Bright Data Web Unlocker vs Oxylabs Web Unblocker: 2026 Pricing per 1K and per GB

A web unlocker is sold in two incompatible units, so "what does it cost per 1,000 requests" has an answer only for half the market. On the rates published on 2026-09-05, Bright Data Web Unlocker charges $1.50 per 1,000 successful requests pay-as-you-go, IPRoyal Web Unblocker $1.00 per 1,000, Thordata Web Unlocker $1.30 per 1,000 delivered responses, and Decodo Site Unblocker $1.25 per 1,000 on a request meter it keeps behind a toggle — while Oxylabs Web Unblocker does not sell requests at all, only gigabytes, from $9.40/GB. The vendor page a buyer actually lands on for this question answers none of it: brightdata.com/products/web-unlocker/pricing returned an HTTP 301 on 2026-09-05 to a hub page that shows no Web Unlocker price at all.

What each unlocker charges, and in which unit

The naming collision is real — Bright Data says Unlocker, Oxylabs and IPRoyal say Web Unblocker, Decodo says Site Unblocker, Thordata says Web Unlocker — but the products compete for the same job: fetching public pages that reject ordinary automated traffic. Every figure below comes from the vendor's own page as displayed on 2026-09-05, in USD, with the interaction needed to see it stated wherever a figure is not visible on arrival.

Sources: Bright Data Web Unlocker pricing, Oxylabs Web Unblocker pricing, Decodo Site Unblocker, IPRoyal Web Unblocker pricing, Thordata Web Unlocker pricing, all fetched 2026-09-05.

That table poses the question it cannot answer: two of these vendors bill for bytes and three bill for requests, and nothing in either rate card converts between them. The conversion needs one number no vendor publishes — your average page weight — so the tables below compute it from the published entry rates, and let you substitute your own workload.

Three things about those figures are worth stating in words. Nothing in them ranks a per-GB rate against a per-1,000 rate: those are different products until a page weight is supplied, which is exactly what the break-even row supplies. Bright Data's cheapest published rates are overage rates — what requests beyond an included allowance cost — so its comparable entry figure stays $1.50 per 1,000, and it is marked as such rather than being sorted beside a bundle price. And IPRoyal prints no bundle totals anywhere on its pricing page, so any total shown for IPRoyal, in the table above or anywhere else on this site, is ProxyFacts arithmetic over its published per-1,000 rates rather than a figure the vendor displays.

Engineer's take (Hinata): The only input in this whole comparison you control is your own average page weight, and most teams do not know theirs. Fetch a few hundred real target URLs through Bright Data's recurring free allowance or the Oxylabs trial, divide bytes transferred by successful pages, and only then pick a meter — because on a per-GB product you are buying your own markup: Oxylabs states that both upstream and downstream traffic counts, so every retry and every redirect is billable weight that a per-request meter charges nothing for. My second rule is to budget the rate you can actually see today. WU40 is a six-month promotion, and a pipeline sized against $5.64/GB inherits the list rate of $9.40/GB in month seven. Per-request pricing is the safer default when you cannot predict page weight; per-GB wins only when you have measured and your pages are genuinely light.

How unblockers differ from raw proxies and scraping APIs

An unblocker is the middle layer of the managed-scraping stack. Raw residential proxies give you IP addresses billed per gigabyte, and everything else — rotation, retries, browser fingerprints, block detection — is your problem; our guide on how to scrape without getting blocked shows how much engineering that "everything else" actually is. Full scraping APIs sit at the other extreme, adding parsing, templates and structured output on top; we compare those in best scraping APIs.

Unblockers do exactly one job: given a URL on a public site that rejects datacenter or naive automated traffic, return the page. They bundle proxy selection, fingerprint management, JavaScript rendering and CAPTCHA handling behind a single endpoint — Decodo's and IPRoyal's even present as ordinary proxy endpoints, so existing scraper code can adopt them by changing a config line. You keep your own parsing and orchestration. If you are unsure whether you need this layer at all, scraping API vs proxy walks through the build-versus-buy decision from both ends, and proxy pricing comparison gives the raw per-GB baseline these products are marked up from.

