Skip to content

PDF-MD

Convert PDFs into clean, LLM-ready Markdown text. Available at chat-lab.ai/pdf-md.

How it works

Your PDF is base64-encoded and sent to the Jina Reader API (r.jina.ai), which performs layout-aware extraction (including OCR where needed) and returns the document as Markdown. The request timeout scales with file size, and transient network failures are retried automatically.

No language model is called by this tool; Jina Reader is a purpose-built extraction service. The output is designed to be pasted directly into an AI chat or pipeline.

User options

  • Single PDF tab: one file up to 50 MB; results appear in an editable text box and download as a .md file.
  • Batch tab: up to 15 PDFs, each up to 50 MB. Files are processed sequentially with rate-limit pacing, then delivered as a ZIP of .md files. Any failures are listed with their reason and downloadable as a .txt report.

Rate limits

Requests are paced automatically to respect Jina Reader's rate limits (the app uses an authenticated key for higher throughput). Large batches simply take proportionally longer; progress is shown per file.

Data handling

PDFs are processed in memory and sent to Jina Reader for extraction; the app stores nothing. Extracted text exists only in your browser session. See Privacy & Data Retention.