upAPI vs ScraperAPI
A catalog of operations, not a credit balance against one endpoint.
TL;DR: Is ScraperAPI or upAPI the better fit?
Pick ScraperAPI when the job is unblocking pages you choose yourself, at high sustained volume, with per-request control over rendering and proxies. Pick upAPI when you want structured JSON from named sources, such as Maps, search suggestions or page metadata, on one key with a free plan and a published unit cost for every call.
How ScraperAPI charges
Six credit tiers running from $49 to $1,975 a month plus an Enterprise band, with pay-as-you-go overage only unlocked once you are on the $475 plan. A trial of 7 days and 5,000 credits, no card.
How upAPI charges
Free at $0 with 1,000 weighted units, then $29 for 100,000. The unit price is printed beside the plan price — $0.29 per 1,000 on Starter — and there is no threshold you have to buy your way past before overage behaves sensibly.
USD 29 / mo · USD 0.29 per 1,000 units · full pricing
Last checked 2026-10-05. ScraperAPI figures here were read from their own pages on that date. We do not restate a competitor’s number we have not seen ourselves, and where they publish none we say so; if this date has aged, check theirs before ours.
ScraperAPI and upAPI side by side
| Feature | ScraperAPI | upAPI |
|---|---|---|
| Pricing | Monthly credit plans from $49 (100,000 credits) up to $1,975, then custom Enterprise. A normal request is 1 credit, a request with render=true is 10, and a Google or Bing request is 25. Pay-as-you-go is available from the $475 Scaling plan. | Free at $0, then Starter $29 for 100,000 weighted units a month, Pro $99 for 500,000 and Scale $499 for 5,000,000. Each operation declares its own unit cost, and the quota is pooled across the catalog. |
| Free tier | A 7-day trial with 5,000 credits and no credit card, per their pricing page. | The Free plan: 1,000 weighted units a month across every public operation, with no trial clock. |
| Endpoints | One scraping endpoint that takes any URL, a proxy-port mode and an async API for batch jobs, plus structured endpoints for Google (Search, News, Jobs, Shopping, Maps). | Named operations, each with its own request and response schema: Maps, search suggestions, page metadata, sitemaps, screenshots, PDF and more. There is no generic fetch-any-URL-as-HTML endpoint. |
| Output format | Raw HTML by default; markdown or text with output_format, and JSON or CSV from the structured endpoints. | JSON on every operation, shaped by its published output schema. The page-to-Markdown operation returns Markdown. |
| JavaScript rendering | Yes, with render=true on any request, at 10 credits instead of 1. | Only inside operations built on a browser: screenshot and HTML-to-PDF. Other operations have no render flag, and the page-to-Markdown operation reports a client-rendered page as an error rather than returning it empty. |
| Proxies | Rotating pools on every plan, premium proxies at 10 credits and ultra-premium at 30. Country targeting costs nothing extra, but Hobby and Startup are limited to US and EU regions. | Included in the unit price. The exit is chosen per operation, and many operations accept your own proxy URL. |
ScraperAPI is a focused tool and the comparison should be read that way. If your problem is "this page blocks me", their per-request knobs are more expressive than anything in our catalog, and at heavy sustained volume on that one job their credit tiers win on price.
The comparison changes when the job is "assemble data from a dozen places". A credit balance against one generic endpoint means you still write and maintain the parser for every source, and the parser is where the ongoing cost actually lives — sites change, your extraction breaks, and the credits kept flowing the whole time. upAPI ships the parsed result: a Maps search returns structured places, not HTML you have to interpret.
The other structural difference is the entry point. Their smallest paid plan is $49; ours is $0 forever and $29 when you outgrow it, which matters when you are still finding out whether the data is useful at all.
Where ScraperAPI wins
- It is very good at the one job. Unblocking a stubborn page — retries, IP rotation, JS rendering, geotargeting — is their whole product and it is more configurable per request than ours.
- Their credit tiers are cheaper per request at high, sustained scrape volume than paying for a general catalog you are not using.
