Twitter (X) Data API: Tweets, Profiles, Followers, Search

Read public Twitter/X data: tweets, search results, profiles, followers, replies, retweeters, list and community tweets, and trends, as clean JSON, CSV or Excel. You don't need an X account, a login, cookies or proxies, and nothing runs in a browser.

This Actor gets its data from TwittrAPI, a REST API for public Twitter/X data, so runs are quick and use little memory. Every tweet and user comes back with the same flat fields, so you can load the results into a spreadsheet or database without cleaning them up first.

What can it collect?

Mode	You give it	You get
Search tweets	Search queries (X search operators work)	Tweets, or accounts with the People tab
User tweets	Usernames or user IDs	Tweets from the profile timeline, including the pinned tweet and retweets
User profiles	Usernames or user IDs	One profile per user
Followers of a user	Usernames or user IDs	Accounts that follow the user
Accounts a user follows	Usernames or user IDs	Accounts the user follows
Tweet replies	Tweet IDs or URLs	Replies in the tweet's conversation
Tweet retweeters	Tweet IDs or URLs	Accounts that retweeted the tweet
List tweets	List IDs or URLs	Tweets from an X list
Community tweets	Community IDs or URLs	Tweets posted in an X Community
Trends by location	WOEIDs (1 = worldwide)	Trending topics with rank and volume

How to use it

1. Pick a Mode.
2. Fill in the field for that mode, for example Search queries or Usernames. URLs work too: https://x.com/nasa, https://x.com/nasa/status/1844000000000000001, https://x.com/i/lists/123.
3. Set Max items per input. Each query, user, tweet, list or community is collected up to this number.
4. Click Start and download the results from the Output tab as JSON, CSV, Excel or HTML, or read them through the Apify API.

Search operators

Search queries work the same way as the X search box:

• from:nasa gets tweets from one account. to:nasa gets replies sent to it.
• "exact phrase" matches the exact words. Add -word to leave a word out.
• lang:en limits results to one language.
• since:2025-01-01 until:2025-02-01 sets a date range.
• min_faves:100, min_retweets:20 and min_replies:5 set an engagement minimum.
• filter:media, filter:links and -filter:replies filter by content type.

Use the Latest search tab for the newest tweets first. Use Top for X's own ranking.

Input examples

Search for recent English tweets about a topic:

{
  "mode": "searchTweets",
  "searchQueries": ["openai lang:en -filter:replies", "from:nasa since:2025-01-01"],
  "searchCategory": "Latest",
  "maxItems": 200
}

Get profiles for a few accounts:

{
  "mode": "userProfiles",
  "usernames": ["nasa", "https://x.com/spacex"],
  "userIds": ["11348282"]
}

Get the first 1,000 followers of an account:

{
  "mode": "followers",
  "usernames": ["nasa"],
  "maxItems": 1000,
  "maxPages": 60
}

Get replies to a tweet:

{
  "mode": "tweetReplies",
  "tweetIds": ["https://x.com/nasa/status/1844000000000000001"],
  "maxItems": 500
}

Get trends for the whole world and the United States:

{
  "mode": "trends",
  "woeids": ["1", "23424977"]
}

Limits

• Max items per input (default 100) caps the number of results for each query, user, tweet, list, community or location.
• Max pages per input (default 20) caps the number of requests for each input. A page usually holds about 20 results, so for 1,000 items set it to at least 50.
• Include raw X data adds a raw field to each item with X's original JSON. Turn it on only if you need fields that the flat output doesn't include, because it makes items much larger.

Output

Each dataset item has a type field set to tweet, user or trend. It also has an input field (the query, user, tweet, list, community or location it came from) and a mode field. The Output tab has separate table views for tweets, users and trends.

