters on these).
  • Uses up to 6 times fewer tokens than feeding pages to a vision model, so every downstream LLM call costs less.
  • Runs fully offline after a one-time first-run setup. No API key, no account, nothing uploaded.
  • Batch-converts a whole folder with progress, cancellation, and per-file diagnostics.
  • Optional cleanup: off, rule-based, or a local AI pass.

Install

  1. Download MDFlux_0.1.0_portable.zip below.
  2. Extract it anywhere (for example C:\Apps\MDFlux\).
  3. Run MDFlux.exe.

This is a portable app, so there is no installer. The first launch sets up a private, self-contained Python environment (one time, needs internet). After that it runs offline.

Notes for the first run

  • Windows SmartScreen may show a one-time "Windows protected your PC" message. The build is open source and unsigned, so click "More info" then "Run anyway". Code signing is on the roadmap.

Verify your download (optional)

SHA-256 of MDFlux_0.1.0_portable.zip:

6736097a13e307a60deb7d6b428921064c5655dc8f610b080c1e321d12e1a032

MIT licensed. Built on Microsoft's MarkItDown.

Assets 3
Loading