MarkItDown

MarkItDown

Convert virtually any file format to clean, usable Markdown using MarkItDown — Microsoft's open-source Python library for document-to-Markdown conversion.

What you can do

• Convert PDFs: Extract text content from PDF files into structured Markdown.
• Office documents: Convert DOCX, XLSX, and PPTX files preserving headings, tables, and structure.
• Images with OCR: Extract text from scanned documents and photos via an LLM vision plugin.
• Audio transcription: Convert MP3, WAV, and M4A recordings to Markdown transcripts.
• Batch processing: Convert entire folders of mixed file types in a single pipeline run.
• RAG pipelines: Feed the output directly into vector stores for LLM retrieval.

Prerequisites

pip install markitdown
# or with all extras:
pip install "markitdown[all]"

Example Requests

• "Convert all PDFs in the docs/ folder to Markdown for RAG indexing."
• "Extract the text from this Excel spreadsheet as a Markdown table."
• "Transcribe this MP3 meeting recording to Markdown."
• "Convert this PPTX presentation to Markdown so I can summarize it."
• "Pull the text from this scanned PDF image using the LLM plugin."