Create TAR archives.
Bundle files into a tar archive — built on your device.
Drop any files here
or browse your files
Paste with ⌘V · or drop files anywhere on the page
Files never leave your device. Everything runs in your browser, nothing touches a server — tools you've used even work offline.
A tar file glues many files into one stream without compressing them — the format unix tooling has expected since the tape-drive era. Drop files, download one .tar. Built entirely in your browser; combine it with gzip or xz here too if you want it compressed.
How it works
- Drop files anywhere on the page, click to browse, or paste with ⌘V.
- Pick a quality or preset — or set an exact target size and let the tool find it.
- Compress, compare before/after, and download — individually or as a ZIP.
tar, tar.gz, tgz — the family tree
tar bundles; gzip/bzip2/xz compress the bundle. tar.gz (or .tgz — same thing) is the everyday combination, and Create TAR.GZ builds it in one step. A plain tar from this page can also be compressed later with Gzip — the result is byte-for-byte a tar.gz.
tar for machines
tar’s real audience today is software, not people. The format is one strictly sequential stream — no index to seek, no directory at the end — which makes it perfect for pipes: a producer can start writing while the consumer starts reading, and two tars concatenate into a valid third.
That is why machine interfaces keep asking for it: docker build sends your context to the daemon as a tar stream, kubectl cp moves files in and out of containers the same way, CI systems tar cache directories between jobs, and plenty of upload APIs accept exactly one uncompressed .tar. When the consumer is a program, plain tar is not a missing feature — it is the spec.
Frequently asked questions
Why is my tar as big as the inputs combined?
Because tar does not compress — it only concatenates files with headers. That is by design: compression is a separate layer (gzip, bzip2, xz) applied over the tar. Pick TAR.GZ on this page instead if you want the compressed kind.
When is a plain uncompressed tar actually right?
When the consumer expects one: docker build contexts, some upload APIs, streaming pipelines, and cases where the content is already compressed (photos, video) so a gzip layer would only waste time.
Does it preserve folder structure?
Files land at the archive root with their names — the browser does not hand websites full folder trees on drop. For nested structure, tar an existing archive after converting it, or accept the flat layout most transfers actually need.
Is it private?
Yes. Archives are built and converted entirely in your browser — neither the archive nor the files inside it are ever uploaded, and any password you set is applied locally. The server does nothing but deliver this page. Want proof? Run one file through, switch your connection off, and run another — it still works.