Free Twitter API (X API) in 2026: What Is Free and What Is Not

Type “free Twitter API” into Google and you will find pages that still describe a free tier. That tier is gone. Since February 2026, X sells its API by usage, and every post you read costs a fraction of a cent.

We read X's developer docs, its Terms of Service and each vendor's pricing page on 2026-10-23, then did the maths from them. Nothing here comes from running anyone's product, so there are no speed or success-rate numbers. Where a page printed no figure we write “not verified.” Serpent sells one of the options below, so every calculation is shown; its post lookup by id is $0.10 per 1,000.

The options at a glance (read 2026-10-23)

One row per option. Each paid rate is in the unit that vendor bills, so read the second column before you compare the fourth. Rates are for one post read or lookup unless the row says otherwise.

What X's official API costs now

X's pricing page opens with one sentence: “The X API uses pay-per-usage pricing with no subscriptions.” You buy credits in the Developer Console, and each request spends some. It also says there are “no contracts, subscriptions, or minimum spend.”

The page does not mention Free, Basic or Pro plans at all. It lists these read prices, each per resource returned:

• Posts: $0.005 each, which is $5.00 per 1,000. Lists, Spaces, Communities and Notes are the same.
• Users: $0.010 each, or $10.00 per 1,000. Followers and following lists are the same per user.
• Likes, mutes and blocks: $0.001 each.
• Writes: $0.015 to create a post, or $0.200 if the post contains a URL.

Three details matter for a budget. Pay-per-usage is capped at 3 million post reads per monthly billing cycle, and the page says to upgrade to Enterprise for more. The same resource requested more than once in a 24-hour UTC day is charged once, a “soft guarantee.” And you can set a maximum spend per cycle, after which requests stop.

What changed, and when

X's changelog gives the dates. On 2026-02-06 X launched pay-per-use pricing. In the same entry, Public Utility Apps kept free scaled access, recently active Legacy Free users received a one-time $10 voucher, and Basic and Pro plans “remain available.” Existing subscribers can opt in to pay-per-use.

On 2026-04-16 X also announced “Owned Reads” at $0.001 per resource, valid from 2026-04-20 and only for your own data, such as your bookmarks or followers. On 2026-10-02 it added the free credits covered next.

What Basic, Pro and Enterprise buy

We found no current official price for Basic or Pro, and we will not quote one from a third-party blog. The pricing page we read lists only pay-per-usage prices and names no plans, and the changelog says only that Basic and Pro remain available.

Enterprise is the top step. X's rate limits page says Enterprise customers have custom rate limits, and no price is printed on the pages we read.

Rate limits and billing are separate on X. Post lookup by id allows 450 requests per 15 minutes for an app token and 900 per user token. Past the limit you get a 429 until the window resets, and every successful read is still billed.

How far the free credits go

X's free credits page describes two one-time incentives for pay-per-use accounts:

• $20 after your first purchase. The page says “Adding a card on its own no longer qualifies (updated October 13, 2026).” The card must be your only payment method, the account must have no earlier successful payments, and prepaid cards do not count.
• A match on your first auto-recharge, up to $50. A $20 first recharge earns $20, and a $100 first recharge earns $50. Turning auto-recharge on earns nothing until a charge succeeds.

Together that is $70 for a new account. The credits expire three months after they are granted and are spent before purchased credits. The page also says the incentives are “currently rolling out,” so an account that has not been enabled yet gets no credit.

Here is the arithmetic. At $0.005 a post read, $20 buys 4,000 reads and $70 buys 14,000. That is enough to try an idea, but it is not a standing free tier: you must pay first and the credit lapses.

Older articles still describe a free read allowance, often 1,500 posts a month. We found no current official page that describes one, so do not plan around it.

The rules on X’s own API

This is not legal advice. We quote short lines so you can read the originals.

• Developer Policy scope. The Developer Policy says the Free, Basic and Pro plans are “designed for hobbyists, commercial prototyping, early-stage X product integrations, and supporting applications with limited end-users.” Use beyond that scope needs an Enterprise plan.
• Limits. The same policy says “You may not exceed or circumvent rate limits.”

Before you build on any data source, read X’s Terms of Service and Developer Policy and ask your own counsel.

