If you tried to pull Twitter data for free in 2023 and came back in 2026 expecting the same setup, you're in for a rude awakening. The free API is gone, the academic research track is discontinued, and half the open-source tools that once powered entire research labs are now digital fossils.
X's pricing changes forced researchers, sales teams, marketers, and data scientists to rebuild workflows that had depended on free or inexpensive access. Twelve options were operating at the August 2026 research cutoff, with different costs, data-acquisition methods, and target jobs. “Still work” means that a current product or listing could be verified—not that every option is authorized by X, complete, or suitable for production.
Why "Twitter Data Providers" Is Even a Search Term Now
A few years ago, nobody Googled "Twitter data providers." You just used the free API, or you spun up snscrape, or you grabbed a library like Tweepy and went to town. Then came the Musk acquisition, and the dominoes fell fast:
- Early 2023: Free API v1.1 access shut down. Paid-only tiers introduced.
- Mid-2023: Academic Research track discontinued — a lifeline for thousands of researchers, gone overnight.
- 2023–2024: X used named subscription tiers, including Basic and Pro. Those tiers are historical context, not current buying guidance.
- 2025–2026: X shifted to a prepaid, pay-per-usage credit model. No more flat monthly plans. Standard Post read: $0.005. User read: $0.010. Self-serve cap: two million Post reads per month.
The free-library era weakened in parallel. snscrape's X search is no longer dependable, Twint is archived, and the public Nitter ecosystem has repeatedly been disrupted by X changes. That's why “Twitter data providers” became a real search query: teams needed durable alternatives.
What's Dead, What's Broken: A Free Twitter Tool Audit for 2026
Before the twelve current options, it helps to separate API clients from acquisition methods and to retire recommendations that no longer provide dependable access.
| Tool | Status (2026) | Notes |
|---|---|---|
| snscrape | ⚠️ Broken / unmaintained | Last reliable commit in 2023; X endpoint changes killed it |
| Twint | ❌ Dead | Archived repo; hasn't worked since API v1.1 deprecation |
| Nitter | ⚠️ Unreliable as a provider layer | Public instances and access methods have repeatedly been disrupted by X changes |
| Tweepy | ✅ Works (with paid API key) | Maintained, but requires a paid X API key — no longer free |
| twarc | ✅ Works (with paid API key) | Academic-community maintained; same paid-key requirement |
Tweepy and twarc are API clients, not data providers; they remain useful only when paired with valid current X API access. snscrape, Twint, and Nitter should not be recommended as durable acquisition layers for a new production workflow.
There are current alternatives, but none recreates the old combination of broad, durable, no-cost access with minimal setup.
How Twitter Data Providers Actually Get Their Data
Not all "Twitter data providers" work the same way, and the differences matter — for reliability, compliance, and cost. Here's the taxonomy I use:
-
Official API access — You use your own X API key, hit X's endpoints directly, and play by X's rate limits and developer policy. Tweepy and twarc fall here. So does the X API itself.
-
Scraping-backed APIs — A vendor operates automated public-web retrieval (browsers, reverse-engineered endpoints, etc.) and sells you the output as a clean API. Faster and often cheaper, but X's Terms of Service and automation rules prohibit crawling or scraping without prior written consent. Bright Data, SocialData.tools, and most Apify Actors fall here.
-
Licensed social listening platforms — Enterprise vendors like Brandwatch and Meltwater historically had firehose-level agreements with Twitter. Some still hold official partnerships; others operate under evolving contracts. Coverage, export rights, and data redaction rules vary by package.
-
Browser-based extraction tools — Tools such as Thunderbit can structure visible data on compatible pages where the user has permission to automate. This is different from a bulk API or firehose and must not be treated as a way around X's access controls.
Understanding which category a provider falls into tells you a lot about its durability, legal posture, and what happens when X changes something.
What to Look for When Choosing a Twitter Data Provider
Each option is evaluated against six practical criteria:
- Data access method — Official API, scraping-backed, licensed listening, or browser-based?
- Pricing transparency and entry-level cost — Can you see prices before talking to sales?
