Know the document.
Choose the right extractor.
Selectable text or scanned image?
A text PDF already contains a text layer; a scan requires OCR. The downloadable local extractor handles text and simple tables with supplied column boundaries. It flags pages without a text layer. Compare 12 extraction candidates: local libraries, managed OCR, layout/table extraction and document fields. Cloud services require your own account; local tools require dependencies and suitable hardware. Their general accuracy has not been tested here.
Match this task →Measure the actual result
Check extracted pages, missing text, cell values and layout. Check the output contract for the selected operation: JSON table cells, CSV and XLSX are different outputs. Complex layouts and merged cells require task-specific validation. Private documents stay on your machine when using the local example.
Read the dated assertions and failures · Download local extractor InstructionsKeep a useful personal toolbox
Find a tool, save its public setup page, and leave yourself a private note after trying it. Your toolbox stays on this browser.
Open my toolbox →