One measurement makes every number on this page actionable, and it is the one to take before you buy anything: the real bytes a successful fetch of your own targets moves, including retries and rendering traffic. Our web scraping cost estimation guide covers how to measure page weight, retry overhead and render ratio properly.

1. Bright Data Web Unlocker — turnkey per request, and the page that will not show a price

Bright Data Web Unlocker is the turnkey per-request option: $1.50 per 1,000 requests pay-as-you-go, billed for successful requests, with a recurring free tier of 5,000 requests a month that needs no credit card, per its pricing page fetched 2026-09-05. The Scale plan is not one plan but a three-step slider: $499/mo includes 383,000 requests, $999/mo includes 908,000 and $1,999/mo includes 2,000,000, with additional requests beyond each allowance at $1.30, $1.10 and $1.00 per 1,000. Enterprise is quote-only. Full browser rendering, CAPTCHA handling and unlimited concurrency come with the paid plans.

The thing worth knowing before you go looking for those numbers yourself is that the obvious URL does not have them. On 2026-09-05, brightdata.com/products/web-unlocker/pricing returned an HTTP 301 to Bright Data's general pricing hub, which shows no Web Unlocker price at all; the figures above are on brightdata.com/pricing/web-unlocker, a different path. The site-wide menu compounds it by advertising "Starts from $1/1k req" — that $1.00 rate exists only as the overage rate at the deepest committed slider step, not as anything a new buyer pays.

Two corrections to what this page said before. Bright Data's published Web Unlocker floor is $1.00 per 1,000, not the $1.30 per 1,000 quoted here until 2026-09-05; our July capture recorded only what it found and never what it looked for and did not find, so we cannot tell from our own record whether the two cheaper slider steps were added since or were simply never opened. And the "98% success rate" this page attributed to Bright Data appears nowhere on the cited page; the page's own wording is "typically 100%" and "a near 100% success rate", both vendor claims, and the unsourced figure has been removed rather than restated. The first-deposit match of up to $500 is promotional — a one-time credit, not a rate change — and no code or duration for it is shown on the page. Bright Data also runs the strictest customer vetting of the products priced here, geared to registered companies; details in our Bright Data review. This page compares prices within the unlocker family only; Bright Data's whole catalogue — residential proxies, Web Unlocker and Web Scraper API — sits on the Bright Data pricing ledger.

2. Oxylabs Web Unblocker — per GB, with its floor behind a tab

Oxylabs Web Unblocker is the per-GB incumbent: $9.40/GB at 8 GB, $8.60/GB at 38 GB and $7.50/GB at 88 GB, per its pricing page fetched 2026-09-05. Those three plans are what an arriving visitor sees. Behind the page's Enterprise tab sit three more, all priced: $7.00/GB at 128 GB, $6.00/GB at 333 GB and $5.00/GB at 700 GB ($3,500/mo), with only a Custom+ band unpriced. That correction matters more than any other on this page — until 2026-09-05 this article described Oxylabs as having "an Enterprise tier without listed prices" and reported its floor as $7.50/GB, which is half again the vendor's published floor of $5.00/GB. The pricing navigation corroborates it in plain HTML, rendering "-40% Off Starts from $5 $ 3 /GB" — the $5.00 list floor beside the $3.00 the WU40 promotion takes it to.

Billing is success-based, and the product page states the mechanic in full: failed attempts are excluded, and "both upstream and downstream traffic is calculated". A billable gigabyte is request plus response, which is the hidden markup of every per-GB unblocker and the reason a heavy-page workload drifts away from this meter faster than the rate card suggests. Rate limits run 50 requests per second on the open plans and 100 on the two largest Enterprise ones. Top-ups extend Micro to 73 GB and the larger plans to 143 GB and beyond, but the top-up rate per GB is not published — unverified, and not estimated here.

The WU40 code takes 40% off for six months, giving $5.64, $5.16 and $4.50 per GB on the three open plans. It is a promotion, continuously displayed since at least 2026-07-17 with no end date shown, and month seven reverts to list. The free trial covers 1 GB with no card at 10 requests per second, but the vendor contradicts itself on its size — the pricing card says up to 10K results while the product-page FAQ says 2K, both live on 2026-09-05 — and the "1-week" duration survives only in the pages' meta description rather than in copy a visitor reads. Web Unblocker returns raw HTML only, with no parsed output, and the "98%+ at an effective $0.60 per 1K" figure often quoted from Oxylabs is a customer testimonial (a vendor claim) rather than a rate: this product has no per-1,000 meter. More in our Oxylabs review. The same handoff applies to Oxylabs: this page prices only its unblocker, and the whole catalogue — residential proxies and Scraper API alongside Web Unblocker — is on the Oxylabs pricing ledger.