- Data types returned — Posts, profiles, engagement metrics, historical archive?
- Compliance posture — ToS alignment, GDPR/CCPA readiness, deletion handling?
- Technical barrier — Developer-only, or can a non-coder use it?
- Reliability evidence — Does the provider publish an SLA, status history, maintenance signals, and a change-notice process?
No single provider wins on all six. The right choice depends on your volume, budget, technical skill, and how much maintenance you can tolerate. With that framework in mind, here are the twelve.
All 12 Twitter Data Providers Compared (2026)
Before the deep dives, here's the full comparison table. Prices are current at publication — always verify on the provider's site, because this market moves fast.
| Provider | Access Method | Pricing Starts | Historical Data | Real-Time | No-Code Option | Best For |
|---|---|---|---|---|---|---|
| X Official API | Direct first-party API | $0.005/Post read (pay-per-use) | Full Archive to March 2006 | Yes (Filtered Stream) | No | Developers needing official endpoints |
| Bright Data | Managed scraper + datasets | Free 5K records/mo; PAYG $1.50/1K | Dataset/date-window options; no complete-archive guarantee found | Live scrape on request | Partial (control panel) | Managed public-web collection |
| SocialData.tools | Scraping-backed API | ~$0.20/1K items | Query/cursor dependent | Yes | No | Low-cost developer experiments |
| Apify | Marketplace (varies by Actor) | Platform free $5/mo; Actor-specific pricing | Actor-dependent | Yes | Partial (Actor GUI) | Automation workflows, prototyping |
| Netrows | Public-data enrichment API | €99/mo for 10K credits | No archive | Yes (live, uncached) | Partial | B2B enrichment, prototyping |
| PhantomBuster | Session/browser automation | $56/mo (annual effective rate at cutoff) | No guaranteed archive | Scheduled polling / watcher mode | Yes (GUI) | Small-scale lead-gen workflows |
| RapidAPI endpoints | Marketplace (varies by listing) | Free 200 req → $9.99/10K | Varies | Varies | No | Quick API testing, prototyping |
| Phyllo | Cross-platform API infrastructure | Custom quote | Unknown X-specific depth | Webhook/real-time claims | No | Creator/influencer analytics |
| Brandwatch | Enterprise listening (vendor identifies as Official X Partner) | Custom quote | Product- and package-dependent; full-firehose offerings available | Yes, product-dependent | Yes (dashboard) | Brand monitoring, sentiment, research |
| Meltwater | Enterprise media + social intel | Custom quote | 15 months instant social history | Yes | Yes (dashboard) | PR, comms, media intelligence |
| Sprout Social | Social mgmt + Listening add-on | $199/seat/mo (annual) + Listening extra | Sampled; 30-day auto-backfill | Yes (sampled) | Yes (dashboard) | Social teams, publishing + analytics |
| Thunderbit | Browser-based extraction | Free tier available | Page-level only | Page-level | Yes (extension) | Non-technical users extracting visible data |
Now, the details.
1. X (Twitter) Official API
The X Official API is the only first-party source on this list and the only option with a documented complete search archive back to March 2006. If compliance and data provenance are your top priorities, this is where you start.
How it works: Direct REST and streaming endpoints. You register an application, get API keys, authenticate via OAuth, and query endpoints for Posts, Users, likes, lists, and more.
Key features:
- Recent Search covers the previous 7 days (up to 100 Posts per response)
- Full-Archive Search reaches back to March 2006 (up to 500 Posts per response, pay-per-use and Enterprise)
- Filtered Stream delivers matching public Posts near real time (~6–7 seconds P99 latency)
- Complete Post objects, user profiles, engagement metrics, compliance endpoints
Pricing: X moved to prepaid credits with no subscription or minimum commitment. At the August 2026 research cutoff: Post read $0.005, User read $0.010, Like read $0.001. Self-serve accounts are capped at two million Post reads per month. Enterprise pricing is custom. Verify the Developer Console before budgeting because X says prices can change.
Watch out: As of X's May 4, 2026 search-index migration, keyword REST search no longer returns reposts. Filtered Stream is unaffected. This changes corpus composition relative to older tutorials.
