Images
Pixel data is never decoded, redrawn or re-encoded.
- PNG — removes eXIf, tEXt, zTXt, iTXt, caBX and any c2* chunk; keeps IHDR, PLTE, IDAT, IEND and colour chunks.
- JPEG — removes APP1 through APP15 and COM comments, including the APP11 JUMBF segment; keeps APP0 and the entropy-coded scan.
- WebP — removes EXIF, XMP and C2PA chunks and corrects the VP8X feature flags.
- AVIF and HEIC — replaces C2PA, XMP and metadata uuid boxes with equal-size free boxes so media offsets stay valid.
- BMP — removes trailing bytes after the pixel payload and rewrites the file-size field.
- GIF — removes comment, XMP and application extensions; keeps the NETSCAPE looping and ICC extensions.
- TIFF and BigTIFF — removes descriptive tags, the EXIF sub-IFD and the GPS sub-IFD in place, keeping every structural offset intact.
Audio and video
Media samples are never re-encoded. Files above 8 MB are walked header by header through file slices, so a large video never has to be held in memory.
- MP4, MOV, M4A and M4V — removes C2PA and XMP boxes and the moov/udta user-data box.
- WAV — removes C2PA, LIST INFO and id3 chunks and rewrites the RIFF size.
- MP3 — removes the ID3v2 tag, including a truncated one, by locating the first valid audio frame.
- FLAC — removes the ID3v2 carrier used for C2PA manifests; native FLAC comment blocks are left alone.
Text and markup
TXT, MD, JSON, CSV, TSV, HTML, SVG and XML are read as UTF-8 and cleaned with the invisible-character engine. Structure and markup are preserved; only hidden code points change.
Limits and what is not supported
A single file is capped at 64 MiB by the engine. PDF, DOCX, ODT, EPUB and RAW camera formats are not supported, and no partial handling is offered for them.
A file whose container ends mid-block is reported as incompletely inspected. The tail is preserved rather than dropped, and the result is never described as verified.
Frequently asked questions
Can I remove video metadata for free without uploading the video?
Yes, for supported fields in MP4, MOV, M4A and M4V files up to 64 MiB. Processing runs in your browser and does not re-encode the media. Not every possible metadata field is supported. The free daily cleaning allowance still applies.
Does the MP3 cleaner remove every music tag?
No. The current MP3 cleaner removes ID3v2 tags while preserving audio frames. It is not a full tag editor and does not promise to remove every tag format. Preview the downloaded copy in your player and keep the original.