Will EXIF and colour profile data be preserved?

Your file moves through a short, well-defined pipeline: upload over HTTPS, write to a temporary location on our processing server, run the conversion, save the result, hand the download URL back to your browser, and then queue both the input and output for cleanup. The raw file is never put behind a public URL — only the converted output is fetchable, and only via a signed link tied to your session.

We do not look inside the file, do not scan it for content, do not fingerprint or hash it, and do not feed it to any training pipeline. Our infrastructure has no human-readable file browser exposed for the upload directory.

The retention policy is short by design:

• Input file: removed as soon as the conversion job finishes successfully. If a job errors out, the input is still removed within 24 hours by the cleanup task.
• Output file: stays in storage for 24 hours so you can re-download it from the same browser. After 24 hours it is permanently deleted by the nightly cleanup task.
• Job record: an internal database row that tracks status (queued / processing / done) is retained for 7 days for billing and rate-limit accounting, then pruned. The row contains a job ID, format, size, and tier — never the file contents.

If you want files removed sooner, just close the tab — there is no user action that keeps them alive longer than 24 hours.

• No permanent copy. Your original file is never archived. Once the 24-hour window passes, it is gone from our storage.
• No content scanning. We don't run OCR, face detection, or content moderation over your files. The only thing the server looks at is the file type and dimensions, so it can pick the right encoder.
• No ML training. Your files are not used to train any machine learning model.
• No third-party sharing. The only systems that touch the file are our conversion workers and (for paid users opting in) our Wasabi S3-compatible storage.

For workflows where files cannot leave the browser at all, our photo editor, aspect ratio calculator, image metadata viewer and color palette extractor all run 100 % client-side — nothing is uploaded to our servers at all. Read the full privacy policy for the exact data we collect and the legal basis for processing.