Residential proxies that exit through real home broadband, not a data center
90M+ compliantly sourced residential IPs across 195 countries. Four billing models — per GB, per Mbps, per IP-day, per IP-month — so you pay the way your workload actually runs.
Not because it is the cheapest. Because someone picks up when it breaks.
Rotation you control
Rotate on every request, or hold a sticky session for up to 90 minutes. Intervals are settable from 1 to 360 minutes.
Sourcing you can audit
Every residential IP is KYC-authorised and traceable to its origin. No botnet-sourced nodes enter the pool.
Continuous purity checks
Reputation scores and fraud fingerprints are monitored continuously. Flagged IPs leave the pool automatically, holding request success at 99.9%.
API and sub-accounts
A full API alongside the console — sub-accounts, isolated quotas and automated extraction, so it drops into the stack you already run.
HTTP(S) and SOCKS5
Both protocols in full, unlimited concurrency, and targeting down to city, state or ASN.
Engineers, not ticket bots
24/7 support in English and Chinese, with an engineer on your integration — someone who can actually read the request that failed.
FAQ
What people ask before buying
Anything else, ask an engineer directly — no ticket bot in between.
What is the difference between rotating and static residential proxies?
Rotating IPs swap the exit address on a schedule you set and are billed by traffic or bandwidth — right for scraping, where it does not matter who sent the request. Static IPs stay fixed for the lease, are dedicated to you, and are billed per IP — right for store back-ends and social accounts, where it matters a great deal. The test is simple: does the target platform re-verify you when the address changes?
What is a native ISP proxy, and how is it different from a broadcast residential IP?
A native ISP address is handed out by a local carrier to an actual household, and the hardware carrying it sits in someone's home — reverse ASN and WHOIS lookups both return a residential network. A broadcast residential IP is registered as residential but announced from a data center over BGP: faster and cheaper, but it still carries a hosting fingerprint. Use native where scrutiny is high (social accounts, banking, limited drops) and broadcast where throughput matters more (scraping, ad verification).
How much do residential proxies cost, and does unused traffic expire?
Rotating traffic starts at $0.8/GB, bandwidth at $32/day for 100Mbps, short-term static at $0.08 per IP-day and long-term static at $3.8 per IP-month, with unit prices stepping down as volume grows. Traffic plans have no expiry — the balance sits in your account until you use it. Static IPs bill for the period you bought, and renewing keeps the same address rather than issuing a new one.
Can I target a specific city, state or carrier?
Yes. Country-level targeting covers all 195 regions; 33 priority markets add city and ASN (carrier) targeting, and the US adds state. This matters most in markets with wide internal variation — in the US, Brazil or Türkiye, sales tax, fulfilment times and platform pricing all shift by region, so scraping from the wrong one returns data that simply does not apply.
How do I connect? Do you support SOCKS5 and anti-detect browsers?
Both HTTP(S) and SOCKS5 are supported, with either user:pass auth or an IP allowlist, plus an API for bulk extraction. Anti-detect browsers — AdsPower, BitBrowser, Multilogin, Dolphin Anty — take the proxy string directly, and there are ready snippets for Scrapy, Playwright, Puppeteer and Selenium. Sub-accounts let you isolate usage per team.
How many accounts can I run behind one static IP?
For anything you intend to keep, one account per IP. When several high-value accounts share an exit, the platform can cluster them — and one suspension takes the rest with it. Low-risk read-only accounts can share, but live streaming, store back-ends and ad accounts are the wrong place to economise. Worth knowing: IP is only one clustering signal, so pair it with an isolated browser profile.
Get one request working. Scale after that.
Test before you commit, with an engineer on your integration.