Twitter (X) Academic Research API — A Practical Guide

By Alex Chen•6 min read
TL;DR
  • The old Twitter (X) v2 Academic tier is sunset — no direct replacement product exists
  • twitterapi.io covers the same full-archive search at $0.00015/tweet with no institutional gating
  • X official offers full-archive search on pay-per-use at $0.005 per post read (and on Enterprise), about 33x our per-tweet price
  • Grant-eligible institutional projects sometimes get X's academic access reinstated case-by-case — worth applying if you have institutional affiliation

For years, X (formerly Twitter) offered a dedicated Academic Research product tier — free / discounted access to the full historical archive of public tweets, gated by institutional affiliation and research-purpose review. Academic institutions worldwide built research programs on top of it. In 2023 X sunset the tier. Full-archive search still exists on the official API, but as a paid product billed per post read.

This guide covers the practical replacements for academic and research workloads in 2026 — what's still available, what costs, and how to run the same full-archive search patterns programmatically. Pricing references URL-cited to each provider.

Cost of a 100K-tweet research corpus

~$15 via twitterapi.io vs $500 on X pay-per-use full-archive search (100,000 × $0.005)

01 — Section

What sunset — and what replaced it

Sunset (2023): the v2 Academic Research tier. This was the free-with-institutional-affiliation product that gave verified academic users full-archive search back to Twitter's first tweet in 2006.

Current state (2026): no dedicated Academic tier exists in X's product line. Full-archive search lives at /2/tweets/search/all, which docs.x.com lists as available to pay-per-use and Enterprise customers; pay-per-use bills $0.005 per post read. Case-by-case reinstatements for specific institutional research programs have been reported but are not a documented consistent product.

Third-party alternatives: providers like twitterapi.io that maintain a full-archive index at commercial pricing per tweet returned. No institutional-affiliation gating, no Developer Console onboarding — API key from an email signup.

02 — Section

Auth via X-API-Key. Full-archive search depth included (back through the platform's history for indexed tweets). Same advanced-search operator syntax as X — from:user, since:YYYY-MM-DD, until:YYYY-MM-DD, min_faves:N, lang:xx, filter:verified, #hashtag, "exact phrase", boolean combinations.

Pricing per twitterapi.io/pricing: $0.00015 per returned tweet. A 100,000-tweet corpus for a research project (moderate size) costs ~$15. A 1M-tweet corpus for a discourse-analysis or longitudinal study costs ~$150. No monthly minimum, no Developer Console onboarding. Relative cost framing: $0.00015 per tweet at twitterapi.io vs $0.005 per post read on X's pay-per-use full-archive search (docs.x.com/x-api/getting-started/pricing), so about 33x cheaper per read ($0.005 / $0.00015 = 33.3).

python
import os, requests, json
from pathlib import Path

HEADERS = {"X-API-Key": os.environ["TWITTERAPI_IO_KEY"]}
BASE = "https://api.twitterapi.io"

def research_pull(query: str, out_path: str, max_pages: int = 100):
    """Pull a research corpus to JSONL. Paginate cursor until exhausted."""
    Path(out_path).parent.mkdir(exist_ok=True)
    tweets, cursor = 0, None
    with open(out_path, "w") as f:
        for _ in range(max_pages):
            params = {"query": query}
            if cursor: params["cursor"] = cursor
            r = requests.get(
                f"{BASE}/twitter/tweet/advanced_search",
                headers=HEADERS, params=params, timeout=15,
            )
            r.raise_for_status()
            resp = r.json()
            for t in resp.get("tweets", []):
                f.write(json.dumps(t) + "\n")
                tweets += 1
            cursor = resp.get("next_cursor")
            if not cursor: break
    return tweets

# Example: climate-change discourse in 2020
n = research_pull(
    '"climate change" since:2020-01-01 until:2021-01-01 lang:en',
    "climate_2020_corpus.jsonl",
)
print(f"pulled {n} tweets")
# Cost: ~$15 for a 100K-tweet corpus at $0.00015/tweet
03 — Section

The /2/tweets/search/all endpoint keeps the full-archive functionality, and docs.x.com lists it as available to pay-per-use and Enterprise customers. On pay-per-use it bills $0.005 per post read (docs.x.com/x-api/getting-started/pricing): a 100,000-tweet corpus costs 100,000 × $0.005 = $500, and a 1M-tweet corpus costs $5,000.