Best for: Developers who need official, reproducible raw data. Compliance-sensitive organizations. Anyone who needs the full historical archive.
Not ideal for: Non-technical users, teams on tight budgets, or anyone needing bulk data beyond the self-serve cap.
2. Bright Data
Bright Data offers both a managed X scraping API and pre-built datasets. The reviewed product pages establish current collection and delivery options, but not an official X partnership or a complete historical archive.
Key features:
- Profiles, Posts, URLs, and discovery jobs
- Synchronous jobs (up to 20 URLs) and asynchronous jobs (up to 5,000 URLs)
- Output in JSON, NDJSON, or CSV via API, webhook, cloud storage, or data warehouse
- 5,000 free records per month; PAYG at $1.50 per 1,000 successful records; Scale plan at $499/month including 384,000 records
Compliance note: Bright Data publishes compliance and security controls, but the reviewed pages do not claim an official X authorization or partnership. This is a managed public-web scraper, not an official API reseller. The operational tooling reduces collector maintenance, but it does not remove the buyer's obligations under X terms, privacy law, or downstream-use restrictions.
Best for: Enterprise teams needing large-scale, ongoing X data pipelines and willing to accept the compliance posture of scraping-backed collection.
Bright Data's Dataset vs. Live Scraping
Bright Data offers both a pre-built dataset product and live scraping through its infrastructure. The dataset is useful for one-time research or historical snapshots; live scraping suits continuous monitoring. Choose based on whether you need a static corpus or an ongoing feed.
3. SocialData.tools
SocialData.tools advertises 30+ read-only endpoints for Posts, profiles, search, followers, mentions, replies, quotes, threads, lists, Communities, and Spaces. Its common price is $0.20 per 1,000 items, with some specialist endpoints priced higher.
Key features:
- REST/JSON API with MCP server support
- Pay-per-request, no subscription
- Covers a wide range of X data types
Caveats (and they matter): SocialData.tools' own terms explicitly describe the service as content scraping. Data sources are identified only as "various sources." Customers must warrant that retrieval is permitted. The vendor can suspend service if a target source objects. Their search documentation acknowledges omissions and cursor failures. Despite "Try for Free" language, there's no published free plan — only a possible support-provided demo.
Best for: Developers who want low-cost Post and profile data and are comfortable with the provenance and policy risk after their own diligence.
4. Apify
Apify is a web scraping and automation platform hosting community-built "Actors" — pre-made scrapers that run in the cloud. Multiple independent X Actors exist on the marketplace, each with different pricing, quality, and maintenance schedules.
Key features:
- Platform starts at $5 free usage per month, then $29/$199/$999 per month
- Representative Actor: xquik/x-tweet-scraper — from $0.00015 per delivered row on paid plans; supports Posts, replies, profiles, search, threads, quotes, retweeters, media, and date filters
- Integration with Zapier, Make, webhooks
Critical caveat: Apify's terms state that Community Actors are not reviewed, vetted, endorsed, or guaranteed for quality, security, accuracy, legal compliance, or maintenance. The article must name the specific Actor being evaluated — "Apify costs X" is meaningless without specifying which Actor and billable unit.
Best for: Automation workflows, one-time data collection projects, and developers who want to test multiple scraping approaches in one platform.
Choosing the Right Apify Twitter Actor
Before committing, check the Actor's last-updated date, user ratings, and maintenance status. Some Actors lag weeks behind X changes and can return stale or broken results. The marketplace activity badge helps triage but doesn't guarantee reliability.
5. Netrows
Netrows is a newer B2B public-data enrichment API offering X endpoints alongside LinkedIn data. It lists 26 X endpoint types covering profiles, followers, Posts, replies, quotes, threads, search, lists, Communities, trends, and Spaces.
Key features:
- €99/month for 10K credits (50 req/min), €499 for 100K, €1,499 for 500K
- Live, uncached responses (vendor claim)
- Public status page was operational at the research cutoff; no public X-specific SLA was verified
Caveats: The exact credit cost per endpoint and response schema require an API key to verify. Netrows' own materials say it does not offer a historical archive. Its terms describe automated retrieval from public sources, disclaim affiliation with X, and make users responsible for third-party terms.
