Compared · checked 2026-10-24
The best free OCR tools for images and scans, compared
OCR turns a photo, screenshot or scanned PDF into text you can copy and edit. The free options differ a lot in languages, limits and where your scan goes. Here is what each says about itself.
We make OCR, one of the tools on this page. Every fact about the other tools comes from their own pages, linked under each one, as they read on 2026-10-24. Prices and features change: check each site before you choose.
How we compared them
- Languages it can read
- Free limits: files, pages and size
- Outputs: text, Word, searchable PDF
- Where the scan is processed: your device or their servers
- Whether you need an account or an install
At a glance
| Tool | Price | Free limits | Languages | Files processed on | Account or install |
|---|---|---|---|---|---|
| OCR ours | Free | None set by us | 85, one model per script (Latin, Cyrillic, Arabic, Chinese, Japanese, Korean, Hindi and more) | Your device | No account; a 13 to 26 MB model per script downloads once |
| Google Drive | Included in Google Drive | Files of 2 MB or smaller | Many; detected automatically | Google Drive | Google account |
| OnlineOCR.net | Free for guests; paid plans | 5 files an hour as a guest, 15 MB per file | 46 | OnlineOCR servers (guest files deleted after conversion) | None for guests |
| i2OCR | Free; premium bulk OCR | One image or PDF page at a time | 128 | i2OCR servers (files deleted after processing) | No install; account not stated |
| Tesseract | Free, open source (Apache 2.0) | None | 100+ | Your computer | Install; command line |
| Feature | OCR ours | Google Drive | OnlineOCR.net | i2OCR | Tesseract |
|---|---|---|---|---|---|
| Photos and screenshots | Yes | Yes | Yes | Yes | Yes |
| Scanned PDFs | Yes | Yes | Yes | Yes | No |
| Many pages in one go | Yes | Yes | Partly | Partly | Yes |
| Languages | 85 | Many | 46 | 128 | 100+ |
| Plain text | Yes | Yes | Yes | Yes | Yes |
| Word (DOCX) | Yes | Not stated | Yes | Yes | No |
| Searchable PDF | Yes | Not stated | Not stated | Yes | Yes |
| Fix the text before saving | Yes | Yes | Not stated | Not stated | No |
| Boxes on the page for each line | Yes | Not stated | Not stated | Not stated | Not stated |
| Free file size | Device memory | 2 MB | 15 MB | Not stated | Not stated |
| Free use without a cap | Yes | Not stated | No | Partly | Yes |
| No account needed | Yes | No | Yes | Not stated | Yes |
| Nothing to install | Yes | Yes | Yes | Yes | No |
| Scan stays on your device | Yes | No | No | No | Yes |
1. OCR (ours)
Best for: Private scans in 85 languages, many pages at once
Reads text from photos, screenshots and every page of a scanned PDF in the browser, shows a box around each line, and lets you fix the text before saving.
Good
- Nothing is uploaded: the scan is read on your device
- Every page of a PDF, read at 200 dpi
- Edit the text before you save it
- Save as text, Markdown, Word or a searchable PDF, or copy
- Works on your phone, tablet and computer in the browser, even offline
Not so good
- Fewer languages than i2OCR or Tesseract: 85, and you choose the language (or it suggests one after a first read)
- Needs a one-time model download (13 to 26 MB for each script) before the first read
- The searchable PDF's invisible text keeps Latin letters only
- Slower on old phones and computers than a server
2. Google Drive
Best for: People who already use Google Drive and Docs
Good
- Open an image or PDF with Google Docs and the text appears in a document
- Detects the language itself, with a long list of languages
- Keeps bold, italics, font size and line breaks in many cases
Not so good
- Files should be 2 MB or smaller
- Lists, tables, columns and footnotes are not likely to be detected
- The file is uploaded to your Google Drive
Google Drive website · Sources: Convert images to text (Google Drive Help)
3. OnlineOCR.net
Best for: Getting Word or Excel files out of a few scans
Good
- 46 recognition languages
- Outputs Word, Excel, PDF, RTF and plain text
- No registration for guest use
Not so good
- Guests can convert 5 files an hour
- Files are uploaded to the service
- Registered users' output files are stored for a month
OnlineOCR.net website · Sources: OnlineOCR.net
4. i2OCR
Best for: Rare languages, one page at a time
Good
- 128 languages
- Download as text, Word, HTML or searchable PDF
- Single or multi-column layout setting
Not so good
- Free OCR handles one image or page at a time
- Bulk OCR needs a premium plan
- Files are uploaded for processing
i2OCR website · Sources: i2OCR
5. Tesseract
Best for: Developers and batch jobs on your own computer, in 100+ languages
Good
- Recognises more than 100 languages out of the box
- Outputs plain text, hOCR, PDF, invisible-text PDF, TSV, ALTO and PAGE
- Runs on your own computer; can be trained for other languages
Not so good
- No graphical app: it is a command-line program and library
- Opens images, not PDF documents
- Better results often need you to clean up the image first
Tesseract website · Sources: Tesseract README
Which one to choose
For private documents in its 85 languages (Swedish, German, Polish, Russian, Arabic, Hindi, Chinese, Japanese, Korean and more), OCR fits our criteria best: no account, no cap, many pages at once, and the scan never leaves your device. For a language outside that list it is the wrong tool.
For rarer languages, i2OCR (128) and OnlineOCR.net (46) work in the browser but upload your file and limit free use. Google Drive is handy if you live in Google Docs, within 2 MB per file. Tesseract is the choice for developers and large batches on your own machine, if you are happy with the command line.
Questions
What is the best free OCR that does not upload my scan?
OCR reads the text in your browser, so the image or PDF stays on your device. Tesseract also runs on your computer, as a command-line program. Google Drive, OnlineOCR.net and i2OCR process the file on their servers.
Which free OCR reads the most languages?
i2OCR lists 128 languages and Tesseract more than 100. OCR reads 85, with one model per script: Latin-script languages such as Swedish, German and Polish with their accents, Russian and other Cyrillic languages, Arabic, Persian and Urdu, Hindi, Chinese, Japanese, Korean, Greek, Thai, Tamil and Telugu. OnlineOCR.net lists 46.
Can I get a searchable PDF for free?
Yes. OCR, i2OCR and Tesseract can save a PDF with an invisible text layer, so you can search and copy the words.
Can OCR read a whole scanned PDF?
OCR reads every page of a PDF in one go. Google Drive opens multipage PDFs, while i2OCR's free plan does one page at a time and OnlineOCR.net allows 5 files an hour for guests.
Names of other products are trademarks of their owners. ITpris Verktyg is not affiliated with them.