Last updated

PDF to Text

Every page read, not just the first. No account, no email.

No file yetReady

Drop a PDF to read it

PDF — up to 3 MB. Your file is read on our servers in Frankfurt and deleted within the hour.

5 files an hour3 MB a file3 pages a filedeleted after 1h

VisionParse extracts text from scanned PDFs free in your browser, reading every page rather than only the first.

Each page is rasterised at 150 DPI and recognised separately, then joined in page order. A five-page PDF completed in 3.64 seconds in our 2026-08-01 measurement. Uploaded files are deleted one hour after conversion.

Last updated: 2026-08-01 · Measured by VisionParse engineering

What this tool does, in numbers.

  • SPEED

    VisionParse read a five-page PDF in 3.64 seconds, including 1.03 seconds of rasterisation.

  • ACCURACY

    VisionParse rasterises PDF pages at 150 DPI, the lowest resolution that held 100% accuracy.

  • LIMIT

    The free PDF tool reads the first 3 pages; an account raises this to 200 pages per document.

  • LANGUAGES

    VisionParse recognises 161 languages inside PDFs, including Arabic, Sinhala and Tamil.

  • ACCESS

    The PDF to text tool requires no account, no email address and no payment card.

  • RETENTION

    VisionParse deletes free-tool PDFs one hour after conversion and never trains models on them.

Where the 3.64 s comes from.

Every figure on this page traces to a dated run on hardware we name. Nothing here is a vendor estimate.

Measured 2026-08-01

Claim
A five-page scanned PDF is rasterised and read in 3.64 seconds — 1.03 s rasterising, 2.61 s recognising.
Method
VisionParse standard recognition tier, single-threaded. Full methodology.
Sample
Five-page PDF built from the synthetic invoice, rasterised at 150 DPI and read page by page. n = 1 document, 5 pages.
Measured
2026-08-01 on Contabo Cloud VPS 8 — 8 vCPU AMD EPYC, 24 GB RAM, Ubuntu 24.04.4
Limitations
Timing scales with page count and page density; a full A4 text page takes longer than this invoice. A single synthetic document per variant is a small sample and is not a substitute for a corpus. These figures describe format behaviour, not general accuracy. A 500-document corpus run is scheduled and will supersede them.
Contact
VisionParse@senithu.lk

What this tool gets wrong

DISQUALIFIED

Single-image rasterisation reads page one and stops. We found this in our own pipeline on 2026-08-01: a five-page PDF returned one page of text with no error. Silent truncation is worse than a failure because the caller cannot tell. VisionParse now rasterises every page explicitly and reports the page count it actually read.

KNOWN LIMIT

PDFs that already contain text are still re-read optically. A born-digital PDF has a text layer that could be copied directly and perfectly. VisionParse currently recognises it as an image instead, which is slower and can introduce errors a direct copy would not.

HELD UP

Page order survives past page nine. Pages are sorted numerically rather than by filename, so page 10 follows page 9 instead of landing between 1 and 2 — a common failure in naive batch OCR.

Three steps, no account.

  1. Drop your PDF

    Drag a PDF onto the workbench, or choose one. Nothing leaves your device until you drop it.

  2. Every page is rasterised

    Each page is rendered at 150 DPI and recognised separately, so nothing is skipped and page order is preserved.

  3. Take the text

    Text appears page by page with a confidence score. Copy it or download a .txt; the file is deleted within the hour.

What this tool accepts.

PropertyFree tool
Input formatPDF, scanned or born-digital
OutputPlain text, UTF-8, joined in page order
Rasterisation150 DPI per page
Maximum pages, free tool3
Maximum pages, account200 per document
Maximum file size3 MB
Rate limit5 files per hour, per IP address
Languages161
File retentionDeleted 1 hour after conversion
Processing locationFrankfurt, Germany

When to use the API instead.

This page converts one file now. To convert files from your own software, the same engine runs over HTTP from Starter upward.

# same engine, from your code
curl -X POST https://api.visionparse.app/v1/ocr \
  -H "x-api-key: $VISIONPARSE_KEY" \
  -F "file=@image.jpg"

→ 202 { "jobId": "c0df8e2c…" }

Use the API when

You convert more than a handful of files, convert on a schedule, or need results inside another system. Batch, webhooks and bounding boxes are API-only.

Frequently asked

Does this read every page of my PDF?

Yes. Every page is rasterised and recognised separately, and the result reports how many pages were read. The free tool covers the first 3 pages; an account raises that to 200 pages per document.

How long does a multi-page PDF take?

A five-page PDF took 3.64 seconds in our 2026-08-01 measurement — 1.03 seconds to rasterise and 2.61 seconds to recognise. Longer or denser documents take proportionally longer.

Does it work on scanned PDFs?

Yes, that is the main case. A scanned PDF is a picture of a page, so it needs optical recognition rather than a text copy. Born-digital PDFs also work.

Can I get a searchable PDF back instead of text?

Use the scanned PDF to searchable PDF tool, which keeps the original page images and adds an invisible text layer underneath.

Do you keep my PDF?

Free-tool files are deleted one hour after conversion. Account uploads are deleted after 30 days, and no customer file is ever used to train models.

Measured by VisionParse engineering. Method and raw data are published.Updated Benchmark · Methodology · Data handling