The Twitter Search API at $0.04 per 1,000 Tweets

Why reach for a Twitter search API

Typing a query into the X web app is fine for a one-off. The trouble starts the moment search has to live inside a product, a dashboard, an alert pipeline, or an AI agent, where you need it as a callable endpoint. The official X API does ship search, and it is the sanctioned channel, yet the price and the gatekeeping shut a lot of teams out:

• Price plus a standing monthly floor. Sanctioned search reads land around $5 per 1,000 inside a paid tier, so even a light monitoring job owes a recurring monthly minimum before a single tweet comes back.
• Sign-off and tier ceilings. First you request a developer account and wait on review, then you meet hard read ceilings and a short lookback on the lower tiers. The throttle decides your ceiling, not your code.
• Roll your own and watch it crack. Free libraries that scrape X internals collect IP bans and snap the next time the site shifts, so you swap a monthly invoice for proxy churn and endless patching.

TwitterAPIs sits in the managed middle. One Bearer key, one documented search endpoint, structured JSON in return, and $0.04 per 1,000 tweets with no monthly floor and one flat 600 req/min ceiling. You send a query, matching tweets come back, and none of the plumbing underneath is yours to babysit.

Inside the search endpoint

Everything routes through a single endpoint, GET /twitter/tweet/advanced_search. Put your query in the query parameter, set a product ranking, and step through pages using cursor. A call costs $0.0008 and brings back about 20 tweets.

One request, matching tweets

curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.twitterapis.com/twitter/tweet/advanced_search?query=from:elonmusk&product=Latest"

What comes back is structured JSON plus a cursor for the next page, so there is no markup to scrape and no proxy pool to keep rotating. See the advanced search endpoint docs.

The official X API: pay-per-use rates

X moved to consumption billing in February 2026, so the comparison is now rate against rate rather than plan against plan. Post reads bill at $0.005 each and user reads at $0.010, which puts a thousand tweets at roughly $5 before any profile lookups, against $0.04 here. Writes bill separately, at $0.015 a post and $0.20 when the post carries a URL. The ceiling matters as much as the rate: pay-per-use caps post reads at 3 million a month, and the only route past it is an Enterprise contract with no published price and a procurement cycle attached. There is still a free tier in name, but it is write-centric with roughly 100 reads a month, which is not enough to collect anything. Full workings and every scenario sit on the Twitter API pricing page.

The legacy Basic and Pro tiers

Anyone who priced the X API before 2026 is remembering two plans that no longer work the way they did. Basic ran at $200 a month for 50,000 posts and 15,000 reads; Pro ran at $5,000 a month for 300,000 posts, a million reads, full-archive search and the filtered stream. Both stopped taking new signups at the February 2026 cutover, and Basic went further: X deprecated it outright, monthly and annual alike, and moved remaining subscribers onto pay-per-use from 1 June 2026 at the end of their billing cycle. So being grandfathered in is not the situation any more, and a bill that used to be a fixed $200 is now whatever the call volume prices out at. Pro is a narrower case, closed to new signups, and X has not said publicly whether it faces the same migration, so treat its future as open rather than settled. Search here has no tiers to age out of: one flat rate per call, the same on your first request and your millionth.

Flat, readable pricing

A search call costs $0.0008 and returns close to 20 tweets, so 1,000 tweets settles around $0.04. The full breakdown sits on the Twitter API pricing page, and you can project a monthly bill with the Twitter API cost calculator.

Go deeper on any single tweet

Search hands you matching tweets and their IDs. Once you hold a tweet ID, three read endpoints expand it into the full picture, all on the same Bearer key at $0.0008 per call. Pass the ID you already have, and structured JSON comes back with no extra auth step.

A common pattern is one search call to find the tweets that matter, then a detail, replies, or retweeters call on each ID to expand the ones worth a closer look. Both steps bill at the same flat rate, so a deep read on a handful of tweets stays well under a cent.

Read what is trending, worldwide or by location

Search answers what people are posting about a query. Trends answers the other direction: what X itself is surfacing right now. Two read endpoints cover it at $0.0008 per call. Call trends locations first to resolve a city or country to its WOEID, then pass that WOEID to trends for the live list in that place.

# 1. Find the WOEID for a place
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.twitterapis.com/twitter/trends/locations"

# 2. Read the live trend list for that WOEID (1 = worldwide)
curl -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.twitterapis.com/twitter/trends?woeid=1"

That pairing drives trend dashboards, timing calls for a launch, and regional monitoring where a topic breaks in one country before it travels. No developer account, one 600 req/min ceiling, same key as every other read.

When search alone is not enough

This is a single endpoint. Need entire timelines, follower lists, and profiles on top of it? The broader Twitter scraper reaches that wider surface on the very same key. Wiring search into an AI agent instead? Point it at the MCP server. Still weighing vendors? Line up the full slate of Twitter API alternatives and compare them head to head.