Uploading
Processing file 0 of 1
Preparing file... (--)
Preparing file... (--)
在线将 PDF 转换为 Markdown,以用于 AI 工作流程、文档和发布。针对普通文本 PDF 使用 Basic(快速),针对扫描件、表格和复杂布局使用 Deep Parse。
Choose between lightweight MarkItDown extraction and AI-powered structured parsing based on the file type and document complexity you upload.
Basic supports: PDF (.pdf); Word (.docx); PowerPoint (.pptx); Excel (.xlsx, .xls); CSV (.csv); Markdown/Text/JSON (.md, .markdown, .txt, .text, .json, .jsonl); HTML/XML/Feeds (.html, .htm, .xml, .rss, .atom); EPUB/ZIP (.epub, .zip); Outlook Message (.msg); Jupyter Notebook (.ipynb); Images (.jpg, .jpeg, .png)
Deep Parse supports: PDF (.pdf); Word (.docx); Images (.jpg, .jpeg, .png)
Turn supported documents into Markdown that is easier to search, chunk, reuse, and send into AI or RAG workflows.
Use Basic for PDFs, Office files, text, EPUB, ZIP, notebooks, images, and Outlook messages. Switch to Deep Parse for PDF, DOCX, or image files when scans, tables, formulas, or layout structure need stronger parsing.
Deep Parse also returns structured outputs such as JSON and layout artifacts, which are useful for document pipelines and secondary processing.
Deep Parse uses an open-source parser. Source code and license details
PDF 转 Markdown 转换器
Basic mode creates a Markdown file from any supported MarkItDown input. Deep Parse supports PDF, DOCX, and image inputs and creates a ZIP bundle with Markdown plus structured parsing outputs for downstream workflows.
The task is limited to 1 file and maximum file size of 20 MB.
Please upgrade to the Premium plan to be able to process up to 10 文件 per task and up to 100 MB of file size.