Nothing goes to a server
Your browser downloads the site once, like any web page. After that, each image you pick stays inside your browser and is read there.
OCR without upload
Your image stays on your device
Add up to 10 images or 100 MB. They are read one after another inside your browser, and you get editable text back. No image or result is sent anywhere.
✓ No uploads✓ English + Chinese✓ Works offline after first use
The page is locked down
A browser security rule (a Content Security Policy) only lets this page talk to its own site, for its files, its models and anonymous use counts, plus Cloudflare’s page-visit counter. There are no ad, account, upload, or OCR addresses for it to reach.
One image open at a time
The list keeps small previews and your results, not every full-size image, so memory stays free.
Works offline after first use
Fast works offline once your browser has saved its files. High accuracy works offline only after you have downloaded it too.
Common questions
Frequently asked questions
Where is OCR processed?
Inside your browser (using ONNX Runtime Web). The site has no address that accepts images or the text read from them.
Can the site see my text?
No. Nothing is ever read on a server, not even as a backup. The only things counted, anonymously, are page visits and three steps (started, got a result, copied or downloaded it), never images or text.
What are the batch limits?
Up to 10 images, 50 MB per image, and 100 MB in total. Images are opened and read one at a time.
What is downloaded?
The two Fast models (one finds the text, one reads it) download when you start. The 21 MB High accuracy model downloads only after you confirm.
What actually crosses the network
Checked: · Chrome 152.0.7977.76
The browser downloads the page, JavaScript, dictionary, ONNX models and WASM runtime from this site. OCR runs in a worker. The local test observed GET requests for these assets and POST requests to /api/demand-event for anonymous task steps. The production check on this date found no Cloudflare page-analytics beacon; the CSP allows it if enabled. Zero upload does not mean zero network activity.
The task endpoint accepts only task_started, task_completed and result_exported with a page path and an enumerated coarse source. It does not accept image bytes, filenames or recognized text. Use your browser Network panel to inspect request payloads while trying the synthetic image.
/models/ppocrv6-tiny-det.onnx+/models/ppocrv6-tiny-rec.onnx(Fast)/models/ppocrv6-small-rec.onnx(High accuracy)/ort/ort-wasm-simd-threaded.wasm
When offline OCR works
This sample was processed again after the models and page assets were cached, the browser went offline and the same page reloaded. The recorded offline output appears below. First use needs a connection. High accuracy needs its optional model downloaded first. Other pages, formats and PDF assets must be opened or used online before relying on them offline. Clearing site data, private browsing or storage eviction can remove the cache.
Observed offline result: 烯OCR纬摧PRIVATE