Frequently Asked Questions

Sign up with an email address and a password, then confirm the address from the message we send. Confirming it releases the free allocation for eligible accounts. No card required.

Yes. A verified account gets 3 free shared proxies with 1 GB/month bandwidth and 10 concurrent threads, HTTP(S) only, with no card required. The free plan has no end date. If your free proxies carry no traffic for 30 days, we pause them and return their addresses to the pool; one click on your dashboard reactivates a fresh set, with new IP addresses and the same proxy username and password. Eligibility rules apply. Use it to run your own code against your own target before you buy. Paid plans add SOCKS5 and larger allocations.

Paid plans start as soon as the payment clears: credentials are generated when the subscription is created, so card and PayPal orders have no provisioning queue, and a Bitcoin order provisions once the payment confirms on-chain, which can take longer. Eligible free accounts start once their email address is verified.

HTTP(S) and SOCKS5 on every paid product, using the same credential for both. Free accounts are HTTP(S) only.

Node4.io is operated by Tekari LLC, a US company. Node4 owns the IP blocks its dedicated, shared and rotating datacenter proxies run on, in the United States, Italy and Spain, so allocation and subnet spread are under its own control. Residential exits come from a partner network in 170+ countries and are resold, not owned. Company details are on /about.

High-volume work against targets that accept datacenter addresses: Rotating Unmetered rotates exits per connection and is priced by concurrent connections with no data cap, and dedicated datacenter proxies give you static IPs of your own, priced per proxy with no data cap. Shared datacenter is the low-cost option for development, QA and light crawling. For targets that refuse hosting ranges, residential exits add country, state, city and ASN targeting. If you are unsure which a target needs, /tools/block-replay fetches a URL from both kinds of address and compares the answers.

Anything unlawful or against the destination site's terms. The Acceptable Use Policy names fraud, phishing and spam; malware; denial-of-service attacks; credential stuffing, account takeover and brute force; carding or testing stolen payment details; harvesting personal data in breach of GDPR, CCPA or similar laws; piracy; harassment; and reselling without written permission. The full policy is at /acceptable-use.

Support and sales are both available 24/7, and we reply within 12 hours. Technical questions go to support@node4.io or a support ticket from the dashboard, sales questions to sales@node4.io, and refunds and cancellations to billing@node4.io.

Dedicated datacenter proxies from IP blocks Node4 owns and announces, totaling thousands of IPs. Each address is allocated to one customer at a time, and bandwidth is included with no data cap or per-GB charge, so a heavy month costs the same as a light one. Run whois on any address we assign you to check the allocation.

A datacenter proxy comes from a hosting provider's IP range: cheaper, low latency, and identifiable as hosting, which some sites reject outright. A residential proxy routes through a real consumer connection, so it is much harder to tell apart from an ordinary visitor, at the cost of per-gigabyte billing and more variable latency. If you are not sure which your target needs, send the same request through both and compare the responses; the free allocation is shared datacenter, so that test costs nothing. See /help/choosing-the-right-proxy-type.

Yes. One account can hold plans for several products at once, managed from the same dashboard, and your one credential works on every product you hold an active plan for. A common arrangement is bulk work on datacenter addresses with residential kept for the targets that refuse hosting ranges.

Yes, on residential and on both rotating products. Add a session ID (letters and digits, up to 32 characters) to your proxy username and every request carrying that ID keeps the same exit. On Rotating Datacenter and Rotating Unmetered a session holds one exit for up to 30 minutes, counted from when the exit is first assigned rather than from your latest request. Residential sessions are best effort, with a 30-minute ceiling: the exit can drop sooner if its device goes offline, so switch to a new session ID after repeated failures. A new ID starts a new pin. Details and examples: /help/sticky-vs-rotating-sessions.

Yes. Contact support and we will move you to the new plan and settle any difference. Plan changes are handled by us rather than self-service, so your proxies and billing are both correct on the new plan.

Credit and debit cards (Visa, Mastercard and American Express), PayPal, Bitcoin through BTCPay (on-chain only), and USDT on the Tron network (TRC-20). Card and PayPal plans renew each month automatically until you cancel. Crypto cannot be collected from your wallet, so a Bitcoin or USDT plan does not renew on its own unless your account credit covers the renewal in full; otherwise you pay the renewal invoice from your dashboard. For invoiced or enterprise billing, contact us.

