PDF2MD icon
PDF2MD icon

PDF2MD

A REST API and a hosted MCP that turn PDFs into clean, LLM-ready Markdown for agents and RAG. Own engines (MinerU, Docling), real tables, formulas and OCR. Web app and Chrome extension too.

Get clean, LLM-ready Markdown from any PDF: real tables, formulas and OCR, in the browser, via a REST API or a hosted MCP.

Cost / License

  • Freemium (Subscription)
  • Proprietary

Platforms

  • Online
  • Software as a Service (SaaS)
  • Google Chrome
Convert PDF to Markdown in the browser: drop a file or paste a URL, watch jobs convert in your free slots.
Clean Markdown output: real tables and formulas rendered as typeset math, with a rendered or raw preview.
+2
Developer hub: one engine across a REST API, a hosted MCP and a ChatGPT Custom GPT - convert from code or an agent.
PDF2MD alternatives 
0likes
0comments
8alternatives
0articles

Features

  1.  Support for MarkDown
  2.  OCR
  3.  Markdown preview
  4.  Save as Markdown
  5.  REST API
  6.  Convert PDF to Text
  7.  Model Context Protocol (MCP) Support

 Tags

PDF2MD News & Activities

Highlights All activities

Recent activities

PDF2MD information

  • Developed by

    AM flagSHIFT LLC

  • Licensing

    Proprietary and Freemium product.

  • Pricing

    Subscription ranging between $9 and $49 per month + free version with limited functionality.

  • Alternatives

    8 alternatives listed

  • Supported Languages

    • English
    • Spanish
PDF2MD was added to AlternativeTo by Dmitry Petrakov on and this page was last updated .
No comments or reviews, maybe you want to be first?

What is PDF2MD?

A developer / AI platform that turns PDFs into clean, LLM-ready Markdown. It is built around a programmable engine, not a one-off web tool: the same conversion is a hosted REST API and a hosted MCP (Model Context Protocol) endpoint, so scripts and AI agents convert PDFs as a built-in tool. It runs its own document-understanding engines, the open-source MinerU (default, robust on dense/complex layouts) and Docling (fast on clean documents), so it is not an LLM wrapper.

One engine, multiple ways to integrate:

  • REST API – create a job, poll status, download Markdown; Bearer API key, idempotent create, webhooks and batch create on higher tiers.
  • Hosted MCP – a Model Context Protocol endpoint so AI agents can convert PDFs as a native tool, with a ready-made ChatGPT Custom GPT.
  • Web app – for people: drop a PDF or paste a URL and get Markdown in the browser, no install.
  • Chrome extension – convert any PDF straight from a tab, sharing the same engine and account.

The output is real Markdown: reading order preserved, real tables instead of broken columns, formulas kept as LaTeX, images embedded or as lightweight placeholders, and OCR for scanned and image-only PDFs across many languages (including Cyrillic). Markdown is compact, structure-preserving and token-efficient, so it drops straight into ChatGPT, Claude, Gemini or a RAG pipeline. Files auto-delete after a short retention window and are never used for advertising or model training.

It's freemium: a capable free tier handles everyday conversion within set limits (file size, concurrent slots, processing time, retention), and paid plans raise those limits and add webhooks, batch creation and higher queue priority for heavier or automated workloads.

Official Links

AppStores & Other Links