Free ZIP, RAR & 7z Extractor

Worked example

Drop holiday.zip, a 38 MB archive with 14 files. The page shows "14 files" and "52.3 MB uncompressed". Files inside folders sit under a heading such as "photos", and files at the top sit under "(top level)". Type itinerary into "Filter files…" and the count changes to "Showing 1 of 14 files". Press "Preview" on that text file to read it in the page, then "Hide" to close it. Clear the filter and tick two files. The buttons now read "Download 2 selected files as ZIP" and, in a browser that can pick a folder, "Extract 2 selected files to a folder". The first one saves holiday-selected.zip and the page says "Extracted 2 files." With nothing ticked, the buttons read "Download all 14 files as ZIP" and "Extract all 14 files", and the new ZIP is named holiday-files.zip.

Which settings to use, and the limits

• Preview a file before you save it. Press "Preview" beside a file. Images, PDFs, audio and video open in the page if they are 60 MB or smaller, and text files if they are 5 MB or smaller. A text preview shows the first 50,000 characters, then "… (truncated — download for the full file)", but "Copy text" copies the whole file. Other file types are checked only up to about 200 KB, and a binary one says "This looks like a binary file — download it to view." Larger media or text says "This file is too large to preview — download it instead."
• Choose between one new ZIP and a real folder. "Download all 14 files as ZIP" packs the files into one new ZIP named after the archive, with "-files" or "-selected" added. It works in every browser. "Extract all 14 files" appears only when the browser can open a folder picker, which the code treats as a Chromium feature. It asks you to pick a folder, rebuilds the folders inside it and shows "Extracting… 3/14". Tick a folder's box to select all its files, and use "Clear" to start over. On an iPhone or iPad that can share files, the finished ZIP waits behind a "Save holiday-files.zip (4.2 MB)" button, because iOS needs a fresh tap.
• Enter the password for protected archives. When files are protected, the page says "2 files are password-protected. Enter the password to preview or extract them." Type it into "Archive password" and press "Unlock". The tool tests it by unpacking the first protected file, then shows "Password accepted — protected files can now be previewed and extracted." A wrong one gives "That password didn't work. Check it and try again." Classic ZIP encryption and AES-encrypted ZIPs both work. Encrypted RAR and 7z archives do not: the names are listed, but the page tells you to use 7-Zip or WinRAR. The password is used only inside this tab.
• Fix garbled file names with the encoding menu. Older ZIP tools save names in a local code page instead of UTF-8. When the tool finds such names it shows "File name encoding" with the note "This archive stores file names in an older code page. If names look garbled, pick the matching encoding." "Auto" keeps UTF-8 when the bytes are valid UTF-8. Otherwise it guesses from the page language: CP866 for Russian, Ukrainian, Belarusian, Bulgarian, Serbian, Macedonian and Kazakh, CP1254 for Turkish, Shift-JIS for Japanese, GBK for Chinese, EUC-KR for Korean and CP437 for the rest, including English. Picking another entry relists the files and clears your ticks and preview.
• Know which formats open. ZIP and ZIP-based files such as .docx, .xlsx, .pptx, .epub, .jar and .apk open, and so do RAR (versions 4 and 5), 7z, TAR, .tgz and .tar.gz, .tbz2 and .txz. A lone .gz, .bz2 or .xz file with no tar inside is shown as one file, named from the gzip header or else from the archive name without .gz. RAR, 7z, TAR and some ZIPs need a reader of about 1 MB, announced as "Loading the RAR/7z reader (about 1 MB, first time only)…". One part of a split archive, such as .part1.rar, .7z.001 or .z01, cannot be opened.
• Know the size limits and warnings. The whole archive is held in memory, and a file over 2 GB is refused on every device. A notice appears above 1 GB, or above 150 MB on a phone or tablet: "This archive is 400 MB. It is held in your device's memory while open, so a phone or tablet may slow down or reload the page." Before a save, the tool estimates peak memory. If it looks too high you see "This may be too much for this device" with "Choose files instead", "Continue anyway" and "Cancel". A total over 1 GB or a ratio over 100:1 raises a red warning but does not block you.

Common problems and how to fix them

It says "Could not extract that file."

This is the general message when unpacking fails for a reason the tool has no specific text for. The usual cause is an archive that is cut off or damaged, so one file's data cannot be read or inflated. It can also appear when the browser cannot save the new ZIP. Press ⬇ beside single files to find the bad one, untick it and save the rest, because one failing file stops a whole bulk save. If the archive was downloaded, download it again.

It says "This file is 2.4 GB — larger than a browser can hold in memory (2 GB). Open it on a computer with a desktop archive tool."

The size shown is your file. The tool refuses any archive over 2 GB before reading it, on every device, because the whole archive is held in memory. Open it with 7-Zip, WinRAR or your computer's built-in unzip, or ask the sender for several smaller archives. Below 2 GB it can still fail with "This archive is too big for your device's memory. Try a computer, or a smaller archive." A computer usually has more room than a phone.

It says "1 file could not be saved under its name in this folder and was skipped: notes.txt. Use "Download as ZIP" to get it." (with your own count and names).

The browser's folder API refuses names the operating system cannot hold, and Chromium browsers such as Edge on Windows throw a TypeError for them. The tool first cleans every name: it replaces control characters and < > : " | ? * with "_", drops trailing dots and spaces, puts "_" in front of names like CON, NUL and COM1, adds "_" to names ending in .lnk, .local, .url or .scf, and cuts names at 200 characters. A name the browser still refuses is skipped and the rest is saved. Saved names can therefore differ from the archive's.

It says "Could not read this archive. Make sure it's a valid, complete archive file."

A ZIP whose end-of-archive records cannot be read, usually because the download was cut short or the file is damaged. The same message appears if the browser cannot read the file from disk at all. Download the file again and compare its size with the source. A file that is not an archive of any supported type gets a different message: "This doesn't look like an archive this tool can open. It opens ZIP, RAR, 7z, TAR, GZ, BZ2 and XZ files."

It says "This file is packed with Deflate64 compression, which this tool can't unpack yet. Open the archive with 7-Zip or your computer's built-in unzip to get it."

A few compression methods inside ZIPs cannot be unpacked here. Deflate64, Zstandard, JPEG and WavPack are named in the message, and other unknown methods appear as a number. The code notes that Windows' own Compressed folder writes Deflate64 for big files. Such files are marked with ⚠ in the list, a banner counts them, and "Download as ZIP" and "Extract" skip them and save the rest. Stored, Deflate, BZIP2, LZMA, XZ, PPMd and AES entries all unpack.

How it compares

7-Zip, WinRAR or your computer's built-in unzip are the better choice for archives over 2 GB, split archives, encrypted RAR or 7z files and ZIPs that use Deflate64 or Zstandard. This tool only reads archives. It has no way to create one, apart from "Download as ZIP", which repacks the files you chose into a new ZIP with no password and no compression setting. Its strengths are that nothing needs installing and you can list, filter and preview before unpacking. The files are unpacked in the browser and are not uploaded. Because the whole archive sits in memory, a phone has less room than a computer. "Extract all" to a real folder needs a Chromium browser.