String is the better fit when a fixed list of protected pages has to return through one API. Oxylabs is the better fit when you need a raw residential proxy network, dedicated e-commerce parsers, or unusually strong PerimeterX coverage. In the August 2026 Web Data Frontier Benchmark, String returned verified content on 97.0% of 495 requests. Oxylabs returned it on 76.8%. String's failure-aware latency score was 9.98 seconds, compared with 14.67 seconds for Oxylabs.
Oxylabs still wins meaningful cases. It passed 97.8% of requests against PerimeterX, ahead of String's 88.9%, and returned Skyscanner on four of five attempts while String returned nothing. Its Web Scraper API also offers dedicated Amazon and Google rates, JavaScript rendering, custom parsers, and browser instructions. Buyers should test their own targets rather than treating the overall score as universal.
Last updated August 31, 2026. Benchmark figures come from the August 11, 2026 run. Oxylabs product, pricing, billing, and compliance claims were checked against its published Web Scraper API pricing, residential proxy pricing, documentation, and Trust Center on August 31, 2026.
| String | Oxylabs | |
|---|---|---|
| Product shape | One fetch API, plus search, sitemap crawling, remote Chrome, and MCP | Web Scraper API, Web Unblocker, and residential, mobile, ISP, and datacenter proxies |
| Benchmark success | 97.0% (480/495) | 76.8% (380/495) |
| Failure-aware latency score | 9.98s, second of 15 | 14.67s, seventh of 15 |
| Sites passed on all 5 attempts | 92 of 99 | 68 of 99 |
| Sites passed on no attempts | 2 of 99 | 18 of 99 |
| PerimeterX requests | 88.9% | 97.8% |
| Entry offer | 5,000 free requests; Starter $20 plus usage | 2,000-result free trial; Micro $49 |
| Roughly $100 tier | Growth $100 plus $0.20 to $4.00 per 1,000 verified requests | Starter $99; $0.45 to $1.30 per 1,000 counted results |
| Raw proxy network | No | 175M+ residential IPs, plus ISP, mobile, and datacenter products |
| Best fit | Protected pages that must return consistently | Proxy infrastructure, dedicated parsers, and PerimeterX-heavy target lists |
The Web Data Frontier Benchmark sends every provider the same 99 URLs five times. A response passes only when it contains a marker from the real page. A CAPTCHA or block page that returns HTTP 200 still fails. The August run made 7,425 live requests across 15 providers.
String built the benchmark and sells the product that ranked first. The test is public so that conflict is inspectable: the target list and pass criteria, Oxylabs adapter, and complete results JSON are available for review. The adapter used Oxylabs' general Web Scraper API, not every dedicated source the company sells.
| Provider | Successful requests | Success rate | Latency score |
|---|---|---|---|
| String | 480/495 | 97.0% | 9.98s |
| Oxylabs | 380/495 | 76.8% | 14.67s |
The latency score is failure-aware. For each target it takes the p75 of successful attempts; a provider that never returns the page inherits the p75 of providers that did. That prevents a fast block page from looking like good performance.
Oxylabs returned nothing on all five attempts for 18 sites: Amazon, Facebook, Booking.com, Kayak, Ticketmaster, SeatGeek, AXS, Zara, StockX, Saks Fifth Avenue, Louis Vuitton, Trustpilot, Apartments.com, Realtor.com, Stack Overflow, Roblox, Mouser, and Temu.
String returned content on 17 of those 18 and passed all five attempts on 15. Temu shut out both providers. Oxylabs was partial on another 13 sites; String was partial on five.
The Amazon result needs context. The benchmark used the general universal source. Oxylabs separately publishes dedicated Amazon sources with parsers for products, search results, pricing, sellers, and bestsellers. The 0-for-5 benchmark result describes the general endpoint, not the dedicated Amazon product.
The same caution applies to every buyer. Test the actual source and options you plan to purchase using a subset of the public harness.
| Anti-bot system | String | Oxylabs |
|---|---|---|
| Cloudflare | 100% | 77% |
| Akamai | 98% | 67% |
| DataDome | 98% | 77% |
| PerimeterX | 89% | 98% |
Oxylabs' PerimeterX result was the best in the field. It returned Skyscanner four of five times where String returned nothing, the clearest Oxylabs win in the run.
String's larger advantage appeared by business category. Oxylabs passed 25% of ticketing requests, 55% of review sites, and 63% of real-estate requests. String passed 100% across all three. Oxylabs' residential proxy FAQ also says some categories, including ticketing, government, streaming, banking, and mailing, are restricted on that proxy product. The benchmark tested Web Scraper API, so the published proxy restriction should not be treated as a proven cause of an API result.
Oxylabs Web Scraper API is a managed retrieval and extraction product. Its universal source accepts a URL and can add geolocation, JavaScript rendering, browser instructions, and parsing. Dedicated sources cover Amazon, Google, YouTube, and other common targets.
