PDF to Word

Reconstruct native PDF text as editable DOCX paragraphs, lists and confirmed tables, or use positioned text frames with raster graphics. Review reading order and correct text before export.

Processed on your deviceFree · No account needed

Up to 5 PDFs, 10 MiB each; select at most 20 pages across the batch. Native PDF text is read first. No Office engine download is needed for this export.

PDF stores positioned marks, not the original document structure. Reading order, table boundaries, fonts and page layout are reconstructed and require review. Lists and confirmed tables become editable objects. Positioned mode retains a raster graphics layer; Word reading flow includes detected raster images only.

PDF files

Output mode

Reading flow includes detected raster images. Vector drawings are not carried into this mode. Choose positioned layout to retain the page graphics as raster images. This is a mode limitation, not a claim that this document contains vector drawings.

Scanned pages · optional OCRText language

Optional recognition loads about 8.2 MB of static OCR software. It starts only when you check this box and read the pages. Uncertain words must be corrected. OCR does not rebuild scanned graphics as editable objects.

Download OCR if needed and recognize pages without text

Preparing your tool…

How to use it

Select PDFs and page ranges. Read native text first; explicitly enable optional OCR only for pages without text.
Compare each source preview with extracted text. Correct words, review table rows and columns, and adjust reading order for a reading-flow document.
Choose reading flow or positioned layout. Confirm table candidates and font substitution; rotated text needs reading-flow mode.
Create real DOCX files and download individual documents or the ZIP with conversion reports. Reopen in Word or LibreOffice and check line breaks.

The two-page synthetic PDF includes two text columns, IDs 0012 and 0042, a three-column table and an image. Change an ID before export and check the actual Word text and table cells.

Details & limits

Up to 5 PDFs, 10 MiB each, at most 20 selected pages across the batch, 6,000 text fragments and 120,000 characters per document. Bounded PNG previews and 100 image regions. Encrypted, signed or active PDFs and unsupported PDF image filters are rejected by the shared PDF preflight. English/German scan OCR is an optional separate download. Editable text uses Liberation Sans; original fonts and document semantics are not recovered automatically. Word reading flow keeps detected raster images; vector drawings require positioned mode.

Is the Word file a page image with hidden text?

Reading-flow export creates native paragraphs, list items and confirmed tables; image regions remain images. Positioned export creates editable text frames over a graphics-only raster layer. Table values require a verified source cell grid and receive separate frames; uncertain table geometry requires reading flow. It approximates layout and does not recover original styles, footnotes or semantic tags. Scanned pages require reviewed OCR, and their original graphics are not retained in editable mode.

PDF to Word

How to use it

Details & limits

Related tools