scrapeprices 
ScrapingAnt
The steepest ladder of the six: 1 credit plain, 10 rendered, 25 residential and 125 for a browser on a residential IP, so the same URL can cost 125 times more depending on how it has to be fetched. On the $19 Enthusiast plan that is $0.19 per 1,000 plain pages and $23.75 per 1,000 at the top. The default matters more here than anywhere else, because browser is on unless you turn it off, so the naive request costs ten credits and the cheap rate is something you have to ask for. There is a dedicated Markdown endpoint that bills off the same table, which makes LLM-ready text as cheap as HTML if you disable the browser. A genuinely free 10,000 credits arrive every month with no card, failed requests cost nothing, and there is a proxy port that prices identically to the API. Two caveats in the price table itself: one row prices a datacenter browser request at 50 credits while another prices what reads as the same thing at 10, and nothing reconciles them, and Google domains are billed at 10 credits even without a browser. There is no anti-bot tier with a name and no published SLA, and plans hard-cap with no metered overage; the offered fix is restarting your plan early. Left null: the annual discount, since no annual option exists.
Plain page, by volume
| Pages per month | Monthly bill | Per 1,000 | Cheapest route | Allowance unused |
|---|---|---|---|---|
| 1k | $19.00 | $19.00 | Enthusiast | 99,000 units |
| 10k | $19.00 | $1.90 | Enthusiast | 90,000 units |
| 100k | $19.00 | $0.190 | Enthusiast | - |
| 1M | $249 | $0.249 | Business | 2,000,000 units |
How ScrapingAnt compares
At 10,000 pages a month ScrapingAnt costs $1.90 per 1,000, 14.6x the cheapest published route and 2.7x AnyAPI.
| Provider | 1k a month | 10k a month | 100k a month | 1M a month |
|---|---|---|---|---|
Z  Zyte API | $0.130 | $0.130 | $0.130 | $0.100 |
A  AnyAPI | $0.700 | $0.700 | $0.700 | $0.700 |
| S ScrapingAnt | $19.00 | $1.90 | $0.190 | $0.249 |
D  Decodo | $19.00 | $1.90 | $0.490 | no plan reaches this volume |
S  ScrapingBee | $19.00 | $1.90 | $0.490 | $0.099 |
List prices, dollars per 1,000 Plain pages at the cheapest published route to each volume. How this is worked out
Every published plan
| Plan | Price | Included credits | Overage each | Note |
|---|---|---|---|---|
| Free | $0 per month | 10,000 | hard cap | 10,000 credits every month, no credit card. Documentation-only support. |
| Enthusiast | $19.00 per month | 100,000 | hard cap | 100,000 credits a month, email support, docs-only integration help. No expert assistance, custom proxy pools or custom anti-bot tuning. |
| Startup | $49.00 per month | 500,000 | hard cap | Marked "Most Popular". 500,000 credits, priority email support, custom code snippets, expert assistance. |
| Business | $249 per month | 3,000,000 | hard cap | 3,000,000 credits, priority email support, debug sessions, custom proxy pools, custom anti-bot avoidances, dedicated account manager. |
| Business Pro | $599 per month | 8,000,000 | hard cap | 8,000,000 credits, priority debug sessions, same custom proxy and anti-bot options as Business. |
| Custom | on request | - | not published | "Talk to Sales" tier for 10M+ credits a month with priority and dedicated support and full enterprise onboarding. The card prints "$699+/mo", which is a starting point rather than a price for a stated number of credits, so monthlyUsd is null. |
The credit-cost doc prints one table for the whole API: "Simple request (no browser usage) + standard (datacenter) proxy 1; Headless browser (no JS rendering, return_page_source parameter is true) + standard (datacenter) proxy 2; Headless browser (JS rendering) + standard (datacenter) proxy 10; Simple request (no browser usage) + premium (residential) proxy 25; Headless browser (no JS rendering, return_page_source parameter is false) + standard (datacenter) proxy 50; Headless browser (JS rendering) + premium (residential) proxy 125; Any request to Google domain + standard (datacenter) proxy 10." The 50-credit row contradicts the 10-credit row on its face, since both describe a datacenter browser request, and no page reconciles them. Note the default: the browser parameter defaults to true, so the FAQ's "default request - Chrome rendering with standard proxies - costs 10 credits", and the 1-credit rate only applies when you pass browser=false. antibot is 125 because that is the dearest configuration ScrapingAnt publishes, a rendering browser on the residential pool; there is no separately named stealth, unblocker or ultra tier, and the CAPTCHA and Cloudflare page describes anti-bot avoidance (real latest-version browsers, fingerprinting, country selection, cookie chaining) as something included in every request rather than as a priced mode. Markdown is 1 because the /v2/markdown endpoint "accepts the same request structure as the general endpoint" and bills off the same table, so with browser=false a page of Markdown is the same single credit as HTML; left at the default it is 10. The AI extractor is priced differently again, by characters: one credit per 30 characters of the Markdown plus the output text, added to the underlying scrape cost, which is not a per-page figure and is left out of the verticals. Every response carries an Ant-credits-cost header with what the call actually cost.
Cost by kind of fetch
Per 1,000, at 10,000 pages a month.
What it can do
Only what the provider's own pages state.
Where these numbers came from
- https://scrapingant.com/#pricing read 2026-09-14 / the five-column plan grid: Enthusiast $19 for 100K credits, Startup $49 for 500K, Business $249 for 3M, Business Pro $599 for 8M and Custom at $699+ for 10M+, with the support, integration-help, custom-proxy-pool, custom-anti-bot and account-manager rows; "10,000 free credits every month", "No credit card required", "Pay only for successful scrapes - failed requests cost 0", the mid-month plan restart, and the FAQ line "The default request - Chrome rendering with standard proxies - costs 10 credits". No annual billing option is offered.
- https://docs.scrapingant.com/credits-cost read 2026-09-14 / the full credit table quoted in unitCost.note, "Unused API credits are not transferred to the next subscription period", the AI extractor formula of one credit per 30 characters of Markdown plus output added to the scrape cost, and the Ant-credits-cost response header
- https://docs.scrapingant.com/request-response-format read 2026-09-14 / the /v2/general endpoint and its parameters: browser (default true), return_page_source, proxy_type, proxy_country, js_snippet, wait_for_selector, cookies, timeout
- https://docs.scrapingant.com/proxy-settings read 2026-09-14 / proxy_type=datacenter (50,000+ IPs, the default) versus residential (2,000,000+ IPs), the 24 selectable proxy countries, per-request IP rotation, and mobile proxies available only on request
- https://docs.scrapingant.com/llm-markdown read 2026-09-14 / the /v2/markdown endpoint, which "accepts the same request structure as the general endpoint" and whose "cost of using this endpoint is based on the number of API credits consumed per request", pointing back at the same credit table
- https://docs.scrapingant.com/proxy-mode read 2026-09-14 / proxy port at proxy.scrapingant.com:8080 and :443 with parameters passed in the username; "Request cost is calculated in the same way as for the scraping endpoint"
- https://scrapingant.com/legal/terms-of-use read 2026-09-14 / operating company DATAANT SP. Z O.O., "registered and acting under the laws of the Republic of Poland under the business registration number (KRS) 0000848755"
Something out of date? Send the provider's own page to hello@scrapeprices.com and it is corrected on the next capture.