Best for: Prototyping and SMB use cases, especially teams that also need LinkedIn data from the same API.
6. PhantomBuster
PhantomBuster is fundamentally a different kind of tool than the APIs above. It's session-based browser automation for prospecting, not a data API or licensed feed. I'm flagging that distinction up front.
PhantomBuster connects to a user's X session through a browser extension and can export visible search results, profile data, and follower information to CSV or JSON. It can run on a schedule or in watcher mode.
Key features:
- No-code GUI designed for sales and marketing teams
- Annual effective rates: $56/mo (Start), $128/mo (Grow), $352/mo (Scale)
- Focused on lead-gen-relevant fields, not full tweet archives
Important risk: PhantomBuster's own documentation warns that exceeding platform limits can produce warnings, restrictions, flags, or bans. This is automation running inside your X account session — not an authorized data feed. Use conservatively and understand the account risk.
Best for: Small-scale sales prospecting workflows where you need to build targeted lists from X profiles and followers.
7. RapidAPI Twitter Endpoints
RapidAPI is a marketplace hosting dozens of third-party X API endpoints from independent developers. It's not a single provider — it's a bazaar.
Representative listing: datahungrybeast/twitter154 ("The Old Bird") — free Basic tier with 200 requests, $9.99 for 10K, $49.99 for 100K, $299.99 for 1M. Exposes user, follower, Post, reply, search, and list endpoints.
Key features:
- Unified REST interface across many third-party APIs
- Free tiers available on some listings
- Quick way to test multiple approaches
The risk is real: This listing had no product website, no API-specific terms of service, and no X authorization, archive guarantee, real-time SLA, or deletion-compliance process. RapidAPI's own docs confirm that publishers control their listings, pricing, versions, and endpoints. Many listed endpoints are abandoned or broken.
How to Avoid Dead Endpoints on RapidAPI
Check response time stats, user ratings, last-updated date, and always test with the free tier before committing to a paid plan. A high subscription count is not an SLA.
Best for: Quick API testing and prototyping. Not recommended as a production data source without thorough provenance and legal review.
8. Phyllo
Phyllo markets normalized public and authenticated social-data APIs across many networks, including X. It describes keyword and competitor mentions, structured Post fetches, creator analytics, asynchronous jobs, polling, and webhook alerts.
Key features:
- Cross-platform API infrastructure (X, Instagram, YouTube, TikTok, etc.)
- Consent-based model for creator-scoped data
- Webhook and real-time marketing claims
Caveats: The reviewed first-party pages do not publish an X-specific latency SLA, completeness percentage, history start date, full-firehose claim, or named X partnership. Pricing is custom (requires contacting sales).
Best for: Developers building creator, influencer, or cross-platform data products who need a normalized API layer — after verifying X-specific coverage details directly with Phyllo.
9. Brandwatch
Brandwatch calls itself an Official X Partner. Its Consumer Research product advertises the full X firehose and public historical Posts; Listen advertises full-firehose access with twelve months of X history.
Key features:
- Deep historical archive (one of the deepest available commercially)
- Sentiment analysis, trend analysis, competitive intelligence dashboards
- Enterprise-grade compliance documentation, GDPR-ready
The export catch: Brandwatch's developer documentation says X Post text and most metadata are stripped from API responses for compliance. To get full Post bodies, you may need to use Post IDs and rehydrate via the X API. This is a crucial distinction between in-product dashboard access and unrestricted raw export.
Pricing: Custom quote. No current numeric list price was published on the reviewed Brandwatch plans page.
Best for: Global brands, research teams, and crisis monitoring operations that need deep X archives and enterprise-grade analytics.
10. Meltwater
Meltwater combines social listening with news and media intelligence. Its product tour establishes fifteen months of instant historical social data including X, plus real-time alerts, Boolean search, dashboards, competitive intelligence, and reporting.