Two limits matter for research budgets. First, the per-read price: the same 100,000-tweet corpus is about $15 on path 1 against $500 here. Second, the ceiling: pay-per-use is capped at 3 million Post reads per monthly billing cycle (per X's pricing page, as of 2026), so a larger corpus has to be spread over several billing cycles or moved to Enterprise, which has no public price. For institutions that already hold an X Enterprise contract, research use can ride on that agreement.

04 — Section

Common academic workloads — running them cheaply

Discourse analysis on a topic over years: pull all tweets matching your topic + date range, feed to your qualitative-analysis tool (NVivo, Atlas.ti, custom Python). Storage cost typically dwarfs API cost.

Longitudinal study of a user's posting patterns: from:username since:YEAR1 until:YEAR2 — pull one user's full archive in a date range for behavioral analysis. Sub-$1 per user for typical multi-year pulls.

Sentiment / classification training corpora: pull matched tweets, apply your classifier, iterate. Iteration cost is the classifier (LLM API + compute), not the tweet pull.

Cross-language comparative research: same query with different lang: filters, compare corpora across languages. Full archive supports going back years across all major-language subsets.

Network / graph analysis: pull tweets + author metadata to reconstruct interaction graphs. Combine advanced_search with /twitter/user/info for network-node enrichment.

05 — Section

Side-by-side — 2 paths for academic research

Costs derived from each provider's published pricing. Both paths are open to an individual researcher; the rows that differ are price and the monthly ceiling.

Dimensiontwitterapi.io advanced_searchX official full-archive search (pay-per-use)
Archive depthBack to platform history (indexed corpus)Back to platform history (full official)
Per-tweet cost$0.00015 (twitterapi.io/pricing)$0.005 per post read (docs.x.com/x-api/getting-started/pricing)
Institutional gatingnone — API key from email signupnone — X developer account with a saved payment card
100K-tweet corpus cost~$15$500 (100,000 × $0.005)
Monthly read ceilingnone3 million Post reads per monthly billing cycle; Enterprise above that
Best forindividual researchers, small-lab projects, large corpora on a grant budgetprojects that must cite the official endpoint, or institutions with X Enterprise already in place

Two practical observations: (a) individual-researcher and small-lab projects usually land at twitterapi.io on price, since the per-read gap is about 33x ($0.005 / $0.00015 = 33.3); (b) corpus size decides the rest, because anything above 3 million posts in one billing cycle does not fit X pay-per-use and needs Enterprise.

06 — Section

Reproducibility + citation notes

Cite the API + version: research using twitterapi.io should cite the provider name, endpoint, and access date in the data section of any resulting paper. Same for X official.

Store the raw JSONL, not just processed outputs: reviewers may request the raw data. Store per-query JSONL corpora in institutional archival storage per your research-data policy.

Rate limits + reproducibility: your query may return different result counts if re-run months apart (deletion, account suspension, index changes). Store the count + captured-at metadata alongside the corpus.

Ethics review: your IRB / ethics board may have specific requirements for social-media data collection. The twitterapi.io path doesn't automatically satisfy those; the compliance layer is on the researcher regardless of provider.

07 — Section

Picking the path

Individual researcher, small lab, thesis / paper project: twitterapi.io. Immediate access, about $15 per 100,000 tweets against $500 on X pay-per-use, no monthly read cap.

Project that must cite the official endpoint, or an institution with an existing X Enterprise contract: X official /2/tweets/search/all. Budget $0.005 per post read on pay-per-use and plan around the 3-million-read monthly cap.

Case-by-case X academic access reinstatement: worth applying if you have institutional affiliation + a well-defined research question aligned with X's stated academic-support priorities. Not a reliable path but occasionally granted.

