File Extractor

Unpack ZIP, TAR, TAR.GZ, and GZIP archives into their files and folders, then save the folder or download every file, all in your browser.

runs locally on your browser. Your data never leaves your device.

Drag & drop an archive here

ZIP, TAR, TAR.GZ, TGZ, or GZIP. The type is detected from the file's contents.

Common Use Cases

Open a ZIP of contracts or scans from a client without uploading it to a third-party converter site
Unpack a .tar.gz release or server backup on a Windows or Chromebook machine that has no tar command
Check what a downloaded ZIP really contains, including the paths it would extract to, before opening anything
Restore a whole project folder tree from a TGZ straight into a directory on your disk with Save to a folder

About File Extractor

File Extractor opens an archive and shows you what is inside: every file with its folder path and size, ready to download one at a time, or save as a real folder on your disk. It reads ZIP, TAR, TAR.GZ (also written TGZ), and plain GZIP. You do not have to say which one you have. The tool looks at the first bytes of the file, where each format leaves a fixed signature, so a ZIP renamed to .dat, or a .tar.gz that lost its extension in an email, still opens correctly.

Most online unzip sites upload your archive to a server, unpack it there, and send the files back. File Extractor never does. ZIP files are read with the JSZip library running inside your tab, TAR files are parsed by a small reader that walks the 512-byte header blocks itself, and gzip streams are inflated with the DecompressionStream API built into current browsers. None of that opens a network connection, so a backup full of contracts, a client deliverable, or a folder of unreleased designs stays on your machine from the moment you drop the archive in.

Browsers cannot hand you a folder as a download, which is why the tool offers three ways to get the result. In Chrome and Edge, Save to a folder uses the File System Access API to write the real directory tree to a place you choose, inside a fresh sub-folder so nothing already on your disk is overwritten. In every browser, Download all saves every file separately (without the folder structure), and each file in the list has its own download button. A TAR.GZ is unpacked in one step: the gzip layer is removed first, and because the result is a TAR, its entries are listed as a normal folder tree. A plain .gz holds a single file, so you get that one file, named the way the gunzip command would name it.

Archives are untrusted input, so the extractor defends against the usual tricks. Paths that start with a slash, carry a drive letter, or climb out with .. segments are moved inside the output folder, and the tool tells you how many entries needed that. Names that differ only in capital letters, which would overwrite each other on Windows and macOS, get a (1) suffix instead. Total unpacked size is capped at 1 GB and checked while data streams out, so a small file that expands into gigabytes, a zip bomb, is stopped early rather than freezing the tab. Password-protected ZIPs, corrupt archives, and formats such as RAR, 7z, XZ, and BZIP2 produce a plain explanation instead of a blank result.

Frequently Asked Questions

Is my archive uploaded anywhere to be extracted?
No. The archive is read and unpacked in your browser tab. ZIP files are opened with the JSZip library, TAR files by a short piece of local code, and gzip data by the DecompressionStream API that ships inside Chrome, Edge, Firefox, and Safari. No request carrying your file, its name, or its contents is ever made, which you can confirm by watching the Network tab in your browser's developer tools while you extract something.
Why can't I just download the extracted folder?
A web page can only trigger downloads of single files, and browsers have no folder download at all. That is why this tool gives you alternatives. Save to a folder, available in Chrome and Edge, writes the real directory tree to a location you pick. Download all saves every file one after another, without folders, and your browser may ask permission to allow multiple downloads. Every file in the list can also be downloaded individually. Firefox and Safari do not offer the folder picker yet, so they show the Download all and per-file options.
Which archive formats can it open, and which can't it?
It opens ZIP (including files from Office, Java, and Android, which are ZIPs inside), TAR, TAR.GZ or TGZ, and single-file GZIP. It cannot open RAR, 7z, XZ, BZIP2, or Zstandard, because those need decoders that browsers do not include, and the tool names the format when it recognizes one of them. Password-protected ZIP files are also not supported. For those formats, a desktop program such as 7-Zip is the right tool.
What does the tool do with a .tar.gz or .tgz file?
It removes the gzip layer first, checks whether the result is a TAR archive by validating its header checksum, and then lists the TAR's files and folders. You do not need to unpack twice, and you do not need to rename anything. If the decompressed data is not a TAR, as with a gzipped log or SQL dump, the result is a single file named after the archive without its .gz ending.
Is it safe to open an archive from someone I don't fully trust?
Extraction is safer here than with many desktop tools, though the files inside are still yours to treat with care. Nothing in the archive runs. Paths that try to escape the output folder with a leading slash, a drive letter, or .. segments are moved inside it, and symbolic links are skipped. The total unpacked size is limited to 1 GB and enforced while data streams out, so a decompression bomb is cancelled early. Scan any executable you extract before you open it, as you would anywhere.
Is there a size limit?
The unpacked contents are capped at 1 GB, and a ZIP file itself can be up to 1 GB because the ZIP library reads it fully into memory. TAR and gzip input is read as a stream, so those handle large files more comfortably. Everything runs inside one browser tab, so very large archives can be slow or fail on phones and older computers. The tool shows a message when a limit is hit, and extracts nothing rather than half of it.