MarkItDown
Online Markdown tool for multi-format file conversion and AI OCR
About MarkItDown
MarkItDown is an online converter inspired by Microsoft's open-source tool that converts PDFs, DOCX, PPTX, spreadsheets, web pages, text, structured data, e-books, notebooks, and images into Markdown. It supports AI OCR, table recovery, and LaTeX math formula conversion for scanned PDFs, and can send clean output to RAG, LLM prompts, vector databases, and knowledge bases.
Best for
Best use cases
Batch document conversion
Text extraction from scanned PDFs
Preparing Markdown for RAG workflows
Vector database and knowledge-base imports
Compare before choosing
The official comparison page states that the Free plan allows up to 20MB per file, up to 3 files per batch, and has no monthly server-conversion limit; MarkItDown Pro raises the per-file limit to 100MB and the per-batch limit to 30 files, and supports unlimited saved results.
Features
Multi-format conversion
Supports PDF, DOCX, PPTX, XLS/XLSX, HTML/HTM, TXT/MD, CSV/TSV, JSON, XML, EPUB, IPYNB, public URLs, pasted tables, and PNG/JPG/WebP.
PDF and OCR processing
Extracts selectable PDF text while preserving structure as much as possible; scanned PDFs can use AI OCR, table recovery, and LaTeX math formula conversion.
AI workflow output
Clean output can be sent into RAG workflows, LLM prompts, vector databases, and knowledge bases.
Temporary file processing
Uploaded files pass through Cloudflare to a self-hosted Microsoft MarkItDown service, are processed as temporary files, and are deleted after conversion; source files and file-content logs are not retained, and results are saved only when selected.
Pros
- Supports many common file and image formats
- Scanned PDFs can use AI OCR and recover tables
- Output can be used for RAG, LLM prompts, and knowledge bases
Cons
- Free accounts include 10 OCR pages per month
Pricing Plans
MarkItDown Pro Yearly
- 单文件最大100MB
- 单批最多30个文件
- 无限保存结果
* Pricing information is for reference only. Please refer to the latest announcement on MarkItDown's official website.