Dedicated Datacenter Proxies with Unlimited Bandwidth, on IP Blocks We Own

# HTTP(S)
curl -x http://USER:PASS@203.0.113.10:4443 https://api.ipify.org

# SOCKS5 - same credential, different port
curl --socks5-hostname USER:PASS@203.0.113.10:1443 https://api.ipify.org

# Both return 203.0.113.10: the address you connected to is the
# address you egress from.

import requests

PROXY = "http://USER:PASS@203.0.113.10:4443"
r = requests.get("https://api.ipify.org",
                 proxies={"http": PROXY, "https": PROXY}, timeout=30)
print(r.text)  # -> 203.0.113.10

# Spread load across your own allocation
MY_IPS = ["203.0.113.10", "203.0.113.11", "203.0.113.12"]
for ip in MY_IPS:
    p = {"https": f"http://USER:PASS@{ip}:4443"}
    print(requests.get("https://api.ipify.org", proxies=p).text)

// HTTP and SOCKS5 are different transports and need different clients.
// fetch takes an undici dispatcher; a SOCKS agent is not one, so SOCKS
// goes through node:https. Looping one fetch over both cannot work.
import { fetch, ProxyAgent } from "undici";
import { SocksProxyAgent } from "socks-proxy-agent";
import https from "node:https";

const dispatcher = new ProxyAgent("http://USER:PASS@203.0.113.10:4443");
const res = await fetch("https://api.ipify.org", { dispatcher });
console.log(await res.text());

const agent = new SocksProxyAgent("socks5://USER:PASS@203.0.113.10:1443");
https.get("https://api.ipify.org", { agent }, (r) => r.pipe(process.stdout));

package main

import (
	"fmt"
	"io"
	"net/http"
	"net/url"
)

func main() {
	p, _ := url.Parse("http://USER:PASS@203.0.113.10:4443")
	c := &http.Client{Transport: &http.Transport{Proxy: http.ProxyURL(p)}}
	r, err := c.Get("https://api.ipify.org")
	if err != nil {
		panic(err)
	}
	defer r.Body.Close()
	b, _ := io.ReadAll(r.Body)
	fmt.Println(string(b))
}

Do you really own the IPs?

Yes. The datacenter, shared and rotating products run on IP blocks allocated to and announced by Node4; whois on any address will show it. The residential product is a partner network and is labeled as one.

Are the IPs shared with anyone else?

Not on this product. Each dedicated datacenter IP is assigned to one customer at a time. If you want a lower price and can accept up to 5 users per IP, shared datacenter proxies are the lower-cost option.

What locations are available?

The United States, Italy and Spain, chosen at checkout; Italian and Spanish stock is limited. For other markets, the residential product covers 170+ countries.

Is there a bandwidth limit?

No. Dedicated datacenter proxies are billed per proxy per month with unlimited bandwidth.

Is this a forward proxy or a reverse proxy?

A forward proxy. Your client sends its requests through our address to reach a target, which is what people mean when they buy proxies by the IP. A reverse proxy is the opposite arrangement and a different category of software: it sits in front of servers you already own, accepting requests on their behalf for caching, TLS termination or load balancing. If that is what you need, you want a load balancer in your own stack; nothing in this catalog replaces one.

Do you sell ISP proxies?

No. An ISP proxy is an address registered to a consumer internet provider but hosted in a datacenter, which buys some of residential's trust at closer to datacenter speed and price. We do not sell one. What we sell sits either side of it: dedicated datacenter, which is the fastest and cheapest option here and openly registered to a hosting range, and residential, which egresses through real consumer connections. If your target rejects hosting ranges, residential is the product that addresses that, and buying more datacenter IPs will not.

Can I get SOCKS5?

Yes, on every node, using the same credential as HTTP(S).

Do these proxies work with HTTPS sites?

Yes, and without installing any certificate. An HTTPS request opens a CONNECT tunnel through the proxy: your TLS session runs end to end between your client and the destination, and the proxy relays encrypted bytes it cannot read. Plain HTTP works over the same port, and SOCKS5 tunnels either protocol.

How fast are dedicated datacenter proxies?

We publish no latency number, because latency depends mostly on the route between the proxy and your target, so a single figure measured against one destination would not describe your case. Structurally, these are machines on datacenter uplinks with no consumer last mile, no gateway hop and no per-GB billing. Measure from the smallest tier against your own target.

What happens if one of my IPs gets blocked by a target?

Blocks are per target, so start by spreading that workload across your other addresses and lowering the request rate against the site. If an address stays dead for the job you bought it for, replace it within your plan's monthly replacement allowance, where it has one; the new address is drawn from our owned pool and has its own history. Dedicated IPs can be blocked like any other; what you get is that no other customer is sending traffic from your addresses while you hold them.

Why choose this over your rotating product, which uses the same owned blocks?

Stability of identity and the billing model. Dedicated datacenter gives you named addresses that stay the same: what you need when a partner API whitelists your egress, when sessions must persist for days, or when you want one customer per address. Rotating gives you breadth of addresses behind one endpoint, billed per gigabyte. Sustained heavy transfer favors a flat per-proxy price; bursty wide-net work favors per-GB. Plenty of setups use both.

How do I get my proxy list, and how does authentication scale across it?

The dashboard's Proxies page lists every address with its host and port, and the same list is available from the public API, so deploy scripts can pull the current allocation instead of hardcoding it. Authentication does not multiply with size: one credential, or one whitelisted source address, works across every proxy you own, while the per-connection ownership check keeps anyone else's credential from using yours.