3. Decodo Site Unblocker — two meters on one page, plus VAT

Decodo Site Unblocker survived the April 2025 Smartproxy rebrand as a standalone, proxy-format product, and the pre-rebrand /scraping/site-unblocker URL still returned HTTP 404 with no redirect on 2026-09-05 — which means a good deal of third-party coverage still points at a dead page. On the current page, fetched 2026-09-05, the per-GB ladder runs $10.00/GB at 1 GB, $9.00 at 5 GB, $8.50 at 10 GB, $8.00 at 25 GB, $7.50 at 50 GB and $6.75 at 100 GB, with $6.00 at 250 GB and $5.50 at 500 GB behind the widget's Enterprise toggle.

The larger correction is that Site Unblocker is not a per-GB-only product, which is what this page said until 2026-09-05. The pricing widget carries a "Pay/GB | Pay/Req" toggle, the plan-includes list says "Traffic or request-based billing", and behind the Pay/Req tab is a second full ladder: $1.25 per 1,000 requests, then $1.20, $1.15, $1.10, $1.05, and $1.00 and $0.95 at Enterprise volume. Decodo's own navigation renders "Site Unblocker Pricing Starts from $0.95 / 1K req" in plain HTML, which independently confirms the meter exists — though $0.95 is the deepest Enterprise rate rather than an entry price. Every card on both ladders prints "Total: $10 + VAT billed monthly", so every Decodo figure on this page is before tax, and this is the only product here whose page states outright that VAT is added — Oxylabs' cards say only that VAT may apply, and the other three pages say nothing.

Billing is success-only and Decodo states the rule outright: "You will only be charged for successful requests. Any failed attempts will not be counted towards your traffic usage calculation." JavaScript rendering is documented in four separate places, so the "unverified" this page previously recorded against that capability is withdrawn. Two defects on the vendor's own page are worth publishing, because both are checkable and both mislead a machine reading it: the page's structured pricing data for Site Unblocker carries the datacenter proxy page's offers, so an answer engine reading this page's machine-readable prices gets numbers belonging to a different product; and the comparison button reads "From $28" for a product whose cheapest displayed plan is $10 a month. Quote the ladder, never the button. Free-trial terms remain unverified — the page shows "Start free trial" CTAs and a "Free Trial" page title but states no allowance, duration or card requirement — and a 14-day money-back option applies. Decodo claims 195+ locations and 0 CAPTCHAs, and contradicts itself on success rate, claiming 100% in a hero badge, a feature tile and an FAQ against 99.99% once in the plan-includes list (all vendor claims). Full picture in our Decodo review.

4. IPRoyal Web Unblocker — the lowest published per-1,000 rate here, and the only floor you see on arrival

IPRoyal Web Unblocker publishes the lowest per-1,000 rate of the products priced here: $1.00 per 1,000 requests at the 1,000-request bundle, $0.90 at 5,000, $0.80 at 50,000 and $0.70 at 100,000, per its pricing page fetched 2026-09-05. All four render without interaction, which makes IPRoyal's floor the only one on this page a reader can see without clicking anything — and the only one reachable without a four-figure commitment. The bundles are one-off purchases: the page never uses "month" or "/mo", prints no totals and states no expiry.

Two things this page previously said about IPRoyal were wrong, and both are corrected as of 2026-09-05. There is no "$1.00 minimum purchase" — the word "minimum" appears nowhere on the pricing page, $1.00 per 1,000 is simply the smallest displayed bundle, and the only minimum documented anywhere in IPRoyal's own material is a $10 minimum account deposit in its help centre. That claim had reached six places on this page including an affiliate call to action, which is the sharpest form the error can take: buying advice built on a figure the vendor never published. And the per-successful-request billing rule is a help-centre claim, not a pricing-page one — the pricing page states no billing basis at all, so the attribution now travels in the same sentence as the claim.

