Image to text (OCR)
Extract text from photos and scans.
Runs in your browserFree — no signup
How to use Image to text (OCR)
Image to text (OCR) reads printed text out of photos, screenshots and scans and returns it as editable text — running entirely on your device, so even sensitive documents stay private.
Step by step
- 1Choose the language of the text in your image.
- 2Drop the image (a sharp, straight, high-contrast photo gives the best result).
- 3Wait while the language model loads and reads the page — progress is shown live.
- 4Copy the text or download it as a .txt file.
What to expect
- Accuracy of roughly 95–99% on clean, straight, high-contrast scans.
- Lower accuracy on handwriting, low light, heavy skew or decorative fonts.
- The first run downloads the language model (a few MB); later runs are much faster.
- You can pass the output straight into AI OCR cleanup to repair broken words.
Common use cases
- Turning a photographed page, receipt or whiteboard into editable text.
- Making a scanned document searchable.
- Pulling a quote or table out of a screenshot without retyping it.
- Digitising forms, invoices or IDs for record keeping.
Privacy & data
Everything runs locally in your browser. Your file is never uploaded to a server, so nothing leaves your device.