Renting the Internet's Backbone: How IPv4 Leasing Actually Works in 2026

The internet has a plumbing problem. Every device, server, and service that communicates across it needs a globally unique numerical address — and the pool of those addresses, fixed by design at about 4.3 billion, ran dry at the source over a decade ago. Since 2011, when the Internet Assigned Numbers Authority handed out its last large allocation, organizations that need IP space have had to acquire it the old-fashioned way: from someone who already has it.

That secondary market has matured in ways most people outside the networking industry don't realize. There are brokers, marketplaces, transfer registries, escrow services, and — increasingly — a leasing sector that looks less like a niche workaround and more like a legitimate infrastructure service. For many organizations, leasing IPv4 addresses is now a routine line item, the same way you'd lease server capacity or office space rather than buy it outright.

What's driving that? A few things. Purchase prices, while currently at historic lows of $10 to $25 per address, still add up quickly when you need thousands of them. Project-based deployments — a temporary campaign, a new service in a new market, a proof-of-concept that might not survive the year — don't justify permanent ownership. And some organizations simply don't want to deal with the registry compliance, routing configuration, and blacklist management that comes with owning a block. Leasing offloads most of that to someone else.

"The number of entrances into the market that want to sell IP addresses is decreasing at a faster rate than the demand for IPs is increasing."

— Jake Brander, CEO, Brander Group

The supply picture makes this more urgent over time, not less. The pool of addresses available for transfer has shrunk from 44.8 million in 2015 to around 18.6 million by mid-2024 — a contraction of nearly 60%. Large contiguous blocks, the kind that serious infrastructure operators need, have become genuinely scarce; availability of those dropped 42% in 2024 alone. Addresses deployed into live infrastructure almost never come back. The annual attrition rate is around 5%, meaning the market is slowly consuming itself. The organizations that need addresses have to compete for a shrinking resource, which is precisely why leasing — locking in access to a block without permanently removing it from circulation — makes structural sense for everyone involved.

What You're Actually Paying For

IPv4 lease rates currently average between $0.38 and $0.47 per IP per month, depending on who you're working with and what's included. In practice, that means a /24 block — 256 addresses, the smallest unit that can be independently routed — costs somewhere between $97 and $120 a month. Scale up to a /20 (4,096 addresses) and you're looking at roughly $1,550 to $1,650 monthly. The math is relatively predictable once you know your block size and your provider.

What varies considerably is what that monthly rate actually buys you beyond the addresses themselves. At the lower end of the pricing spectrum, you get the IP space and not much else — configuring RPKI, updating IRR records, managing BGP announcements, dealing with geolocation databases, and handling any blacklist issues that emerge are entirely your problem. That's fine if you have a capable network operations team. It's a significant hidden cost if you don't.

Full-service providers bundle those operational tasks into the lease price. Brander Group, which operates the IPv4 Connect marketplace and whose CEO Jake Brander personally owns approximately 600,000 leasable addresses, charges around $0.47 per IP — slightly above the industry midpoint — but includes blacklist scanning across more than 100 global lists, RPKI and IRR configuration, BGP support, and geolocation management. For a /24 at $120 a month, that's a meaningfully different product than the bare-address lease you'd get from a cheaper automated platform. For organizations evaluating their options, a direct comparison of auction-based platforms versus full-service brokers illustrates how quickly the apparent savings from a lower per-IP rate can disappear once operational overhead is factored in.

"Without what we do, the Internet will stop working at some point because there are no IP addresses. IP addresses are the only thing driving the growth of the entire Internet right now."

— Jake Brander, CEO, Brander Group

The Platforms Worth Knowing

The leasing market has consolidated around a handful of serious players, each with a distinct model. IPXO, which launched in 2020 as a purpose-built automated leasing platform, has grown quickly — it now manages over 10.9 million addresses across more than 1,500 networks and covers all five regional registries. Its automated abuse management resolves the vast majority of issues without human intervention, which matters for organizations worried about reputational risk from shared address space. At around $0.39 per IP, it sits at the more affordable end of the market and suits organizations comfortable running their own network configurations.

