Extract a tar.gz online

How to extract a tar.gz in your browser

No terminal, no flags, no remembering whether it is -xzf or -xvjf. The browser does the ungzip and the untar in one local step.

• Add the archive. Drag the .tar.gz (or .tgz) onto the page, pick it from disk, or paste a direct URL. The bytes are read locally — no upload.
• Browse the tree. The contents render as a folder tree, with Unix file paths preserved. Expand folders, search by name across the whole archive, and click any entry to preview it first.
• Extract. Use the extract-all control in the file-tree toolbar. On Chrome and Edge, pick a destination folder and the original files are written there with their structure; on Firefox and Safari, the contents download as a clean zip.
• Or download single files. Preview an entry and save just that file.

tar.gz, .tar, and .gz — what each one is

The three extensions get conflated, but they mean different things, and ZipTool handles each honestly.

A `.tar.gz` (or .tgz) is a tar archive compressed with gzip — the common case: many files bundled and then shrunk. A `.tar` is the same bundle with no compression — larger, but sometimes what you receive. A `.gz` is a single file compressed with gzip, not a multi-file container at all — for example backup.sql.gz is one SQL file. ZipTool opens all three: a .tar.gz and .tar render as a folder tree, and a single-file .gz opens as the one decompressed file inside. (Gzip, xz, bzip2, and zstd are single-stream compressors; only when wrapped around tar do they carry multiple files.) For the format background, see what is a tar.gz file and the tarball glossary entry.

Extraction versus previewing a tarball

Previewing means looking inside without writing to disk — ZipTool renders the tree and shows you files one at a time. Extracting decompresses and writes the entries out as real files. For a source bundle this matters: source tarballs are mostly text (code, configs, READMEs, package.json, Cargo.toml), so you can read almost everything inline, decide whether the package is what you want, and only then unpack it.

Because ZipTool previews code in a syntax-highlighting editor, a tar.gz of source code is readable straight out of the archive — useful for auditing a dependency before you build it. For that workflow in depth, see viewing the contents of a zip without extracting, which covers the same browse-before-unpack idea.

Why a no-upload untar matters

tar.gz files are usually source code, configs, or data — exactly the content you may not want on a stranger's server. The upload-based untar sites ask you to send the whole archive to their backend, unpack it there, and trust them with the result. A client-side extractor removes that step: the tarball is decompressed in your tab and written to a folder you chose on your own machine. No complete copy exists anywhere else.

Verify it yourself: open a tar.gz, then watch DevTools' Network tab — you see the page load, but no upload of your archive. Disconnect from the internet and confirm browsing, previewing, and extracting keep working. This is the same privacy model behind unzipping online without uploading and the private zip viewer.

Honest limits

A browser-based untar is genuinely useful, with one format boundary worth stating clearly.

• `.tar` and `.tar.gz` are containers; a lone `.gz` is not. A single-file .gz opens as the one decompressed file inside it (correct), but it is not a folder and never will be — gzip is a stream compressor, not a packager. If you expected many files from a .gz, you likely have a mistyped .tar.gz.
• Other compressors are read-only here. ZipTool decodes gzip streams natively. Archives compressed with xz (.tar.xz), bzip2 (.tar.bz2), zstd (.tar.zst), or LZ4 are not in scope for this page; use the command line or a desktop tool for those.
• Extract-to-folder needs Chrome or Edge. Writing a real directory tree uses the File System Access API, which Firefox and Safari lack. On those browsers you still get every file, bundled as a downloadable zip that preserves the structure.
• Large tarballs use memory. A tarball is a single stream, so unpacking a very large one can be memory-heavy in a browser tab. Open archives from sources you trust, and be cautious with files that expand to many times their compressed size.
• Unix permissions are not restored on extraction. tar records file permissions and timestamps; the browser extractor writes the file contents and folder structure but does not faithfully replay Unix mode bits the way tar -xpf would.

Frequently asked questions