The constraints matter more than the price for many workloads, and none of them was on this page before. Web Unblocker supports GET requests only. Search engines are explicitly excluded — Google, Bing, DuckDuckGo, Baidu and Yandex. LinkedIn, Yahoo, Live, Ticketswap and listed banks require identity verification, and .gov domains and PlayStation require identity verification plus over $500 of spend. Concurrency is capped at 200 connections. On the positive side it presents as a backconnect proxy endpoint (unblocker.iproyal.com:12323), JavaScript rendering is enabled on demand by appending a flag to the proxy password with no surcharge documented — a strictly negative finding, since no page states that rendering is free either — and it draws on a claimed 64M+ residential pool (vendor claim). No trial is shown on either Web Unblocker page, and the free-trial help-centre article this site cited returned HTTP 404 on 2026-09-05. Background in our IPRoyal review.

5. Thordata Web Unlocker — a full ladder that a plain fetch cannot see

Thordata Web Unlocker meters delivered responses rather than requests sent, and publishes a complete ladder: 5,000 responses free once, then $1.30 per 1,000 at the 10,000-response package ($13.00), $1.20 at 50,000 ($60.00), $1.10 at 100,000 ($110.00) and $1.00 at 200,000, per its pricing page fetched 2026-09-05, with a separate contact-sales route above that. Billing is per successful response, stated in the page's FAQ. This product has been absent from this comparison in every language since the article was published, while its figures have been in our dataset since 2026-08-11; that gap is closed here.

Three caveats belong with those rates. Paid packages carry 30-day validity and the free block a 7-day window, so an unused balance lapses — a real budget risk that the "$/Mo" framing on the cards hides. The vendor contradicts itself on the free trial, with a card badge reading "7-Days Free Trial" against an FAQ on the same page offering 30 days. And the ladder is unusually hard to verify: all five cards render $0, "0 responses" and "0 requests /s" in the server HTML, and the prices appear only after the page's scripts run. They are published on the same URL in three mutually consistent machine-readable blocks, which is why they are quotable at all — but a no-JavaScript fetch or an archived snapshot of that page shows $0 everywhere, so cite the ladder rather than a screenshot of the cards.

Thordata publishes no success-rate percentage at all, only an unquantified "high success rate", and the product is documented as incompatible with anti-detect browsers and automation frameworks such as Adspower, Playwright and Puppeteer. That limit is not Thordata's alone: Bright Data's own FAQ says Web Unlocker "is not made for use on browsers or third party tools like Adspower, Puppeteer, Playwright, or Multilogin (MLA)", and both vendors sell a separate browser product for that path — so it is a property of the unlocker category rather than a defect of one vendor. The gap that matters most for a buyer with a compliance function is upstream of price: neither Thordata pricing page states how the residential IP addresses behind the network are acquired, and when we asked directly the vendor declined to provide that documentation, citing confidentiality. We record that as unverified rather than as evidence of wrongdoing, and it is why this page prices Thordata on its published figures without placing it in the recommendation below. Our Thordata review sets out the full picture.

Who is missing from this comparison, and why

SOAX markets an unblocker and does not price it. On 2026-09-05 the product ships as Web Data API, with a feature card headed "Advanced unblocker" and sign-up deep links carrying an unblocker product parameter — but no rate is published anywhere on the public site. The product page's own "View pricing" button targets a URL that renders SOAX's own 404; the purchase link redirects to a login wall; the sitemap contains no unblocker or unlocker URL; and the general pricing page prices the proxy network only, in credits, with no published credits-per-request conversion. The vendor publishes the currency and withholds the exchange rate, so there is nothing to compare.

DataImpulse publishes no unblocker product at all, and says so in its own words. Its blog post on the subject, last updated 2026-09-03, reads: "DataImpulse provides the proxy layer, not a managed unblocker." Its page sitemap enumerates thousands of URLs and none contains "unblock" or "unlock"; its product menus list proxy pools only.

An unpriced product is excluded rather than estimated, and a product that does not exist is not given a zero. A comparison that fills a gap with a plausible-looking guess is worse than one with fewer columns, because the guess is indistinguishable from the sourced figures beside it — and this site has published a coverage false positive that way before.

