What replaced the Twitter academic API

#What replaced the Twitter Academic Research API

No program replaced the Twitter Academic Research API. In September 2026 an X developer-relations engineer put it plainly on the official forum: "There is no separate academic / free research tier today" (opens in a new tab), and pay-per-use is the self-serve path for students and faculty alike.

The track itself opened on January 26, 2021. Approved academics got full-archive search back to 2006, 10 million posts a month and 1,024-character queries, at no cost. It was the data source behind thousands of papers and tools such as academictwitteR and twarc.

The end came in two steps. In February 2023 Twitter announced that free API access would stop. On March 30, 2023 it launched paid tiers (opens in a new tab), retired the old access levels "over the next 30 days" and told academics it was "looking at new ways to continue serving this community". No new program followed, and academic keys stopped working by mid-2023. A survey of 167 researchers (opens in a new tab) published that December counted more than 100 projects canceled, halted or moved to other platforms.

The gap between then and now is easiest to see side by side.

• Price per 1,000 posts

  Academic Research track, 2021–2023

  Free

  Official X API, November 2026

  $5.00, or $15.00 with each post's author from November 22

• Monthly ceiling

  Academic Research track, 2021–2023

  10M posts

  Official X API, November 2026

  3M post reads, then Enterprise from about $42K a month

• Archive

  Academic Research track, 2021–2023

  Back to 2006

  Official X API, November 2026

  Back to 2006

• Who could use it

  Academic Research track, 2021–2023

  Approved academic researchers

  Official X API, November 2026

  Anyone with a developer account and prepaid credits

The X forum gives every new question about academic access the same reply: enroll in the standard pay-per-use API, or apply under Article 40 of the EU's Digital Services Act if the study qualifies.

#Three routes to X data for research in 2026

A researcher has three working routes to X data in 2026: the official X API at standard prices, EU researcher access for studies of systemic risks in the EU, and Sorsa API, a Twitter (X) API and scraper with academic access for non-commercial studies.

The official X API, pay-per-use

The official X API sells posts at $0.005 each and profiles at $0.010 (opens in a new tab), from prepaid credits, and its full-archive search reaches back to 2006. You need an approved developer account, a project, an app and its keys. Pay-per-use stops at 3 million post reads a month; past that, only an Enterprise contract.

Two details hit research budgets hardest. From November 22, 2026, authors and other expanded objects are billed (opens in a new tab) like direct reads, so a post with its author's profile costs $0.015. And X issues no letters for ethics boards (opens in a new tab): institutions rely on the public terms and proof of a paid tier.

EU researcher access under the Digital Services Act

Article 40 of the Digital Services Act obliges very large platforms, X included, to give data to researchers whose work studies systemic risks in the EU. Article 40(12) covers publicly accessible data, for researchers at universities and not-for-profit bodies that meet its conditions. Article 40(4) covers non-public data for "vetted researchers", approved by a national Digital Services Coordinator through the EU Data Access Portal (opens in a new tab), open since October 29, 2025.

The route is real but narrow. A thesis on, say, oil prices or brand sentiment doesn't qualify, and on X's forum a PhD student whose request was declined reported getting no feedback at all.

Sorsa API with academic access

Sorsa API gives researchers the same archive through one API key: search with X's operators back to 2006, posts by ID 100 at a time, timelines, replies, quotes and follower lists, with no developer account. Every tweet arrives with its author's full profile at no extra cost, and every key runs at 20 requests per second.

For non-commercial research in any discipline, Sorsa API adds academic access, shaped around the study: requests granted to your project or a deep discount on a plan. PhD and Master's students, postdocs, faculty, research staff and NGO or newsroom researchers apply from an institutional email, and a person on the team reads every application.

#How to collect a study corpus with Sorsa API

A study corpus is the set of posts your analysis runs on: here, every English post about a heat wave in July 2022. Collecting it takes a key, one test request and a short script that keeps a record of what it did.

1. Get your API key. Sign up with your institutional email: the account starts with 100 free requests, no card needed, and the key works on every endpoint.

2. Test your date range. Send one search with the since: and until: operators and look at what comes back before you plan the full collection.

   curl -X POST "https://api.sorsa.io/v3/search-tweets" \
     -H "ApiKey: YOUR_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{"query": "\"heat wave\" lang:en since:2022-07-01 until:2022-08-01",
          "order": "latest"}'
   

   {
     "tweets": [
       {
         "id": "1549368207425953792",
         "full_text": "Third day above 40C. Cooling centers open until 10 pm.",
         "created_at": "2022-07-19T12:41:07Z",
         "lang": "en",
         "likes_count": 212,
         "retweet_count": 64,
         "reply_count": 18,
         "quote_count": 5,
         "view_count": 48210,
         "conversation_id_str": "1549368207425953792",
         "is_reply": false,
         "user": {
           "id": "1203345678901234567",
           "username": "heatwave_watch_desk",
           "display_name": "Heat Watch Desk",
           "followers_count": 15420,
           "created_at": "2019-12-07T09:15:00Z",
           "location": "London",
           "verified": false
         }
       }
     ],
     "next_cursor": "DAACCgACGW5vTq3AJxAKAAMZbm9Ord-x8AgABAAAAAIA"
   }
   

   Show full response

3. Collect every page and log the run. The POST /search-tweets (opens in a new tab) endpoint returns about 20 posts per page and a next_cursor for the next one; the script follows it until it comes back empty and writes a manifest beside the data.