The near-free X data APIs, one by one

These return X post or profile data as JSON over a plain HTTP call. Each entry names the unit it bills. Serpent comes first because, for a post lookup by id, its rate is the lowest printed rate we read. The comparison table above lets you check that order yourself.

1. Serpent API

• Free: A free account gets 10 free calls, shared across services, and our X docs say both X endpoints can spend them. A hydrated profile call costs one free call, not one and a half. You also get one free demo run a day on the X API page without signing up. That is small: X's $20 credit buys 4,000 post reads and twitterapi.io's $1 about 6,700 tweets.
• Paid: Billed per call, not per item. A post lookup is $0.10 per 1,000 calls on Default, $0.09 after one $100 deposit (Growth) and $0.07 after a $500 or $1,000 deposit (Scale and Enterprise). A profile call is $0.10 per 1,000 at every tier, and hydrate=true makes it $0.15 at every tier. The pricing page has every row.
• Commitment: The first deposit is $10 minimum, which buys 100,000 Default calls by our arithmetic. There is no subscription, and deposited credits do not expire. Promotional bonus credits do, after 30 days.
• What you get: Two endpoints. GET /api/x/tweet takes a numeric id or a permalink url and returns the text, language, created time, author (with a verified flag), favorites and replies counts, media, hashtags, mentions, expanded links, a quoted post and the reply target. GET /api/x/profile takes a handle or profile url and returns name, bio, avatar, website, join date, followers, following and post counts and the recent posts the profile lists. With hydrate=true each recent post also gets its exact text, counts and media. If hydration enriches no post, the extra half-rate is refunded automatically. Every key is always present. The API docs list every field.
• Watch out: It is a lookup, not a search. There is no tweet search, no full timeline, no followers or following list, and no retweet, quote or view counts; only favorites and replies. A profile carries no verified flag. A handle or post that does not exist returns 404, and that call is still charged. The shared batch job route is currently unavailable, so each lookup is its own call. Default allows 100 requests a minute and 10,000 a day.

Cheapest? For a post lookup by id, yes among the printed rates we read: $0.10 on Default is below twitterapi.io's $0.15, SocialData's $0.20 and X's $5.00. Two caveats. Apify's igolaizola actor prints “from $0.13,” and ScrapeCreators, Scrapingdog and others do not print a cost per X request, so we could not compare them. For a plain profile lookup Serpent is not the lowest: twitterapi.io lists users from $0.01 per 1,000 in batches of 100 or more.

2. twitterapi.io

twitterapi.io bills per item returned, at $1 for 100,000 credits. Tweets are 15 credits each, or $0.15 per 1,000. A user profile is 18 credits in a batch of 1 to 99, which is $0.18 per 1,000, and 1 credit each in batches of 100 or more, which is $0.01 per 1,000. The minimum charge per call is $0.00015, and its page lists follower lists and other endpoints too.

New accounts get up to $1 of free credit with no card, after a bot check, once per person or network. It expires after 31 days. Recharged credits never expire, and no minimum spend is required.

3. SocialData

SocialData prints a single standard rate: “$0.20 per 1000 results,” per tweet or user returned, not per request. Its page lists search, post by id, user profile, a user's posts and follower paging, with more beyond those. It has no monthly minimum, balances do not expire, and failed requests do not use balance.

It does not offer a free plan, though its FAQ says support may set up a free demo account for qualifying use. Some specialised endpoints are priced differently, and we could not open its per-endpoint reference.

4. ScrapeCreators

ScrapeCreators has a free tier of 100 credits, plus up to 7,000 bonus credits you can claim with some small tasks. Paid packs are $47 for 25,000 credits ($1.88 per 1,000) and $497 for 500,000 ($0.99 per 1,000), and credits never expire. Its X page lists seven endpoints, including search, profile, user tweets and tweet details. It says most endpoints cost one credit but not what an X request costs, so we left it out of the cost tables.

5. Apify actors

Apify is a marketplace, so the price depends on the actor. The theagents Tweet Scraper lists $0.25 down to $0.15 per 1,000 tweets depending on your Apify plan, plus a $0.005 fee per search query. Fetching one tweet by id is $0.015 per query, or $0.009 on the top tier. We did not confirm whether a per-tweet charge is added.