Advertised floor versus the rate you can actually click

The cheapest rate a vendor advertises and the cheapest rate a visitor can reach are usually different numbers, and on this product family the gap is systematic. Every floor below is published by the vendor on 2026-09-05; what varies is the interaction it takes to see it.

• Bright Data: $1.00 per 1,000, behind a slider. The Scale card's default step shows $1.30 per 1,000; the two cheaper steps exist only once the "Slide to adjust plan" control is dragged. Both are overage rates, and the site menu advertises the $1.00 figure as "Starts from $1/1k req" without saying it needs a $1,999/mo commitment.
• Decodo: $0.95 per 1,000, behind a tab and a toggle. The request meter itself is behind the Pay/Req tab, and its two cheapest rungs are behind an Enterprise toggle inside it. Decodo's navigation advertises the $0.95 rate in plain HTML on every page. Its per-GB floor of $5.50/GB is behind the same toggle.
• Oxylabs: $5.00/GB, behind a tab. The three plans a visitor lands on stop at $7.50/GB. The Enterprise tab holds the rest, and the pricing navigation — which renders "-40% Off Starts from $5 $ 3 /GB" — is the only place the floor appears without an interaction.
• Thordata: $1.00 per 1,000, only after the page's scripts run. The server HTML prices every card at $0.
• IPRoyal: $0.70 per 1,000, visible on arrival. All four bundles render in the page. It is the only floor here that a plain fetch, a screen reader or an archived snapshot will show you.

Promotions are a second layer on top of that. Oxylabs' $5.64/GB is the WU40 rate against a $9.40/GB list, and it lasts six months. Bright Data's first-deposit match changes no unit rate at all. Decodo's "SAVE 15%" and IPRoyal's "Save 0% to 30%" badges are volume-ladder labels rather than time-limited offers, so a budget built on them inherits no later jump; Thordata's "50% OFF" banner ships hidden on every page load and was inactive when we checked. The practical rule is the same in all cases: budget the list rate you can see for yourself, and treat every promotion as a bonus that expires.

Success rates are vendor claims

Every success rate, pool size and uptime figure on this page comes from a vendor's marketing, and none has been independently audited. The vendors do not even agree with themselves: Bright Data's page says "typically 100%" and "a near 100% success rate"; Decodo claims 100% in a hero badge, a feature tile and an FAQ, and 99.99% once in a plan-includes list; Oxylabs publishes no vendor-wide figure and offers a customer testimonial of 98%+ instead; Thordata publishes no number at all; IPRoyal's 99%+ is scoped to a named list of ecommerce, travel and job-board targets.

ProxyFacts has not run a first-party test of any of these products. There is no benchmark behind this page, no measured success rate, and no latency figure — this is a comparison of what the vendors publish, dated, with the interaction needed to reproduce each figure. Success-based billing softens the risk on all of them, since a failed request does not bill, but it does not remove it: failures still cost latency and schedule. The honest test is to run the free allowances against your own hardest target before committing, because the worst URL on your list decides feasibility, not the average one.

Compliance: what unblockers are for — and what they are not

Unblockers exist for one legitimate problem: public pages that indiscriminately block datacenter IPs and automated traffic even when the visitor is collecting data lawfully — price monitoring, SEO rank tracking, ad verification, market research, AI training corpora. That is the use ProxyFacts covers, and it is the use these vendors' own compliance programs are built around. They are not tools for getting behind login walls or paywalls, harvesting personal data, or automating purchases — activities excluded from our coverage and prohibited by every provider here. IPRoyal's published gating is a concrete example of a supplier drawing that line: identity verification for a named list of high-risk targets, and identity verification plus a spend threshold for government domains.

The KYC gradient is a genuine selection criterion, and I read it as a feature: providers that verify customers keep abusers off the infrastructure your traffic shares. Bright Data restricts access to human-verified registered companies; Oxylabs requires a KYC form from every customer at signup with risk-based escalation; Decodo runs automated fraud checks and KYC on every registration and blocks high-risk target categories; Thordata publishes a multi-level vetting policy; IPRoyal's KYC is mandatory for static ISP proxies, with unverified accounts staying partially restricted. This article is not legal advice, and none of it substitutes for advice on your own situation. If the legal side of your project is unsettled, start with is web scraping legal? before buying anything.