Tweet

{
  "type": "tweet",
  "id": "1844000000000000002",
  "url": "https://x.com/oldshape/status/1844000000000000002",
  "text": "@ada_example Here is a clip of it running",
  "created_at": "2024-10-09T13:59:00.000Z",
  "lang": "en",
  "author": {
    "id": "777",
    "username": "oldshape",
    "name": "Old Shape User",
    "followers": 12,
    "verified": false
  },
  "metrics": {
    "likes": 3,
    "retweets": 0,
    "replies": 0,
    "quotes": 0,
    "views": null,
    "bookmarks": 0
  },
  "media": [
    {
      "type": "video",
      "url": "https://pbs.twimg.com/ext_tw_video_thumb/1/pu/img/thumb.jpg",
      "video_url": "https://video.twimg.com/ext_tw_video/1/pu/vid/1280x720/high.mp4"
    }
  ],
  "is_reply": true,
  "is_quote": false,
  "is_retweet": false,
  "conversation_id": "1843999999999999000",
  "in_reply_to_tweet_id": "1843999999999999000",
  "in_reply_to_username": "ada_example",
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "input": "postgres",
  "mode": "searchTweets"
}

• text is the full text, including long posts over 280 characters.
• created_at is in ISO 8601 format, in UTC.
• media[].video_url is the highest-quality MP4 for videos and GIFs.
• quoted_tweet and retweeted_tweet contain the original tweet, with the same fields, for quotes and retweets.
• A field is null when X didn't send it. For example, views is null for older tweets.

User

{
  "type": "user",
  "id": "12345",
  "username": "ada_example",
  "name": "Ada Example",
  "bio": "Writes about databases.",
  "location": "London",
  "website": "https://ada.example",
  "followers": 52000,
  "following": 310,
  "tweets": 8800,
  "likes": 1200,
  "media_count": 40,
  "created_at": "2006-03-21T20:50:14.000Z",
  "verified": true,
  "verified_type": null,
  "protected": false,
  "profile_image": "https://pbs.twimg.com/profile_images/1/ada_normal.jpg",
  "profile_banner": "https://pbs.twimg.com/profile_banners/12345/1",
  "url": "https://x.com/ada_example",
  "input": "@ada_example",
  "mode": "userProfiles"
}

Trend

{
  "type": "trend",
  "name": "#Postgres17",
  "query": "#Postgres17",
  "url": "http://twitter.com/search?q=%23Postgres17",
  "tweet_volume": 15400,
  "rank": 1,
  "context": null,
  "input": "1",
  "mode": "trends"
}

The examples above use made-up accounts.

How much does it cost?

You pay per result: $X.XX per 1,000 items (tweets, users or trends). You don't pay extra for proxies or compute. To control spending, set Max items per input, or set a maximum cost per run in the run options. The Actor stops when it reaches either limit, and it keeps the items it already collected.

FAQ

Do I need an X account or API key? No. The Actor uses its own access to TwittrAPI.

Why did I get fewer results than Max items? X decides how far back a search or timeline goes, and some timelines end early. For example, X shows only some of the replies to popular tweets, and a profile timeline only goes back a limited number of tweets. To go further back on a profile, use Search tweets with from:username since:... until:....

Can I get private or deleted tweets? No. Only public data is available. The Actor skips protected, deleted or suspended content.

A run stopped early with a message about the provider. If the data provider is rate limiting requests or is temporarily unavailable, the Actor stops and keeps the items it already collected. Wait a few minutes, then run it again.

I need a field that isn't in the output. Turn on Include raw X data, or contact us and we may add it.

Can I call this from my own code? Yes, through the Apify API or the Apify clients. If you'd rather call TwittrAPI directly, without Apify, see twittrapi.com.

Is it legal to collect data from X?

This Actor collects only publicly available data. Personal data is protected by laws such as the GDPR, so only collect it if you have a legitimate reason. If you're unsure, ask a lawyer.

Disclaimer

Not affiliated with X Corp. "Twitter" and "X" are trademarks of X Corp., used here only to describe the data.

Powered by TwittrAPI.