More free image tools
Same quality bar. Still free. Still no upload.
Your files never leave your device
Free to use, with no account and no watermark. The first run downloads the model files once and your browser caches them; the model then runs on your device, and the file you process is never sent anywhere.
This is not a promise about how carefully we handle your upload. There is no upload. The file you choose is read into memory by your own browser, processed there, and handed back as a download — the tool has no server component and makes no network request with your data. The only download is the model itself, which contains nothing of yours.
- Nothing is stored. We never receive the file, so there is no copy to retain, no retention period to disclose, and nothing to delete on request.
- Nothing is transmitted. No upload endpoint, no analytics payload carrying file contents, no third-party processor.
- You can verify it. Run the tool once so the model is cached, then disconnect from the internet and run it again. It still works, because everything it needs is already on your machine.
- Nothing persists after you close the tab. Files are held in page memory only — not in local storage, not in a cache, not in a database.
A note on wording, because it matters: we do not describe these tools as “encrypted.” Encryption protects data that travels to somebody else's computer. Your file does not travel, so there is nothing to encrypt and nothing to intercept — which is a stronger guarantee than encryption, not a weaker one. The page itself is served over HTTPS like the rest of the site.
This makes the tools safe for material you are contractually barred from uploading to third-party services — client footage under NDA, unreleased campaign assets, anything covered by a confidentiality clause.
How it works
1. Add images
Drop JPG, PNG or WebP files, pick them, or paste a screenshot with Cmd/Ctrl+V. Up to 10 images, 40 MB each. Large photos are scaled to 1024 px on the long edge. The files stay in this tab.
2. Read the text
Florence-2 base (fine-tuned), Microsoft's model in the onnx-community ONNX build under the MIT licence, reads each image on your device. WebGPU when the browser has it, WebAssembly otherwise. The first run downloads the model once (about 360 MB on WebGPU, about 230 MB on WebAssembly) into your browser cache. Image to Prompt uses the same weights when it picks the same engine.
3. Check the lines
Each image gets an editable box. Line breaks follow the lines Florence returned; words on one line stay on that line. Show text regions to draw the boxes from a second pass. Printed text is the reliable case. Handwriting and dense tables are weaker, so read it before you copy.
4. Copy or download
Copy one image, copy the whole batch, or download a .txt. Nothing is uploaded and there is no watermark.
Questions
Is this image to text tool free?
Yes. No account and no watermark. The model runs on your device, so there is no server cost to pass on. The limits are practical: 10 images at a time, 40 MB each, and a one-off model download the first time.
Is my image uploaded anywhere?
No. The image is read inside this browser tab. The only network request is the first-run download of Florence-2 from Hugging Face, which your browser then caches. Image to Prompt downloads the same files when it uses the same engine.
How accurate is the OCR?
Clear printed type, slides and simple receipts usually come through well enough to edit. Handwriting, tiny type, skewed photos and dense tables are weaker: words get dropped and lines get merged. A full page of small type can also be cut off, because the model stops after a fixed number of tokens. Read the text before you rely on it.
Which model does it use, and can I use the text commercially?
Microsoft's Florence-2 base (fine-tuned), in the onnx-community ONNX build, released under the MIT licence, which allows commercial use. The text is a reading of your image. The licence does not grant rights to text you do not already have the right to copy.
Why is the first run slow?
The model downloads once: about 360 MB on browsers with WebGPU, about 230 MB on the WebAssembly fallback. After that it loads from the browser cache. Each image is two passes, the words and then the boxes: a few seconds on WebGPU, and longer, up to around a minute, on an older laptop. Phones without WebGPU use the slower path, so one or two images at a time is safer there.
Can it read a scanned PDF?
Not directly. This tool takes pictures: PNG, JPG or WebP. Export the page as an image first. If the PDF already has a text layer, PDF to text pulls that out without a vision model.
Related tools
Stay in the same job cluster. Free tools stay on-device; paid tools are labelled as such.
These rearrange files. Versely makes new ones.
Everything on this page works on a file you already have, which is why it costs nothing. Generating something that did not exist (video from a prompt, a voice, a score) runs a model, and that costs credits.