# Twitter (X) scraper API

What a Twitter scraper API gives you, where it breaks, and how to get public X profiles, posts, and replies from one API instead.

Checked September 2026. Public pricing pages. No affiliate links.

The official X API still exists, but read access now costs hundreds of dollars a month before you read a single post, and the caps are tight. That price is why people search for a Twitter scraper API. This page explains the options and what each one really costs.

Openhandle reads public X data on request and returns it in one schema. 15 X endpoints cover profiles, posts, replies, media, followers, lists, and search.

## At a glance

| Criteria | Build a scraper | Rent a scraper | Openhandle |
|---|---|---|---|
| Time to first call | Days to weeks. Proxies, parsers, and blocks come first | Minutes, once you pick a listing | Seconds. No approval wait |
| Who maintains scrapers | You, every time the platform changes | The listing author, on their own schedule | Openhandle handles platform changes |
| Response shape | Whatever you parse out of the page | Different per listing and per platform | One schema across Instagram, TikTok, X, and Reddit |
| Pricing model | Proxies, compute, and engineer time | Monthly tiers with hard caps and overage fees | Pay per request from a balance. No credits, no subscription |
| Failed requests | Your retries and your on-call | Often billed anyway | Failed requests are free. You pay only when Openhandle returns an answer |
| Support | You are the support | A ticket queue or a Discord, if any | Direct support from our team through Crisp and email |
| Uptime | You find out from your users | No SLA unless the listing says so | 99.9% uptime |

Openhandle reads public data on request and returns it in one schema. It is not a scraper you run. Read the public data notice for what we never do.

## Why the official API pushes people away

X sells API access in monthly tiers. The free tier is write-only. Paid read tiers start at a few hundred dollars a month with a fixed post cap, and the next tier jumps by an order of magnitude. Teams that only need a few thousand public reads a month pay for a plan they never use. Scrapers fill the gap by reading the public web interface, and X changes that interface often.

## What a Twitter scraper is used for

- **Brand monitoring.** Posts and replies that mention a brand, a product, or a handle.
- **Social listening.** Search by keyword and read the conversation threads underneath.
- **Profile tracking.** Follower counts, bio, and recent posts for a set of accounts over time.
- **Creator reads.** Views, likes, reposts, and quotes on a creator's recent posts.
- **List and follower graphs.** Who an account follows, who follows it, and the lists it belongs to.

## Where Twitter scrapers break

- **Login walls.** X hides most content from logged-out visitors. Scrapers use pooled accounts, and those accounts get suspended in waves.
- **Guest tokens.** The public web interface needs a guest token that X rotates and rate-limits. When it changes, every scraper built on it fails at once.
- **Protected and suspended accounts.** Many scrapers return an empty success. You cannot tell "protected" from "no posts".
- **View counts.** X shows views only in some places. A scraper that fills in 0 breaks every reach number you compute.
- **Long threads.** Replies load in pages with cursors that expire. Half a thread is a common result.

## What Openhandle does instead

- **Public reads on request.** You ask for a public profile, post, or search. We read it and answer. A protected account returns a clear error.
- **Null means hidden.** A view count X does not show comes back as null, never as 0.
- **Stable pagination.** Replies, followers, and search pages come back with cursors that keep working.
- **Media that stays up.** Images and videos are served from our CDN. The link you store keeps working.
- **One schema with Instagram and TikTok.** A post is a post on every platform. One parser, three networks.
- **Pay afterwards.** $0.0025 per request, down to $0.002 at volume. 30-day cache hits are free. Failed requests are free. You pay only when Openhandle returns an answer. No monthly plan.

## What we never do

Openhandle is read-only. We never post, like, repost, follow, or message on any account. Protected accounts, direct messages, and anything a logged-out visitor cannot see are out of scope. Account owners can ask us to block their profile. The [public data notice](/legal/public-data-notice) lists every limit.

## Start in a minute

Create a key and make your first X call. You get 100 free live requests and no card is needed. See every endpoint on the [X (Twitter) API](/apis/twitter) page, and run a request from the [quickstart](/docs/quickstart) in cURL, TypeScript, Go, or Python.

## How we checked

Official X API tiers come from X's public developer pricing page. Scraper failure modes come from public listings and from customers who moved to Openhandle. Openhandle endpoint counts and rates come from our own catalog and pricing page and update with them.