No. Everything runs locally in your browser - there is no upload, no account needed and nothing stored on a server. Your file never leaves your device.

You mask the logo by hand with Box, Brush and Eraser, and a neural inpainting model reconstructs the masked area from the surrounding image - the same kind of AI fill behind tools like Photoshop's Remove. It runs entirely in your browser, and works best where the background is consistent; a truly unique object hidden under the logo can only be approximated.

Fast uses a small model (28 MB) that is already loaded and returns quickly - good for most images. Best uses a larger, stronger model (LaMa, 208 MB) that reconstructs real photos more cleanly, especially smooth backgrounds. Best downloads once on first use and is then cached in your browser, so later runs skip the download.

Paint slightly beyond the watermark rather than hugging its edge - the anti-aliased fringe is part of the logo, and leaving it behind is what produces a faint ghost. Keep the mask tight otherwise: the smaller the region, the less has to be reconstructed. If Fast leaves a soft patch on a detailed photo, switch to Best.

Video is handled in the Video tab, which masks the logo once and repaints it across every frame while carrying the original audio through untouched.

No. Invisible watermarking such as Google's SynthID is embedded into the pixels themselves and survives cropping, compression and editing. Removing a visible logo does not touch it, and no browser tool can. Metadata such as C2PA Content Credentials is separate again and lives in the file structure - the Metadata Remover strips that.

Use it on images you created or have the rights to. This is an editor for your own files, not a way to strip attribution from someone else's work.