Get Started

Pull the tables out of any PDF into Excel, with free AI that runs on your laptop

Docling, the free document reader started at IBM Research, finds the tables in a PDF (even a scanned one) and rebuilds them as a spreadsheet. It runs on your own machine, so bank statements and invoices never get uploaded anywhere.

36 minBeginnerTools: FreeStep 1 free
Tested 9 Oct 2026 · Docling 2.135.0, RapidOCR 3.9.2, Python 3.12 via uv 0.11.9, Windows 11 laptop with 16GB RAM and no graphics card
Docling installeda free tool that finds the tables in a PDF and rebuilds them, rows, columns and all
A copy-paste script that turns every table in a PDF into an Excel file, with tables that run over several pages joined back into one sheet
Scanned PDFs read with OCR (text recognition), all on your own laptop
  • Cell-by-cell accuracy figures on real and sample PDFs, so you know exactly what to spot-check
  • A final project: your bank statement sorted into spending categories by a local AI, with totals in Excel
Before you start
01/06Install Docling0 done
STEP 01 of 06~5 min

Install Docling

One command installs Docling and the text-recognition engine it needs for scanned pages.

  1. Docling was started by the AI for knowledge team at IBM Research Zurich and is now looked after by the Linux Foundation's LF AI & Data group. It's free and open source (MIT licence ↗). Its trick is that it doesn't just copy the text out of a PDF: small AI models look at each page, spot where the tables are, and work out which words sit in which row and column.

  2. Open PowerShell. If you don't have uv yet, install it first, then close and reopen PowerShell:

    Windows (only if you don't have uv)
    winget install --id astral-sh.uv -e
  3. Now install Docling. Copy the whole line:

    Terminal
    uv tool install "docling[rapidocr]" --with "rapidocr<3.10" --python 3.12
  4. [rapidocr] adds the engine that reads scanned pages. --with "rapidocr<3.10" holds that engine at 3.9: version 3.10.0 breaks Docling's text recognition, and scans come out empty. --python 3.12 tells uv which Python to use, and uv downloads it if you don't have it.

  5. It downloads about 350MB and takes 1.2GB on disk. The biggest part (about 480MB) is PyTorch, the library its AI models run on. Allow a few minutes on a normal connection. Then check it:

    Terminal
    docling --version
    Output
    Docling version: 2.135.0
    Docling Core version: 2.101.1
    Docling IBM Models version: 4.0.3
    Docling Parse version: 7.22.2
    Python: cpython-312 (3.12.13)
    Platform: Windows-11-10.0.26200-SP0
✓
Checkpointdocling --version prints a version number.
If it breaks'docling' is not recognised + 1 more
'docling' is not recognised
uv puts its apps in a folder your terminal doesn't know about yet. Run uv tool update-shell, close PowerShell, open a new one and try again.
'uv' is not recognised
Close every PowerShell window after installing uv and open a new one. New windows pick up newly installed apps; old ones don't.
Still stuck? Ask the community
Next mission · Own your AICancel Copilot: a free, private coding assistant in VS Code on your own laptop
→