PDF to Markdown

Flare indexes your files for your AI agents, on your own machine. This is its PDF reader: drop a PDF, get it as Markdown, or each page's text and metadata as JSON.

Drop a PDF here or click to choose one
Up to 10 MiB; the first 30 pages are read. The file is deleted once it is read.
Scanned pages come back empty here: OCR runs in the app, not in this demo.

API: curl -F file=@doc.pdf https://flare-demo.fra.unikraft.app/api/extract (the JSON has the whole document as markdown)
Flare on GitHub · nicolas-ha.com