- Async and batch endpoints for very large jobs, which we do not offer: upAPI calls are synchronous and edge-capped at 28 seconds.
Where upAPI wins
- The catalog is the product. If what you actually need is Maps places, a Reddit thread, a company profile, an OCR pass and a screenshot, those are five operations behind one key — not five scrapers you write yourself on top of one unblocking endpoint.
- Nothing is gated behind a tier threshold. Pay-as-you-go behaviour that ScraperAPI unlocks at $475 is simply how the meter works here at $0.
- The heavy work is priced honestly and separately. A browser render costs 20 units and a plain read costs 1, so a workload that is mostly reads is not paying browser prices for all of it.
- MCP and the in-app copilot: an assistant can search the catalog, read the schema and run the call, metered on the same quota.
When to pick which
- ScraperAPI. You fetch arbitrary URLs you pick yourself and need per-request control over rendering, geotargeting and sessions.
- ScraperAPI. You run very large batch jobs: their async API accepts up to 50,000 URLs in one batch job. upAPI calls are synchronous.
- upAPI. You want structured results from a named source, such as a Maps search or a page’s Open Graph tags, and would rather not write and maintain the parser.
- upAPI. You are still testing whether the data is useful, and want a free plan with a published per-call cost before you commit to a monthly spend.
- Both. You have a few stubborn pages to unblock and a set of named lookups. Use ScraperAPI for the first and upAPI for the second.
What switching actually costs
Point the request at `api.upapi.io/{slug}` with a JSON body instead of a URL query string, and delete the parsing code — the response is already structured. Run both on the same inputs on the free tier first; the outputs are directly comparable.
Questions about ScraperAPI and upAPI
How much does ScraperAPI cost compared with upAPI?
ScraperAPI sells monthly credit plans from $49 for 100,000 credits to $1,975, with a normal request at 1 credit, JavaScript rendering at 10 and a Google or Bing request at 25 (their docs, read 2026-10-05). upAPI is free at $0 with 1,000 weighted units a month, then $29, $99 and $499. A credit and a weighted unit are different meters, so compare the price of the job you run, not the plan sizes.
Does ScraperAPI have a free plan?
Its pricing page offers a 7-day trial with 5,000 credits and no credit card. Its docs FAQ describes 1,000 free credits a month with up to 5 concurrent connections instead. The two pages disagree, so check the sign-up flow before relying on either. upAPI’s free plan is 1,000 weighted units a month.
Can upAPI replace ScraperAPI for scraping any URL?
No. upAPI has no operation that fetches an arbitrary URL and returns its HTML with render and proxy flags. The closest are page-to-Markdown, which reads any URL over plain HTTP without a browser, plus screenshot and HTML-to-PDF, which use a headless browser, and operations for page metadata and titles. For named sources, an upAPI operation returns the parsed result directly.
Does ScraperAPI charge for failed requests?
Only for successful ones. Their docs say they charge for responses with a 200 or 404 status code, and nothing for a request that still fails after 70 seconds of retrying. A 200 response that contains a CAPTCHA counts as successful, so you are asked to report it to their support team.
Does upAPI have an async or batch API like ScraperAPI’s?
No. Every upAPI call is synchronous and the gateway cuts it off after 28 seconds. ScraperAPI’s async API runs jobs in the background for up to 24 hours and accepts up to 50,000 URLs per batch job, so very large crawls are a reason to keep it.
Sources read on 2026-10-05: https://www.scraperapi.com/pricing/, https://docs.scraperapi.com/getting-started/quick-start/credits-and-requests-costs, https://docs.scraperapi.com/responses-and-formats/api-status-codes, https://docs.scraperapi.com/responses-and-formats/output-formats, https://docs.scraperapi.com/control-and-optimization/geotargeting/standard-geo, https://docs.scraperapi.com/asynchronous-api/overview.
Other comparisons: Apify, Bright Data, Serper, ApyHub, RapidAPI, ScrapingBee, Firecrawl, ScreenshotOne.