The platform goes beyond fetching. Oxylabs provides browser instructions, a custom parser, real-time and asynchronous integrations, and plan-specific rate limits. It also sells residential proxies and other proxy types for teams that want direct session and routing control.
String is narrower. Send a URL to the Web Access API and get Markdown, HTML, or JSON. Proxy rotation, fingerprinting, CAPTCHA solving, and browser escalation happen behind the endpoint. String also provides search, an asynchronous sitemap crawl, remote Chrome over CDP, and an MCP server. It does not provide raw proxy access or Oxylabs' catalog of dedicated parsers.
Both providers publish agent integrations. Oxylabs maintains an official MCP server. String's MCP server is listed as ai.usestring/web-access and exposes fetch, search, and sitemap tools.
Oxylabs also has the stronger procurement package. Its Trust Center lists ISO/IEC 27001:2022, SOC 2 Type 1 and Type 2, and SOC 3 materials. If one of those controls is a buying gate, it may outweigh the benchmark difference.
Oxylabs Web Scraper API pricing lists these self-service tiers:
| Plan | Monthly price | Amazon | Other, no JS | With JS | |
|---|---|---|---|---|---|
| Free trial | $0 | $0.50 | $1.00 | $1.15 | $1.35 |
| Micro | $49 | $0.50 | $1.00 | $1.15 | $1.35 |
| Starter | $99 | $0.45 | $0.90 | $1.10 | $1.30 |
| Business | $999 | $0.30 | $0.60 | $0.75 | $1.00 |
Rates are per 1,000 counted results. The same pricing FAQ defines success-based billing more broadly than verified content: 2xx and 4xx target responses count as successful, while system errors returning 5xx and 6xx do not. That means a target's 403 or 404 can count against the plan even though the benchmark treats a block page as a failure.
String charges a subscription plus requests that return verified content. The first 5,000 requests are free. Starter is $20 per month and Growth is $100. Growth rates are:
| Fetch path | Price per 1,000 verified requests |
|---|---|
| Request, standard proxy | $0.20 |
| Browser, standard proxy | $1.00 |
| Request, premium proxy | $2.00 |
| Browser, premium proxy | $4.00 |
String is cheaper on standard requests. Oxylabs Starter's $1.30 rendered rate is cheaper than String's $2.00 premium-request and $4.00 browser-plus-premium paths, and its $0.45 Amazon rate is designed for a dedicated source String does not provide. The target mix determines the answer.
Do not divide the list rate by benchmark success for either product. Their billing rules are not the same as the benchmark's content-marker rule. Success rate answers whether the page arrives. Billing documentation answers what counts against the plan. The broader 13-provider pricing comparison keeps those questions separate.
Oxylabs' raw proxy product has another unit entirely. Residential plans start at $30 for 5 GB, or $6 per GB, and reach $2.50 per GB at a 1 TB commitment. Headers, response size, assets, and retries all consume bandwidth. A gigabyte cannot be compared with a verified page by looking at one headline price.
Keep Oxylabs for raw proxy workloads and dedicated sources such as Amazon, then send the protected universal URLs that miss through String. The reverse can also work: use String as the default fetch layer and Oxylabs when a workflow needs dedicated parsing, browser instructions, or direct proxy control.
Both products offer a free evaluation. Compare the exact Oxylabs source and String fetch configuration against a representative target list before changing production traffic.
If Oxylabs is one option among several, the Oxylabs alternatives comparison places six substitutes next to the same benchmark.
On the August 2026 Web Data Frontier Benchmark overall, yes. String returned verified content on 97.0% of 495 requests and Oxylabs on 76.8%. Oxylabs was better against PerimeterX, 97.8% to 88.9%, and returned Skyscanner where String failed. Run the public test harness against your own target list.
Sometimes. String Growth's standard request costs $0.20 per 1,000 verified pages. Oxylabs Starter lists $1.10 for most non-rendered results and $1.30 with JavaScript. Oxylabs is cheaper than String when String needs its $2.00 premium-request or $4.00 browser-plus-premium path. Oxylabs also offers lower rates for dedicated Amazon and Google sources.
Its Web Scraper API pricing FAQ says 2xx and 4xx target responses count as successful results, while system failures returning 5xx and 6xx do not. A target 403 can therefore count against the plan even when it contains a block page. String bills only a request that returns verified content.
Oxylabs. Its residential proxy page advertises more than 175 million IPs across 195 locations, and it also sells mobile, ISP, and datacenter products. String does not sell raw proxies.
String on the failure-aware benchmark score: 9.98 seconds against Oxylabs' 14.67. That score includes the cost of targets a provider fails and should not be read as a raw response-time promise for every successful page.
Stay when PerimeterX dominates the target list, when you need raw proxy control or dedicated Amazon and Google parsers, or when Oxylabs' enterprise compliance package is a procurement requirement. The benchmark tested the general API and does not measure every dedicated source.
Yes. The benchmark repository, Oxylabs adapter, and August results file are public. A full run makes thousands of billable requests, so begin with a subset.
