WEIR

Twitter Archive: What's Included and What's Not (2026)

Your Twitter archive holds posts, DMs, likes and followers, but deleted posts for only 14 days. Bookmarks aren't listed, and followers come as IDs.


Published · Updated · 7 min read

Cover image: What your archive leaves out

Your Twitter archive includes your posts, DMs, likes, followers, Lists and media, but a deleted post stays in it for only about 14 days.

X (Twitter) doesn't list bookmarks, search history or watch history, and your followers come as account IDs, not usernames. You get one answer per data type, each checked on 17 November 2026 against X's own pages.

Have the ZIP already? List what's in yours without unzipping it:

# lists every file and its size
python -m zipfile -l twitter-archive.zip

What's in your Twitter archive, file by file

X's help page sums up the archive in one sentence. In a copy saved on 3 November 2026, it lists "your profile information, your posts, your Direct Messages," media, followers, Lists, ad data "and more."

The README.txt inside the ZIP names a file for each kind of data. We read copies from 2022 and late 2025, and they describe the files below almost word for word.

Table of 13 kinds of X data: whether your archive includes each, the file that holds it, and what is missing
X's own list and README, checked 17 November 2026

Before you hunt for a file:

  • A tiny file is usually empty. The README says some files "may not contain any information."
  • Each archive is a snapshot of your account on the day X built it, so keep old ZIPs.

To read the files themselves, see our guide to opening your Twitter archive.

Does the Twitter archive include deleted tweets?

Only recent ones:

  • tweets.js keeps "available Tweets which have not been deleted," edits included.
  • deleted-tweets.js holds posts "deleted in the last 14 days," before X's systems remove them.
  • deleted-tweet-headers.js adds when you deleted each one, and a media folder keeps their photos and videos.

So a post you deleted last week should be in a new archive. One from last year isn't, unless you kept an older archive.

For older deletions, look outside X. Its privacy policy says search engines and others "may retain copies of your posts longer." Our guide to seeing deleted tweets checks the Wayback Machine.

The archive starts "with your first post," X's help page says. It doesn't say how reposts appear, and the README gives them no file of their own.

Does the Twitter archive include DMs and deleted DMs?

The DMs you still have are in. For deleted ones, X doesn't say.

  • One-to-one DMs are in direct-messages.js, with text, sender, time and reactions. Group DMs get their own file.
  • Deleted DMs have no file, unlike deleted posts. X's help says deleting a DM removes it "from your account only," so others still see it.
  • Encrypted Chat messages aren't listed either. The late-2025 README has key-registry.js, your devices' public keys for encrypted DMs, but no Chat message file.

X's Chat help page says those messages stay encrypted on its servers.

Are likes, bookmarks and followers in the Twitter archive?

Likes and followers are in, but each one is missing something.

  • Likes have no dates or authors. like.js gives each post's ID, link and text. Our guide to searching your Twitter likes works from it.
  • Bookmarks aren't listed at all. X's help page and both README copies skip them, though X's privacy policy says it collects them. X's API can export your bookmarks.
  • Followers arrive as IDs. follower.js and following.js give an ID and a link per account, with no username or follow date.
  • Lists are links only. You get the links of Lists you made, were added to or follow.

To put names on those IDs, Weir's GET /v1/x/users?ids= can look up 100 public profiles per call. Each costs 2 credits, plus 1 per call.

What about media, ads and search history?

Media is in, ads are partly in, and only saved searches made it.

  • Photos, videos and GIFs from your posts sit in tweets_media, but not videos you only linked, such as YouTube ones.
  • Logins are in ip-audit.js, with IP addresses, and your old usernames are in screen-name-change.js.
  • Ad data lists promoted posts you saw or engaged with, plus interests X says "may be inferred." Conversion files go back 90 days at most.
  • Saved searches are the only search history. X's privacy policy says it logs "search terms," but the archive has just saved-search.js.
  • Watch history isn't listed in X's help page or the README. See our guide to your X watch history.