The apidojo Tweet Scraper starts at $0.40 per 1,000 tweets, needs at least 50 tweets per query and does not allow single-tweet fetching. The igolaizola actor shows “from $0.13” in its header and “$0.15 per 1000 results” in its title, searches or reads a username, and says tweet-URL collection is not an input mode. On the free plan the first two actors allow 5 runs a month of up to 10 items, and neither runs through the API.

6. Bright Data and Scrapingdog

Bright Data gives 5,000 records a month free with no card. Free credits renew on the 1st and do not roll over. After that it is $1.50 per 1,000 records pay as you go, or $499 a month for 384,000 records and then $1.30 per 1,000. Scrapingdog gives 100 free credits and documents one X endpoint that takes a tweet URL, but the page does not say how many credits a request uses, so we did not price it.

Data365 offers a 14-day trial by request form and an intro call, with no prices on its X page. For a longer vendor-by-vendor view, see our best X API providers guide and the X API cost comparison.

What free really costs you

Credits that lapse and rules you must pass

X's free credits arrive after a first purchase, need a qualifying card and lapse in three months. The Developer Console also asks you to describe your intended use. For a one-off test that is fine. For a product you will run for years, price it at the paid rate from day one.

Maintenance and terms

Per-item billing means one profile with its five latest posts is six charges. Free trials and marketplace actors come with little support, and an actor can change its price or input rules, so put a thin adapter between your code and any vendor.

Cost at 1K, 10K, 100K and 1M lookups

This is arithmetic from the published rates above, per month, for one post lookup each. We ignored free allowances, tax and extras. X's 24-hour dedupe would lower its row if you re-read the same posts, and the 3 million reads a month cap does not bite at these volumes.

How to read it

• The Serpent tier rows assume you made the single qualifying deposit for that tier. Credits do not expire.
• Bright Data's $1,299.80 is $499 plus 616,000 records at $1.30 per 1,000, and assumes one tweet is one record, which its page does not define. At the three smaller volumes pay as you go is cheaper than the plan.
• The Apify row is the no-discount single-tweet query fee; the top tier is $0.009. A bulk search is cheaper per tweet, but that is a search, not a lookup.
• Volume meets rate limits. Serpent Default allows 10,000 calls a day, so a million lookups a month, about 33,000 a day, needs the Growth band (50,000 a day) or $500 in total deposits.

The same maths for a profile with its recent posts

The unit decides this one. X and most vendors bill each item returned, while Serpent bills one hydrated profile call. This table assumes the profile shows five recent posts. Five is our assumption, not a promise about any account.

If you only need the profile counts, a plain Serpent profile call is $0.10 per 1,000 and a batched twitterapi.io user lookup is $0.01, so on that narrow job twitterapi.io is cheaper. Add hydrate=true only for the handles that pass your first filter.

How to choose

• You need to post, reply or read your own account. Use the official X API. Posting is not something a read-only data API does, and owned reads are $0.001 each.
• You need a post by id or URL, or a profile snapshot, at the lowest price. Serpent is the lowest printed post-lookup rate we read, from $0.10 per 1,000 on Default.
• You need search, timelines, followers or communities. Serpent does not have them. Compare twitterapi.io, SocialData and ScrapeCreators, whose pages list search and follower endpoints, and check their terms and prices for the exact endpoint.
• You want the biggest free trial. Bright Data's 5,000 records or X's $20 credit (4,000 post reads, after a first purchase) go further than our 10 calls.
• You need many platforms on one key. See our social media API index, the free Instagram API guide, the free TikTok API guide and the free LinkedIn API guide.
• You want worked examples. See monitoring a known post and auditing a public account.

Quick start and migration

A lookup is a single GET request. These calls come from the API docs:

# A single post by its numeric id
curl "https://apiserpent.com/api/x/tweet?id=1234567890123456789" \
  -H "X-API-Key: YOUR_API_KEY"

# A public profile; hydrate=true adds counts and media to each recent post
curl "https://apiserpent.com/api/x/profile?handle=example&hydrate=true" \
  -H "X-API-Key: YOUR_API_KEY"

And a post by its permalink in Python:

import os, requests

r = requests.get("https://apiserpent.com/api/x/tweet",
                 params={"url": "https://x.com/example/status/1234567890123456789"},
                 headers={"X-API-Key": os.environ["SERPENT_API_KEY"]},
                 timeout=60)

if r.status_code == 200:
    t = r.json()["tweet"]
    print(t["author"]["screen_name"], t["counts"]["favorites"], t["text"][:80])

Try it in the playground first if you want to see the JSON.

What to know when you move from another vendor

• Same idea, different names. You send an id, a permalink or a handle and get JSON back. Field names differ by vendor, so map them in one adapter file.
• Check keys before you read them. Every key in a post or profile is always present, but counts and media on a profile's recent posts appear only on posts that hydration filled in. Use .get() there.
• Engagement is two numbers. Favorites and replies are the only counts. If your dashboard needs retweets or views, this API does not return them.
• Not found is charged. A deleted post or an unknown handle returns 404 and still bills. A bad parameter returns 400 and does not.

Limits and unknowns

• We did not test anything. Every competitor figure is from its published page on 2026-10-23. Prices and allowances change, so treat the numbers as dated.
• Basic and Pro. The X pricing page we read prints no price for either plan, so we quote none. The changelog says they remain available.
• Pages we could not read. SocialData's per-endpoint pricing reference, the credit cost of an X request at ScrapeCreators and Scrapingdog, and any RapidAPI listing.
• Free credits. X says its incentives are still rolling out, so your account may not see the $20. We could not confirm each vendor's free allowance applies to its X endpoints in every account.
• Data quality. We did not compare the fields each vendor returns for the same posts, so a low rate does not prove equal coverage.
• Our interest. Serpent sells one of these options and we wrote the post, so redo any sum yourself from the linked pages.

FAQ

Is the Twitter (X) API free?

Not for reading at any real volume. X's pricing page, which we read on 2026-10-23, lists only pay-per-usage prices, such as $0.005 per post read, and no free read tier. New paying accounts can earn up to $70 in one-time credits that expire after three months, and Public Utility Apps keep free scaled access.

Is there still a Twitter API free tier?

We found no current official page that describes one. X's February 2026 changelog says recently active Legacy Free users got a one-time $10 voucher, and the pricing page lists no free plan. Older articles that quote 1,500 posts a month describe the pre-2026 plan, so check the Developer Console before relying on them.

How much does the X API cost in 2026?

X bills per resource returned: $0.005 per post read and $0.010 per user read, with writes from $0.015 per post. Pay-per-usage is capped at 3 million post reads a month before you need Enterprise. A profile plus five posts is $0.035, or $35 per 1,000 profiles, by our arithmetic.

What is the cheapest alternative to the X API?

For looking up one post by id, Serpent's $0.10 per 1,000 calls is the lowest printed rate we read, against $0.15 at twitterapi.io, $0.20 at SocialData and $5.00 on X itself. For profiles, twitterapi.io lists $0.01 per 1,000 in batches of 100 or more. Units differ, and Serpent has no search.

Can I search tweets or get a timeline for free?

Not through a free official route that we found, since X reads are billed per resource. Serpent does not offer tweet search, full timelines or follower lists, only a post by id and a public profile with recent posts. twitterapi.io, SocialData and ScrapeCreators list search endpoints, so compare those if you need it.

Do I need to check X's terms before using a data API?

Yes. X publishes Terms of Service and a Developer Policy for its own API. Whether a given use is lawful depends on your facts and where you are, so read the terms and ask your own counsel before you build. This is not legal advice.

Sources read on 2026-10-23

• X docs: API pricing
• X docs: changelog
• X docs: free credit incentives
• X docs: rate limits
• X docs: getting access
• X Developer Policy
• X Terms of Service
• twitterapi.io: pricing
• SocialData
• ScrapeCreators
• ScrapeCreators: Twitter API
• Apify: Tweet Scraper (theagents)
• Apify: Tweet Scraper (apidojo)
• Apify: X Twitter Scraper (igolaizola)
• Bright Data: Twitter scraper
• Scrapingdog: Twitter scraper API
• Data365: Twitter data

Related reading on Serpent: best X API providers, X API comparison, SERP APIs with no minimum and a free tier.