IPv4 Depot, which has been in the market for over 13 years, takes a tiered approach — Starter, Professional, and Enterprise service levels that let customers match their support needs to their budget. It's an ARIN-approved facilitator with a long track record, and it serves a meaningful slice of the market, particularly for proxy IP use cases. IPv4 Global, now operating under Hilco Streambank and backed by a significant Japanese financial conglomerate following a $776 million acquisition in 2025, brings considerable institutional resources to the table. Its origins as an auction marketplace mean that its leasing model retains some of that price variability — which works well for buyers comfortable with dynamic pricing and less well for those who need predictable monthly costs.

Prefixx, founded in 2018, takes a no-fee-to-buyers approach that can reduce the total cost of engagement for lease customers. It covers ARIN, RIPE, APNIC, and LACNIC and serves a recognizable client list. For straightforward leasing needs with flexible terms, it's a reasonable option in the mid-market.

The honest difference between these platforms comes down to a single question: how much of the operational work do you want to handle yourself? Automated platforms offer lower per-IP rates but assume you know what to do with the addresses once you have them. Full-service providers cost more per IP but remove the configuration burden entirely. For a detailed look at how service depth translates into real-world cost differences, the comparison between mid-market brokers and full-service providers is worth reading before committing to either model.

The One Risk People Consistently Underestimate

Address cleanliness is the unglamorous factor that separates a good lease from a bad one, and it doesn't get nearly enough attention in the abstract comparisons people make between platforms.

IPv4 addresses accumulate history. A block that spent time hosting spam infrastructure, phishing pages, or botnet command servers will carry blacklist entries that follow it wherever it goes. Those entries don't transfer away when the addresses do. If you lease a dirty block, your outbound email gets rejected, your traffic gets filtered by upstream providers, and unwinding the damage takes months of remediation work with individual blacklist operators — assuming they'll respond at all. The operational cost of inheriting a compromised block can vastly exceed whatever you saved on the lease rate.

Reputable providers scan blocks against major global blacklists before provisioning and can remediate issues before you ever see them. The question to ask any potential provider is specific: how many blacklists do you scan against, and what's your remediation SLA if something is found after provisioning? Vague answers to that question are a meaningful signal about how seriously the provider takes address quality.

Does IPv6 Change Any of This?

The reasonable objection to spending money on IPv4 leasing is that IPv6 — the protocol designed to replace IPv4 with a virtually unlimited address space — should, eventually, make this whole market irrelevant. The problem is that "eventually" keeps receding into the future. Global IPv6 adoption is still below 50% after more than two decades of active promotion. The internet's installed base was built for IPv4, and the cost and complexity of migrating live systems has consistently proved greater than the theoretical benefit.

"IPv6 is like electric cars — it's been around for 20 years, it's supposedly better, but we're creatures of habit. Nobody wants to buy something new when they don't have to. If servers are already running, why spend tens of billions retrofitting entire networks?"

— Jake Brander, CEO, Brander Group

The practical upshot is that any organization deploying internet-facing infrastructure today still needs IPv4 addresses, full stop. The transition will happen — it is happening — but not fast enough to matter for a procurement decision you need to make this quarter. Leasing gives you access to the address space you need now, without locking you into a permanent purchase at a time when the long-term trajectory of IPv4 pricing is genuinely uncertain. That optionality has real value, particularly for organizations that expect their IP requirements to change as the broader transition unfolds over the next several years.

What the market looks like on the other side of that transition is anyone's guess. What it looks like right now is a functioning, competitive infrastructure services sector with real price differences between providers, real operational differences in what those providers deliver, and real consequences for choosing on price alone. The organizations that get this right treat it the same way they'd treat any infrastructure procurement — matching the provider model to their actual operational capabilities, rather than optimizing for the lowest number on the invoice.