Key features:
- Combined news, PR, and social intelligence in one platform
- Real-time alerts and dashboards
- Exports and Analytics/Streaming APIs (package-dependent)
Caveats: No complete X archive or X-specific latency SLA was established. Coverage can depend on package and region. Meltwater's developer changelog documents X/Twitter field redactions for compliance. Meltwater declares itself a full enterprise partner of X, but independent current certification on an X-owned directory was not confirmed in this research.
Pricing: Custom quote, annual contracts.
Best for: PR and communications teams that need X data alongside news monitoring, media intelligence, and cross-channel social analytics.
11. Sprout Social
Sprout Social is primarily a social media management platform — publishing, engagement, care, approvals, owned analytics, and reporting. Its paid Social Listening add-on can monitor public Posts, replies, mentions, quote reposts, and reposts on X.
Key features:
- Core plans: $199/$299/$399 per seat/month (billed annually); Listening is an extra-cost add-on
- Dashboard-based, designed for social media managers
- X's own business site describes Sprout as an official X partner
The sampling issue: Sprout's help center explicitly says X Listening uses sampled data, not the full firehose. Automatic backfill is 30 days or the last 200,000 messages. Sprout's Public API supports owned-profile analytics, but X Listening Topic data is currently unavailable through that API.
Best for: Social media teams who already use Sprout for publishing and community management and want directional X analytics in the same dashboard. Not the right choice if your primary goal is extracting raw X data at scale.
12. Thunderbit
Thunderbit is an AI web scraping and automation tool available as a Chrome browser extension, designed for business users who need to structure permitted web data without writing code.
How to frame it here: Thunderbit can propose fields, follow compatible pagination or subpages, add prompt-based enrichment, and export to Excel, CSV, Google Sheets, Airtable, or Notion. Those are general web-data capabilities. They do not establish an authorized X feed, complete X coverage, or permission to automate X.
Key features:
- No-code field suggestion and structured export
- AI-assisted field detection adapts to layout changes
- Plans and usage limits should be verified on thunderbit.com/pricing
- Suitable only for compatible pages and uses where the user has permission to automate
What Thunderbit is not: It is not an official X data provider, a bulk X API, a firehose replacement, or a prepackaged X dataset. X's current terms prohibit browser scraping without prior written consent. For X data, start with the official API, a documented authorized provider, or your own account archive. Use Thunderbit for compatible, permitted web-data workflows—not as a workaround for X API access or pricing.
For more on how web data extraction works in practice, our blog covers what web scraping is, how to do it without coding, and practical approaches to Twitter data collection.
Which Twitter Data Provider Fits Your Budget and Use Case?
No single provider wins for everyone. Here's a decision framework organized by use case and budget tier:
| Use Case | Budget Pick | Mid-Range Pick | Enterprise Pick |
|---|---|---|---|
| Academic research / archiving | X Official API (pay-per-use) + twarc | Licensed data partner | X Enterprise / contracted licensed partner |
| Brand monitoring / sentiment | Official API with a bounded query | Sprout Social (sampled Listening add-on) | Brandwatch or Meltwater after contract review |
| Lead gen / sales prospecting | Consent-based/manual research | Permitted automation after policy review | Authorized provider + governed CRM pipeline |
| Quick one-off extraction (no code) | Account archive or authorized export | Permitted compatible-page workflow | Enterprise dashboard/export |
| Influencer / creator analytics | Phyllo (after X coverage review) | Phyllo | Brandwatch |
| NLP / data science pipeline | X Official API (pay-per-use) | Authorized provider with export rights | X Enterprise / licensed feed |
A few patterns jump out:
- If compliance and provenance are non-negotiable, the X Official API is your baseline. Everything else involves some degree of trade-off.
- If you need deep historical archives and enterprise dashboards, Brandwatch and Meltwater are the established players — but expect custom contracts and redacted exports.
- If you're a non-technical user, prefer an authorized source with CSV or Sheets export. Use browser automation only on compatible pages where permission is clear; it is not a substitute for a production data pipeline.
A Quick Note on Compliance: ToS, GDPR, and What to Ask Your Legal Team
This section is operational guidance, not legal advice.
