How to Download Video from the Wayback Machine (and What About YouTube?)

You can download video from Wayback Machine captures only when the archive actually captured the video file, not just the page that played it. Find the capture, check whether the video itself loads, then save the stored file using its original-file URL. For many sites, and for most YouTube videos, the archive kept the page, the title and the thumbnail but not the video stream. This guide shows how to check quickly, how to download video from Wayback Machine snapshots when a file exists, what to try for archived YouTube links, and where the honest limits are.

What the Wayback Machine stores when it captures a video page

The Wayback Machine is a crawler-based archive. It saves what its crawler requested and received: usually the HTML of a page, plus some of the images, stylesheets and scripts that page referenced. Whether a video was saved depends on how the original site delivered it.

• Direct video files such as a link or player pointing at an .mp4, .webm or .mov file are the most likely to have been captured, especially if the file was small and linked from a page the crawler visited often.
• Old Flash video (.flv and .swf players) was often captured, because these were single files. You may need a separate player to view them.
• Streaming players that load video in many small segments (HLS or DASH, with .m3u8 or .mpd playlists), or that fetch the stream through JavaScript after the page loads, are rarely complete in the archive.
• Embedded third-party players from video platforms depend on the platform’s own servers. The archive usually holds the embed page, not the video behind it.

So the first question is never “which tool should I use” but “did the archive keep the file at all?”

Step 1: find the capture and check whether the video plays

1. Open web.archive.org, paste the page address and pick a capture from the calendar.
2. Press play. If the video plays inside the archived page, the file is almost certainly stored.
3. If the player is blank or shows an error, open your browser’s developer tools, switch to the Network tab, filter by “media”, and reload. A request to web.archive.org/web/… that returns a video file with status 200 means the archive has it. A 404 or a request to the original site means it does not.
4. Try two or three other capture dates. Video files were often captured on some crawls and not others.

If the page itself is not in the archive, start with our checklist for a website not in the Wayback Machine, which covers address variants and other places a copy might exist.

Step 2: list video files with the CDX API

When a site had many videos, checking pages one at a time is slow. The Wayback Machine’s CDX API can list every captured file under a domain and filter by type. A query such as:

https://web.archive.org/cdx/search/cdx?url=example.com/*&filter=mimetype:video/.*&collapse=urlkey&output=json

returns one line per unique video URL that the archive captured, with its timestamp, MIME type, status and size. Add filter=statuscode:200 to skip redirects and errors. If your videos were Flash, try mimetype:application/x-shockwave-flash as well, and search for .flv in the URL list.

Our Wayback Machine API guide explains the CDX parameters in more detail, including how to page through large results without being rate-limited.

Step 3: download video from Wayback Machine captures with the original-file URL

Once you know the timestamp and the original URL of the video, build the “original file” form of the address by adding id_ after the timestamp:

https://web.archive.org/web/20150312093000id_/http://example.com/media/intro.mp4

That URL returns the stored file as it was captured, without the archive’s toolbar or rewriting. You can then:

• open it in the browser and use “Save video as”, or right-click the player and save;
• download it from the command line, for example curl -L -o intro.mp4 "<id_ URL>" or wget with the same URL;
• add a list of id_ URLs to a text file and fetch them one at a time with a pause between each.

Large files may download slowly or stop partway. If that happens, try again later and check the file size against the size reported by the CDX API. Do not run many downloads at once; the Internet Archive limits heavy traffic and answers with HTTP 429 errors. Our guide to the Wayback Machine 429 error explains how to pace downloads.

Archived YouTube videos: what the Wayback Machine can and cannot give you

Searches like “wayback machine youtube” usually come from people looking for a deleted or private YouTube video. The honest picture:

• Most captures of YouTube are watch pages. They often preserve the title, description, channel name, upload date and thumbnail, which can be valuable on their own, but the player will not play.
• Some videos were archived as files. Coverage is patchy and favours videos that were popular or linked from archived pages.
• A community-documented URL pattern can sometimes reach them. Users share a link format that places the video ID after web.archive.org/web/2oe_/http://wayback-fakeurl.archive.org/yt/. It is not an officially documented feature. When it works, the video plays in the browser and can be saved; when it does not, the archive reports that the URL was not archived.

Before trying anything, find the 11-character video ID in the old link (the part after v= or after youtu.be/). Search the Wayback Machine for the full watch URL first; even if the video does not play, the watch page tells you the title and date, which helps when searching other sources such as old embeds on archived blogs.

Respect copyright and the uploader’s choices. A video that was removed by its owner, or for legal reasons, is not free to republish just because a copy exists. Downloading your own lost content, or keeping a private reference copy for research, is a very different situation from re-uploading someone else’s work.

Recovering videos as part of a whole archived website

If you are rebuilding an old site, video is usually one item on a longer list of missing assets. Treat it the same way as images and documents:

1. List every video the old pages referenced (from the archived HTML).
2. Check each against the CDX list to see which files were captured.
3. Download the captured files with id_ URLs and store them with their original paths.
4. For each file that was never captured, decide whether to replace it with a copy from your own records, re-host it elsewhere, or remove the player and leave a short note.
5. Update the page code so players point at the new local files, not at web.archive.org.

The same approach applies to other media; our guides to recovering missing images and recovering PDFs and downloads cover those file types.

Why the video will not download: common problems

Problem	Likely cause	What to try
Player shows but nothing plays	Only the page was captured, not the video	Check the Network tab and the CDX list; try other dates
Download stops partway	Large file, network interruption or rate limit	Retry later, one file at a time; compare file sizes
File downloads but will not open	Saved an HTML error page or a playlist, not the media	Check the file size and type; use the id_ URL of the actual media file
Only a .m3u8 or .mpd file exists	Segmented stream; segments usually not archived	Look for an older capture that used a direct file
Flash video will not play	Modern browsers dropped Flash	Download the .flv and open it in a desktop media player that supports it

When to get help

Finding one video is a quick job. Finding every surviving media file across a few hundred archived pages, downloading them without tripping rate limits and wiring them back into a working copy of the site takes longer. Our archived website download service includes collecting the media the archive actually holds and pointing the pages at local copies. If you are not sure what survived, ask for a free check of what the archive saved first. We can only deliver files the Wayback Machine captured; if a video was never archived, we will tell you that rather than promise it.

Frequently asked questions

Can I download any video from the Wayback Machine?

Only videos whose files were captured. Many archived pages show a player but the stream behind it was never saved. Check whether the video plays in the capture or appears in a CDX listing. The same applies to clips attached to tweets; see how to find archived tweets in the Wayback Machine.

Does the Wayback Machine have deleted YouTube videos?

Sometimes, but not usually. It more often has the watch page with the title and description. A small share of videos were archived as files and can sometimes be reached through community-documented URLs.

What is the best format to save an archived video in?

Keep the original file exactly as downloaded. Convert a copy only if you need it for a modern player, and keep the original as your reference.

Is it legal to download video from Wayback Machine archives?

The archive makes captures available for research and reference. Rights to the video stay with whoever owns it. Recovering your own material is usually straightforward; republishing someone else’s video needs their permission.