Your proxies pause on the renewal date and the same IPs are held for 14 days. Card renewals retry automatically over that window, and each reminder email links to a page where you can pay the renewal without logging in; paying inside the 14 days restores the same IPs. After 14 days the plan is paused rather than canceled: its IPs are released and nothing more is charged. Pay the renewal whenever you are ready, with Resume your plan on the Plans page, and it comes back straight away with fresh IPs and a new billing period from that day.

Nothing is billed for overage you did not ask for: access pauses at the cap until you top up or the period renews, unless you have switched auto top-up on. We warn you as you approach the limit. You can top up from your dashboard at any time, in sizes scaled to your plan, and a top-up lasts for the current billing period.

Yes. First-time customers can request a full refund within 7 days of their initial purchase under our 7-day money-back guarantee. See our Refund & Cancellation Policy for full details, including how subscriptions and bandwidth add-ons are handled.

Yes, for particular campaigns. Enter the code at checkout and the reduced total is shown on the order before you pay. A code that is live right now, such as the welcome offer for your first month, is shown on the pricing page.

Yes: a REST API at https://node4.io/api/v1 with Bearer token authentication, covering your proxies and gateway proxy strings, usage, the IP whitelist, sub-users with their own bandwidth caps, webhooks, invoices and account credit, and buying plans and top-ups. Purchases require an Idempotency-Key: retry the same request with the same key within 24 hours and you get the original result without a second charge. If the outcome is unknown, check GET /orders (or GET /orders/{id}) rather than buying again with a new key. The reference is at /help/proxy-management-api, and the OpenAPI specification is served at /openapi.json.

Two: username and password, or IP whitelisting. They are alternatives, not layers: a connection with valid credentials works from anywhere, and one with none works only from a whitelisted address. Your account has one credential, generated when you first subscribe, and it works on every product you hold an active plan for. IP whitelisting also works on every product: add an IPv4 address on the Auth & IP Access tab of Account Settings and connections from it need no credentials. On residential, a whitelisted connection sends no username, and the username is where country, city and session targeting live, so that exit is untargeted; on the rotating gateways you can instead send your username with an empty password to keep targeting without the secret. Our proxy endpoints are IPv4, so whitelist the address your machine uses for IPv4 traffic. Full details: /help/authentication.

We target 99.9% uptime on Premium Datacenter Proxies. It is a target rather than a service-level commitment, and the Terms of Service set out what we do and do not undertake. Automated health checks probe every provider every 5 minutes and alert us to degradation, and the current state of each product is on the status page.

It depends on the product. Residential and both rotating products choose an exit for each HTTPS tunnel or SOCKS5 connection, and for each plain-HTTP request. Requests that reuse an open HTTPS tunnel keep its exit until the client closes it, and a later choice can land on an address you had before. Add a session segment to the username to hold one exit on purpose. Dedicated and shared datacenter addresses are static: they keep their IP until you refresh it from the dashboard or the API.

Public API: 60 requests per minute per API key, the same on every plan. Dashboard: 120 requests per minute per session. Sign-in and password endpoints: 10 per minute, as brute-force protection. None of these caps the requests you send through the proxies.

The dashboard and the API are TLS-only. Proxy traffic over HTTPS runs as an opaque CONNECT tunnel, so the TLS session is between your client and the destination and we relay encrypted bytes without terminating them. Accounts authenticate by password, and additionally by IP whitelist on every product; team access is split across four roles, and API keys are scoped and individually revocable.

Yes, each with its own scopes (read:proxies, read:usage, write:orders, read:billing and the rest), so a key issued to a script that only reads usage cannot buy anything or create proxy logins. Any key can be revoked on its own without disturbing the others.

We keep bandwidth usage and a record of each connection: when it happened, the login or whitelisted address used, the IP you connected from, the exit IP, the destination host name and port, and bytes transferred. Records are kept for up to 12 months. We do not inspect, log, or store the content of your traffic, full URLs, or headers. See the privacy policy for details.