X's Terms of Service prohibit crawling, scraping, or scripting the website without prior written consent. X's Developer Policy adds requirements around privacy, transparency, user control, deletion compliance, and restrictions on sensitive inference, matching, redistribution, and model training. Technical availability does not establish permission.
GDPR, CCPA, and other privacy rules may apply when collecting handles, locations, profile bios, or Post text. Public visibility does not by itself settle platform authorization, privacy, contract, copyright, computer-access law, or downstream-use rights. Apply data minimization and retention controls, and obtain legal review for the actual jurisdiction and use case.
Three practical rules:
- Stick to public data you have a legitimate purpose to collect.
- Don't over-store — set retention limits and honor deletion requests.
- Ask your legal team before building anything at scale.
Questions to ask any vendor: Are you an X Official Partner? Which exact methods supply the product? How are deleted Posts and protected accounts handled? Can you retain full text, or only IDs and aggregates?
Conclusion
The old free-access model is gone. These twelve entries span first-party APIs, managed public-web services, enterprise listening suites, marketplaces, and no-code tools; they do not offer equivalent authorization, completeness, or risk.
The right choice depends on the job. For raw, reproducible data with first-party provenance, start with the X API. For enterprise dashboards and governed listening, evaluate Brandwatch, Meltwater, or Sprout against your contract and export needs. Treat independent collectors and marketplace listings as higher-risk options that require live testing, source-method disclosure, and legal review. Use Thunderbit for compatible, permitted web-data workflows—not as an X feed.
X's data landscape keeps shifting. Verify prices live, read the relevant terms, document your data source and query, and ask legal and security reviewers before building a high-volume or sensitive workflow.
Learn more:
- How to Do Twitter Sentiment Analysis for Better Decisions
- How to Master Twitter Data Collection Using Thunderbit
- What Is a Social Media Scraping Tool and How Does It Work?
- Is Web Scraping Illegal? Understanding the Legal Implications
- Web Scraping Without Coding
FAQs
1. What happened to free Twitter API access?
X eliminated the free read tier in early 2023. The current model is prepaid, pay-per-usage credits — a standard Post read costs $0.005, and self-serve accounts are capped at two million Post reads per month. The Academic Research track was also discontinued, leaving researchers without the free bulk access they once had.
2. How do third-party Twitter data providers get their data?
They fall into roughly three categories: (1) official API wrappers that use your own X API key and are subject to X rate limits; (2) scraping-backed APIs that collect data via automated browsers or reverse-engineered endpoints — cheaper but carrying ToS and continuity risk; and (3) licensed social listening platforms like Brandwatch and Meltwater that hold enterprise data agreements with X. Browser-based tools like Thunderbit are a fourth category — they extract what's visible on the page in your own browser session.
3. Which Twitter data provider is cheapest for getting started?
At the August 2026 cutoff, the X Official API listed $0.005 per standard Post read with no minimum commitment. Bright Data advertised 5,000 free records per month, Apify included $5 of monthly platform usage, the named RapidAPI listing offered 200 free requests, and SocialData.tools commonly charged about $0.20 per 1,000 items. These units are not equivalent, and the cheapest authorized, complete option for one use case may be different from the cheapest raw request. Verify every live price before purchase.
4. Can I get historical tweets without paying for X Enterprise API?
Yes. The X Official API's Full-Archive Search reaches back to March 2006 on pay-per-use and Enterprise access, so Enterprise is not the only official route. Brandwatch advertises deep historical research, with availability depending on product and package. Independent collectors may expose date windows or vendor-retained datasets, but their reach and completeness are not equivalent to X's documented full archive.
5. Is there a no-code way to get Twitter data?
Yes. Sprout Social and Brandwatch provide dashboard-based workflows without coding, and authorized providers may offer CSV or spreadsheet exports. Thunderbit can structure data on compatible pages where you have permission to automate; it is not an official X feed. PhantomBuster offers session-based automation but documents account and rate-limit risk. Confirm coverage, authorization, sampling, and export rights before choosing any no-code route.
