Apify pricing in 2026

What $19 buys on each

On Starter, $19 of usage buys about 8,260 Instagram Scraper results at $2.30 per 1,000, or 7,600 Google result pages at $2.50 per 1,000. On SpecStream $19 reads 23,750 pages or runs 19,000 fast searches, and the credit stays until you spend it.

Usage that expires raises the price

Apify usage that is not spent disappears at the end of the billing cycle. A Starter month that uses $9.50 of its $19 pays twice the listed rate. SpecStream credit you buy never expires, and a top-up of $1,000 or more adds 20% more credit.

A compute unit has no fixed number of pages

A pay per usage Actor bills compute units, proxy and storage, so a run costs more with more memory, longer runs and more retries. Apify's own tests put Website Content Crawler at $0.5 to $5 per 1,000 pages with a headless browser. SpecStream charges $0.0008 for each page read with text.

What three popular Actors cost

Tweet Scraper V2 costs $0.40 per 1,000 tweets on every plan, with rules of its own: each query must return at least 50 tweets, one run goes at a time with up to 5 batched queries, and the same query must not run again in short intervals. On the Free plan it allows 5 runs a month of 10 items each. Apify's Instagram Scraper costs $2.70 per 1,000 results on Free, $2.30 on Starter, $1.90 on Scale and $1.50 on Business. The Amazon Bestsellers Scraper costs $5.90, $5.00, $4.10 and $3.20 per 1,000 items on the same four plans.

So the price of a result moves with the Actor and the plan. On SpecStream a page read costs $0.80 per 1,000 and a row you describe 5¢, 15¢ or $1, the same on every plan and every site. How the Instagram and Amazon Actors compare with a request in plain words is on the Instagram profiles dataset and Amazon reviews dataset pages.

The call, before and after

# Before: an Actor run, then its dataset
from apify_client import ApifyClient

client = ApifyClient(APIFY_TOKEN)
run = client.actor("apify/website-content-crawler").call(
    run_input={"startUrls": [{"url": "https://docs.apify.com/"}], "maxCrawlPages": 10}
)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["url"], item["text"][:300])

# After: one call to the SpecStream Extract API
import requests

reply = requests.post(
    "https://api.specstream.ai/v1/extract",
    headers={"Authorization": f"Bearer {SPECSTREAM_KEY}"},
    json={"urls": ["https://docs.apify.com/"], "format": "markdown"},
    timeout=60,
).json()["data"]
for page in reply["results"]:
    print(page["url"], page["status"], (page.get("text") or "")[:300])

There is no run to wait for and no dataset to read: the reply holds the main text of up to 20 URLs. More on the Extract API page.

Scored on the official benchmarks

The SpecStream Search API scores 97.9 F-score on all 1,000 SimpleQA Verified questions and 64.0 item F1 on all 200 WideSearch tasks, and the Extract API scores 0.908 F1 on WCXB. Every score and its method is on the benchmarks page.

Bright Data

Scraper API at $1.50 per 1,000 records pay as you go and $1.30 beyond the volume of its Scale plan, with 5,000 free credits a month. SpecStream returns structured rows for any request you describe, at 5¢, 15¢ or $1 a row. How the two compare is on the Bright Data alternatives page.

Firecrawl

Hobby at $19 a month for 5,000 credits, and 1 credit is 1 page on a basic scrape. SpecStream answers Firecrawl's scrape call at https://api.specstream.ai/firecrawl for $0.80 per 1,000 pages.

ZenRows

Build at $19 a month for 45,000 credits, and a page rendered with JavaScript costs 5 credits. SpecStream answers ZenRows' call at https://api.specstream.ai/zenrows/v1/ for $0.80 per 1,000 pages.