Crawlbase

Crawlbase is a scraping-API-first proxy provider that started life in 2016 as ProxyCrawl and rebranded in 2022 to reflect a wider product suite. Rather than selling raw proxy bandwidth by the gigabyte like Bright Data or Decodo, Crawlbase wraps its residential and datacenter pool behind a single Crawling API endpoint: you send a URL, the platform picks the exit IP, solves the anti-bot challenge, optionally renders JavaScript in a headless browser, and returns the rendered HTML. That positioning makes Crawlbase a competitor to Zyte and ScraperAPI more than to traditional proxy networks, and it is the lens through which any review of Crawlbase pricing should be read.

The underlying network is around 2M+ residential and datacenter IPs spread across roughly 45+ countries, with country-level (not city-level) geo-targeting exposed via a simple country parameter. There are two ways to consume it. The Crawling API is the flagship product and is billed per successful request. The Smart Proxy is a backconnect endpoint that speaks standard HTTP/HTTPS proxy protocol, so it drops into Scrapy, Playwright, Puppeteer, curl, or any tool that takes a proxy URL. Sticky sessions are supported on both products so a sequence of calls reuses the same exit IP, which is essential for login-gated scraping and cart flows.

Pricing is request-based rather than bandwidth-based, which is unusual in this category. The Crawling API starts at roughly $0.21 per successful request on the entry tier, with discounts as volume climbs and as you stay on simpler target domains; harder targets (Amazon, LinkedIn, Google SERPs) consume more credits per call. The Smart Proxy tiers are credit-based and run from around $149/month to $449/month, with a 5,000-credit trial for new accounts. Crawlbase also offers 1,000 free Crawling API calls with no card, expanding to 10,000 once payment info is added, which is generous for evaluation. There is no public refund policy beyond the trial credits, so treat the trial as your due-diligence window.

What Crawlbase is genuinely known for is the combination of a clean developer API, JavaScript rendering on the same endpoint as plain HTML fetches, and a support team that responds quickly relative to price. JS rendering is toggled by using a separate "JavaScript token" — handy because you pay extra credits only when you actually need a headless browser. Cookie forwarding, custom headers, and country pinning are all supported, and the documented rate ceiling on the Crawling API proxy mode is around 20 requests per second per account (~1.7M req/day), which is enough for most mid-sized scraping operations.

Crawlbase is the right pick for engineering teams building scrapers against e-commerce catalogs, news sites, public directories, SERPs, and Reddit-style targets where they want unblocked HTML back without owning the anti-bot stack themselves. It is not the right pick if you need raw GB-based residential bandwidth for ad verification, multi-accounting, sneaker copping, or stealth browsing — Decodo, Bright Data, IPRoyal, and Webshare are stronger there. It is also not the right pick for the most aggressively protected targets (some banking, ticketing, and Cloudflare Enterprise sites) where Zyte or Oxylabs Web Unblocker tend to have a higher success rate.

Limitations worth knowing before signing up: no city-level targeting, no mobile proxy product, no SOCKS5 (HTTP/HTTPS only), and a credit accounting model where harder domains silently cost more, which can make budgeting unpredictable at scale. Documentation is solid but the dashboard is more functional than polished, and analytics are thinner than what you get on Bright Data. Verify current pricing on the official site before purchasing — Crawlbase has revised its credit tiers multiple times since the rebrand.