Our methodology

This is a document-based comparison, not a benchmark. Every price, tier and term above comes from the vendor's own page, read first-hand on 2026-09-05, with the URL cited and the date stamped on the figure rather than on the article. Promotional rates are labelled promotional every time they appear, vendor-supplied performance figures are labelled vendor claims, and where a page shows nothing we write unverified rather than estimating — Oxylabs' top-up rate per GB, Decodo's and IPRoyal's trial terms, and SOAX's unblocker price are all unverified for that reason. Because most of these vendors keep their cheapest rates behind an interaction a plain fetch cannot perform, each floor on this page is published with what it takes to see it; a figure you cannot reproduce is a figure you should not have to take on trust.

This page also corrects itself in public, because a comparison site that quietly fixes its own numbers is worth less than one that shows the correction. Until 2026-09-05 it reported Oxylabs' floor as $7.50/GB and Bright Data's as $1.30 per 1,000. Both were above the vendors' published floors — $5.00/GB and $1.00 per 1,000 — and our July capture recorded only what it found, never what it looked for and did not find, so we cannot tell from our own record whether those rates were added since or were simply never opened. Two further claims were withdrawn on the same date: a 98% Bright Data success rate that appears nowhere on the page we cite, and a $1.00 IPRoyal minimum purchase that the vendor never published. Nothing in the unit prices themselves had moved since July.

ProxyFacts has run no first-hand benchmark of these products, so every performance number here is a vendor's. ProxyFacts is monetised through disclosed affiliate partnerships, and those partnerships never change which numbers are reported or the order they appear in: IPRoyal pays this site nothing and publishes the lowest per-1,000 rate here, which is stated as plainly as any paid partner's figure, and Thordata pays this site a commission yet is still left out of the recommendation below, because neither of its pricing pages says how the residential IPs behind the product are acquired and the vendor declined to document it when asked. The figures on this page are also published machine-readably at /data/proxy-pricing.json under CC BY 4.0.

Which unblocker should you choose?

• Choose Bright Data Web Unlocker for turnkey scale. Success-only per-request billing, browser rendering in the rate, unlimited paid-tier concurrency, and 5,000 free requests every month to validate against your real targets — if you can clear company-level verification.
• Choose IPRoyal for the cheapest published per-request entry here. $1.00 per 1,000 falling to $0.70, no plan to commit to, and no surcharge documented for rendering — provided GET-only, no search engines and the target gating all fit your workload.
• Choose Oxylabs Web Unblocker for measured, light traffic. Per-GB predictability with the deepest published ladder here, sized against list rates rather than WU40, and remembering that upstream traffic counts too.
• Choose Decodo Site Unblocker when you want a proxy-format endpoint or a choice of meter. One product, either billing unit, before VAT, with a 14-day money-back option to test it against your own targets.

Frequently asked questions

Which meter should I pick if I have never measured my page weight? The per-request meter, as a default you can revise. A per-request bill does not move with bytes, so an unmeasured workload cannot surprise you on that meter, whereas a per-GB one prices retries, redirects and rendering traffic you have not counted. Then measure, and use the break-even table above to check whether your real average page weight puts you on the other side of the line.

Should I size an Oxylabs plan around the WU40 discount? No. The 40% cut runs for six months, and the durable figures are the list rates of $9.40, $8.60 and $7.50 per GB on the open plans. Model steady-state cost at list, and treat the promotion as a bonus rather than a baseline.

Can I verify these prices myself? Yes, and each one takes a different action. IPRoyal's four rates and Oxylabs' three open plans render on arrival. Oxylabs' cheaper plans need its Enterprise tab; Decodo's request ladder needs its Pay/Req tab and its cheapest rungs an Enterprise toggle inside it; Bright Data's cheaper rates need the Scale slider dragged; Thordata's whole ladder needs JavaScript enabled, because the server HTML prices every card at $0. That is why each figure on this page is published with the interaction it requires.

Will an unblocker fetch pages behind a login or paywall? That is not what these products are for. Using them that way violates every provider's terms and falls outside ProxyFacts' coverage. Unblockers are built for public pages that block automated or datacenter traffic.

For where unblockers fit in the full data-collection stack — raw proxies, unblockers and scraping APIs — read our web scraping guide pillar.