Open .warc, .warc.gz and .wacz web archives directly in your browser. Choose one or more archives in Load archives below and press Process Locally; every URL they captured appears in Records, grouped by site. Click a capture to replay it in this frame from its captured response — with back and forward, a trail of what you have visited, and the raw HTTP headers. Your archives are parsed on your own machine and never uploaded.
No files selected yet.
No records yet. Select .warc, .warc.gz or .wacz files below and process them.
A WARC (Web ARChive) file is the standard container for a web crawl: every request and response, with its headers and payload, stored as it crossed the wire. It is the format behind the Internet Archive's Wayback Machine, what wget --warc-file writes, and what Webrecorder's tools package into WACZ. Opening one has usually meant installing command-line tooling or uploading the file to someone else's server.
This page is an online WARC viewer that runs in your browser. Choose one or more archives and they are parsed on your own machine into a list of every URL they captured, grouped by site. Click a capture to see the page replayed from its captured response, follow its redirect chain, and read the raw HTTP headers. If you have a .warc, .warc.gz or .wacz and want to see what is inside it — with no account, upload or install — this is the tool.
.warc.gz archives inside are read straight out of the zip; nothing is extracted, and a container is listed as one row per archive it holds. A .wacz.zip — how some browsers rename the download — works the same way.Not supported: the older .arc format, and a .wacz whose archives were re-compressed (deflated) inside the zip rather than stored — which the usual tools do not do.
.warc, .warc.gz or .wacz files. Several at once is fine..warc — the form says how many threads the selected files can actually use..zip.Working outside the browser? How to open a WARC file covers the Python and command-line routes too.
The archive is read as a stream rather than loaded whole, so multi-gigabyte files work. Nothing is sent anywhere: the file is opened from disk by the browser and parsed in Web Workers on your machine, which is why there is no size cap and no queue.
A plain .warc over 128 MB is split into 64 MB ranges and shared between threads. A compressed archive is read by one thread, because gzip members have to be inflated in order.
Records lists every captured URL grouped by origin, filterable by URL segment and by content type — HTML, images, CSS, JavaScript, fonts and the rest.
Opening one replays it from the captured response, with its own address bar, back and forward, a trail of where you have been, and the original HTTP response headers. Links between captured pages work; redirect chains are followed, and a redirect that carries a page of its own is shown rather than skipped.
Parsing runs in Web Workers inside your browser. Archive contents never leave your browser and are never sent to this server. There is no upload step, no account, and no size limit imposed by us.
The only thing fetched over the network is the parser itself — a few hundred kilobytes of code served by this site, cached after the first visit. It is the same parser that indexes this archive's own collection, so the browser and the server read a WARC identically.
.warc.gz takes as long as one core needs to inflate it; the same data as a plain .warc can be split across threads..zip writes straight to disk in Chromium browsers. Elsewhere the zip is built in memory first, which is fine for a page and its assets but not for gigabytes.M4cgyvers Archives is a personal web archive: crawls of forums, personal sites and small communities, stored as WARC and indexed so they can be searched. The viewer exists because checking a fresh crawl should not need a server round-trip, and because opening a WARC should not mean uploading it anywhere. It runs the archive's own parser — one codebase for the server and the browser — so what you see here is what the archive sees.
WARC (Web ARChive, ISO 28500) stores the raw traffic of a crawl. Each record holds one request, response or metadata block with its own headers, so a page and every image, script and stylesheet it loaded are kept as the server sent them. The Wayback Machine, national libraries and wget all write it; the specification is maintained by the IIPC.
Yes — this page is one way. Worth checking on any tool that offers it: some upload your archive to a server and cap the size. Here the file is read locally, so there is no cap and nothing to upload.
Same content. In .warc.gz each record is its own gzip member, so a reader can jump to any record without inflating the rest of the file. Nearly every crawler writes .warc.gz by default. Both open here without you decompressing anything.
WACZ (Web Archive Collection Zipped) is Webrecorder's package format: a zip holding one or more .warc.gz archives plus an index and a list of pages. It is what the ArchiveWeb.page extension saves and what Browsertrix Crawler produces. Open it here like any other archive; each .warc.gz inside is listed on its own.
No. It is parsed on your machine, and nothing leaves your browser except the request for the parser code.
Yes — that is what this page is for. No Python, no warcio, no account. Choose the file, press Process Locally, and browse.
You can filter the record list by URL segment and by content type, which is usually how you find one page in a large crawl. Full-text search of page contents is not part of the viewer.
Yes. Archives are streamed rather than loaded whole, and there is no upload step to impose a limit. What grows with the archive is the record list held in memory, not the bytes.
On the command line, wget --mirror --page-requisites --warc-file=site https://example.com/ writes site.warc.gz as it crawls (see the wget manual). In a browser, the ArchiveWeb.page extension records the pages you visit and saves a .wacz. For whole sites, Browsertrix Crawler produces .wacz too. This archive's own collection is crawled with wget.
If you have an archive of it, yes: pick the file and browse it here. If you do not, search the M4cgyvers web archive — over 1,700 WARC files and more than 11 million captured responses — and view captures from there.
Because your files stay offline. The page is online — it is a website — but what you open never leaves your computer, and that is the property that matters. See also how to open a WARC file for the non-browser options.