Most researchers arriving here on a grant or personal budget land at twitterapi.io. Access is open on both sides; the price gap is what decides it.

python
# Practical example: build a longitudinal research corpus for a topic across years.
import os, requests, json
from datetime import date, timedelta
from pathlib import Path

HEADERS = {"X-API-Key": os.environ["TWITTERAPI_IO_KEY"]}
BASE = "https://api.twitterapi.io"

TOPIC = '"machine learning" min_faves:10'
YEARS = range(2018, 2025)  # 7-year longitudinal window
OUT_DIR = Path("ml_corpus")
OUT_DIR.mkdir(exist_ok=True)

def pull_year(topic: str, year: int) -> int:
    query = f"{topic} since:{year}-01-01 until:{year+1}-01-01"
    tweets, cursor = 0, None
    out_path = OUT_DIR / f"{year}.jsonl"
    with open(out_path, "w") as f:
        for _ in range(200):  # cap pages for budget safety
            params = {"query": query}
            if cursor: params["cursor"] = cursor
            r = requests.get(f"{BASE}/twitter/tweet/advanced_search", headers=HEADERS, params=params, timeout=15)
            r.raise_for_status()
            resp = r.json()
            for t in resp.get("tweets", []):
                f.write(json.dumps(t) + "\n")
                tweets += 1
            cursor = resp.get("next_cursor")
            if not cursor: break
    return tweets

total = 0
for y in YEARS:
    n = pull_year(TOPIC, y)
    total += n
    print(f"  {y}: {n} tweets ({total} cumulative)")

print(f"\nTotal corpus: {total} tweets")
# Cost framing per twitterapi.io/pricing:
#   ~500K tweets × $0.00015 = ~$75 for a 7-year longitudinal corpus
#   Includes JSONL per-year files ready for analysis / archival / IRB review

Researching this?Free starter credits cover end-to-end testing — Google sign-in, no card. Get a key.

08 — Questions

Questions readers ask

Is the old Twitter (X) v2 Academic Research API coming back?

No public roadmap statement to that effect. Since the 2023 sunset, full-archive search on the official API is a paid product: open to pay-per-use customers at $0.005 per post read, and to Enterprise. Third-party providers now cover the low-cost full-archive niche that Academic filled.

Can I use twitterapi.io data in a peer-reviewed publication?

Yes — cite the provider, endpoint, and access date in your data section. Reviewers generally accept commercial-API data sources when the provider is documented and the query is reproducible. Check journal-specific data policies before submission.

What about ethics / IRB review for social-media research?

IRB / ethics-review requirements apply to your research design regardless of data provider. Common requirements: individual identifiability handling, informed-consent considerations for high-profile users, storage security. The API access layer doesn't affect ethics posture.

How far back does the archive actually go?

For twitterapi.io: back through the platform's history for indexed tweets. Very old tweets (pre-2010) may have gaps due to indexing methodology of that era. For X official full-archive search: back to the platform's first tweet, most comprehensive.

Can I get free access to twitterapi.io as a student or researcher?

Signup includes free starter credits (up to $1 for new accounts that pass the automatic bot check, one per person and network, no card required, expiring after 31 days; see /pricing) — enough for method development and a first smoke test. Pilot studies and sustained research corpora move to pay-per-tweet at $0.00015. Reach out to their support if you have specific grant-funded research needs; they may have researcher accommodations.

What's the difference between full-archive search and just polling recent tweets?

Recent-search (last 7 days) is what most consumer-tier APIs offer; full-archive spans years. Academic research typically needs full-archive to construct longitudinal or historical corpora. Both twitterapi.io advanced_search and X's /2/tweets/search/all (pay-per-use or Enterprise) provide full-archive; X's recent search does not.

09 — Further reading

Continue

Sources & further reading
More from this series

Writing about something you built with TwitterAPI.io? Our community rewards program pays API credits for honest tutorials, build logs and videos.

Build it

Stop reading. Start building.

Starter credits cover real testing on real data. Google sign-in, no card, no application queue.

Get an API key

Already building with TwitterAPI.io? Publish an honest tutorial or build log about it and earn up to $200 a month in API credits.