For data beyond the archive, X's privacy policy links a form to "request access to additional information."

Can you get a deleted or suspended account's archive?

Only the owner can get an archive, and only while X still has the account.

Account Archive? What X says
Yours, active Yes Request it in settings
Deactivated, under 30 days After you log back in Request before deactivating
Deleted, over 30 days No No longer in X's systems
Suspended By request Use X's privacy form
Someone else's No Only their authorized agent
A deceased user's No Deactivation only, no access

Already deactivated? Log back in within 30 days, and give it time before you request: X says posts, followers and likes "may take a while to fully restore."

If the archive never arrives, see our fixes for a Twitter archive that's not working.

Need someone's public posts? Save them as a CSV

Weir, the API we run, reads what X still shows in public, so public posts need no archive. It can't see DMs, likes, bookmarks, protected accounts or suspended ones.

This script saves an account's 20 newest posts, without reposts or replies, to a CSV file. The quickstart gets you a key with no card.

import csv, json, os, sys, urllib.request
sys.stdout.reconfigure(encoding="utf-8")  # emoji print on any console
user = sys.argv[1]  # any public account, without the @
body = {"fromUsers": [user], "query": "-filter:nativeretweets -filter:replies", "limit": 20}
req = urllib.request.Request("https://api.weirapp.io/v1/x/search", data=json.dumps(body).encode(),
    headers={"Authorization": "Bearer " + os.environ["WEIR_API_KEY"], "Content-Type": "application/json"})
posts = json.load(urllib.request.urlopen(req))["data"]
with open(user + ".csv", "w", newline="", encoding="utf-8-sig") as f:  # utf-8-sig: Excel shows emoji
    w = csv.writer(f)
    w.writerow(["id", "createdAt", "likes", "reposts", "url", "text"])
    for p in posts:
        text = "'" + p["text"] if p["text"].startswith(("=", "+", "-", "@")) else p["text"]
        w.writerow([p["id"], p["createdAt"], p["likeCount"], p["retweetCount"], p["url"], text])
        print(p["createdAt"][:10], p["likeCount"], "likes", p["text"][:42].replace("\n", " "))
print(f"Saved {len(posts)} posts from @{user} to {user}.csv")

We ran it on 17 November 2026 on @internetarchive, home of the Wayback Machine:

Terminal output of a 15-line Python script saving 20 public posts from @internetarchive to a CSV file through Weir
Our run on 17 November 2026: 20 posts, 21 credits

It saved 20 posts from 24 October to 13 November. The most liked, with 2,494 likes, was about the Wayback Machine itself. Our first run crashed on an emoji on Windows, which the reconfigure line fixes.

20 posts cost 21 credits; 5,000 credits a month are free; $5 buys 25,000. For a full history, see how to download all tweets from a user.

When not to use Weir

We run Weir, so here's where it's the wrong tool:

  • You want DMs, likes or bookmarks. They're private, and Weir never sees your account.
  • The account is suspended or deleted. X shows nothing for it, and neither can Weir.
  • It's your own account and you can wait. X's archive is free and holds far more.

Weir isn't licensed by X, so access can change.

FAQ

Does the Twitter archive include deleted DMs?

X doesn't say. The README has DM files but none for deleted messages. X's help page says deleting a DM removes it from your account only, so the others in the chat keep it.

Does the Twitter archive include deleted tweets?

Only posts deleted in about the last 14 days, in deleted-tweets.js, before X's systems remove them. An archive you downloaded before a deletion still holds that post, since each archive is a snapshot.

Can you get the Twitter archive of a suspended or deleted account?

A suspended account can ask through X's privacy form. A deactivated one can log back in within 30 days and request it. After that, X says the account is "no longer available in our systems."

Can I download someone else's Twitter archive?

No. X sends a code to the account's own email or phone before it builds an archive, and it gives no one access to a deceased user's account. For public posts, use the script above.

Sources

Checked 17 November 2026. X's help pages block our fetcher, so these are Wayback copies, dated: