How we keep your files private
ConvertWarp converts files inside your browser tab. The file you pick is not sent to us. This page explains how that works and how you can check it yourself in a couple of minutes.
The short version
- Your file is read by your browser and processed by code running in the open tab.
- ConvertWarp has no upload endpoint and no conversion server. There is nowhere for the file to go.
- The result is created in your browser’s memory and saved by your browser when you click Download.
- Closing the tab discards the in-memory copy. There is no server copy to delete, because none was made.
- You do not need an account, so there is no profile linking you to the files you convert.
What happens when you choose a file
When you drop a file on a tool or pick one with the file chooser, the browser gives the page a handle to that file (the standard File API). The tool reads the bytes into memory in the same tab. Nothing is posted over the network.
The conversion is done by open-source libraries that ship with the page: pdf-lib for merging, splitting and rewriting PDFs, PDF.js for reading and drawing PDF pages, the browser’s own image decoders and canvas for JPEG, PNG and WebP, a WebAssembly HEIC decoder when the browser cannot open HEIC natively, UTIF.js for classic TIFF, docx and ExcelJS for Office files, fflate for ZIP files, ffmpeg.wasm for video and audio, and @imgly/background-removal for the background tool.
The finished file is held as a Blob in the tab, and the Download button hands it to your browser’s normal download flow. Previews are drawn from that same in-memory copy. They are not sent through an image service.
What the network is still used for
The page itself has to load. Your browser downloads HTML, JavaScript, styles, fonts and codec files so the tool can run. Some of that code loads the first time you use a tool, not when the page opens, so you will see a few requests at that moment. They are downloads of code, made with GET requests, and they are the same for every visitor.
- Site code and styles come from convertwarp.com (paths such as /_next/static/…).
- PDF tools load the PDF.js worker and its helper WebAssembly from /pdfjs/ on this site.
- Video and audio tools load the ffmpeg.wasm core from /ffmpeg/ on this site. Only if that download fails do they fall back to a public package CDN (cdn.jsdelivr.net, then unpkg.com). Either request fetches the codec; it never includes your file.
- The BG remover loads its segmentation model (about 84 MB) and runtime from /bg-removal/ on this site the first time you use it. That GET is the model, not your photo.
- When advertising is active, Google’s ad scripts load from Google’s domains. They see that you visited a page, not what is in your file.
Check it yourself: watch the Network tab
You do not have to take our word for it. Browsers include developer tools that list every request a page makes. If a file were uploaded, you would see a request roughly as large as the file.
- Open any ConvertWarp tool, for example Compress image or Merge PDF.
- Open your browser’s developer tools: press F12 (or Ctrl+Shift+I) in Chrome, Edge and Firefox, or Cmd+Option+I on a Mac. In Safari, turn on Show features for web developers in Settings → Advanced first.
- Select the Network tab, then click the clear button so the list is empty.
- Choose a file in the tool and run the conversion.
- Look at the new rows. Sort by Size and check the Method column: a file upload would appear as a POST or PUT request about as large as your file.
What you should see: a handful of small GET requests for scripts or WebAssembly the first time a tool runs, and then nothing while it converts. You should not see a POST or PUT request, and no request should carry a body the size of your file.
Check it yourself: convert with Wi-Fi off
A server-based converter cannot work without a connection. ConvertWarp can, once the tool’s code is in the browser.
- Open the tool you want to test and convert one small file, so the browser has downloaded the tool’s code.
- Turn off Wi-Fi or unplug the network cable. In developer tools you can also set the Network tab’s throttling menu to Offline.
- Choose another file and convert it.
- The conversion finishes and the download works, because the work happens on your device.
The first conversion is there to make the browser fetch any code that loads on demand, such as the HEIC decoder, PDF.js, or the video codec. If you go offline before that, the tool may report that it could not load, which is the network being off, not your file being sent anywhere.
What we collect, and what we do not
This summary matches the Privacy Policy, which is the authoritative version.
We do not collect:
- The files you convert, or any part of their contents.
- File names, page text, photo metadata (EXIF, location), or the results you download.
- Account details. There are no accounts, logins or user profiles.
- A first-party cookie to run the converters. The tools work without one.
What does exist:
- Ordinary hosting logs at our host (Cloudflare): IP address, browser user agent, referrer, the URL requested and the time. They describe a page visit, not a file.
- Emails you choose to send us, kept only as long as needed to reply.
- When advertising is shown (Google AdSense), Google and its partners may set cookies about the visit. A consent notice is shown where the law requires it.
- Google Analytics counts page visits after you accept the notice. It does not receive the file you converted.
Honest limits
- Browser extensions you have installed can read pages you open, including this one. That is outside our control. A private window with extensions disabled avoids it.
- Your operating system may page memory to disk during a large conversion. That stays on your own device.
- A file you download is yours to keep or delete. We cannot see or remove it.
- Website code can change. That is why the two checks above matter: they test what the page does today, on your machine.
- If a future feature ever needed a server, the Privacy Policy would be updated before it shipped, and the tool would say so on the page.
How the site is delivered
ConvertWarp is a static website served over HTTPS by Cloudflare. There is no application server behind it and no database of conversions. Responses include security headers such as HTTP Strict Transport Security (HTTPS only), a no-sniff content-type policy, a strict referrer policy, a permissions policy that turns off the camera, microphone and location, and a frame policy that stops other sites from embedding the full site. Only the widget pages under /embed/ may be framed, so other websites can offer the same in-browser tools.