   import json, time, requests
   from datetime import datetime, timezone
   
   URL = "https://api.sorsa.io/v3/search-tweets"
   KEY = "YOUR_API_KEY"
   QUERY = '"heat wave" lang:en since:2022-07-01 until:2022-08-01'
   
   manifest = {
       "provider": "Sorsa API (https://api.sorsa.io)",
       "endpoint": "POST /v3/search-tweets",
       "query": QUERY,
       "order": "latest",
       "collected_at": datetime.now(timezone.utc).isoformat(),
   }
   ids, cursor = [], None
   with open("posts.ndjson", "w", encoding="utf-8") as out:
       while True:
           body = {"query": QUERY, "order": "latest"}
           if cursor:
               body["next_cursor"] = cursor
           r = requests.post(URL, headers={"ApiKey": KEY}, json=body)
           if r.status_code == 429:  # above 20 requests per second
               time.sleep(1)
               continue
           r.raise_for_status()  # 403: missing key or no requests left
           page = r.json()
           for tweet in page.get("tweets", []):
               out.write(json.dumps(tweet, ensure_ascii=False) + "\n")
               ids.append(tweet["id"])
           cursor = page.get("next_cursor")
           if not cursor:
               break
   
   manifest["posts"] = len(ids)
   with open("post_ids.txt", "w") as f:
       f.write("\n".join(ids))
   with open("manifest.json", "w") as f:
       json.dump(manifest, f, indent=2)
   

4. Apply for academic access before the full run. Send the application on the academic research page with your study's scope while the free requests cover the pilot; the agreed access appears on the same account and key.

   Good to know

   A post collected today carries today's like and view counts and its author's current profile, even if it was written in 2014. Deleted posts and protected accounts are never returned, so a list collected in 2022 comes back smaller now. Write both points into your limitations section.

For years-long collections split by month, the guide to historical Twitter data covers backfill windows. To rebuild a published dataset from its ID list instead, rehydrate tweet IDs through POST /tweet-info-bulk (opens in a new tab), 100 posts per request.

#What a study's tweets cost

A corpus of 1 million tweets from archive search costs from $70 on Sorsa API and $5,000 on the official X API, or $15,000 once authors are billed. Switch the size to see your study.

The same budget logic holds for other study designs. Rebuilding a published dataset by post ID costs from $0.014 per 1,000 posts, against $5.00 on the official X API. A follower network costs from $0.007 per 1,000 profiles, against $10.00. Prices quoted with "from" are at the lowest per-request rate; academic access lowers what a study pays further.

Time matters on a thesis schedule too. At 20 requests per second, 1 million tweets from search take about 42 minutes, and 1 million posts by ID about 8 minutes.

#Make your data hold up in review

Reviewers and ethics boards ask the same three things about X data: where it came from, whether someone else can rebuild it, and what it leaves out. A record kept during collection answers all three.

Fill in your study below to get a data-collection paragraph for the methods section.

Ethics boards usually ask a short list of questions. Sorsa API's answers:

• What data? Public posts and profiles only: no protected accounts, no direct messages.
• Any interaction with users? None. Sorsa API is read-only: it never posts, likes, follows or messages.
• Whose X account is used? None: Sorsa API never asks for your X login.
• What does the provider keep? Request logs for one year and full responses for 30 days. X never learns who is asking.
• Paperwork? On request, the team writes a short letter about the data source for your ethics board.

Sharing follows X's Developer Policy (opens in a new tab): publish post IDs and user IDs, not full posts. The policy caps sharing at 1.5 million post IDs per entity every 30 days without X's written permission, and lets academic researchers share IDs for non-commercial research approved by X in writing or for peer review. For datasets built to train models, the guide to a Twitter dataset for machine learning covers the training clause.

#How to choose a Twitter API for academic research

The tests that matter for a study are price per 1,000 posts, what each post carries, archive depth, a monthly ceiling above the corpus, setup a student can finish in a day, speed and an academic offer. Sorsa API meets each with a number, and because it is read-only, no one's X account is ever at risk.

• Price per 1,000 tweets, archive search

  Sorsa API

  from $0.07

  Official X API

  $5.00 ($15.00 with authors from November 22)

• Price per 1,000 posts by ID

  Sorsa API

  from $0.014

  Official X API

  $5.00

• Author profile on each post

  Sorsa API

  Included

  Official X API

  Billed as a user read from November 22

• Archive

  Sorsa API

  Back to 2006

  Official X API

  Back to 2006

• Monthly ceiling

  Sorsa API

  Your plan or agreed academic access

  Official X API

  3M post reads, then Enterprise

• Setup

  Sorsa API

  One API key, no developer account

  Official X API

  Developer account, project, app, keys

• Speed

  Sorsa API

  20 requests per second on every plan

  Official X API

  Rate limits per endpoint

• Academic offer

  Sorsa API

  Granted requests or a deep discount

  Official X API

  None since 2023

The first row decides most grant budgets: a 3M-tweet corpus from archive search costs from $210 on Sorsa API and $15,000 on the official X API, before authors are billed.

#Start your study on 100 free requests

X ended free academic access in 2023, and its pay-per-use price of $5.00 per 1,000 posts now applies to every study. For a corpus searched back to 2006 at from $0.07 per 1,000 tweets, use Sorsa API: test your date range on the free requests today, then apply for academic access with your study's scope.

#FAQ