Introduction
Mobile operators have spent the first five years of 5G largely leaning on existing 4G cores. That phase is ending. A new wave of 5G Standalone deployments is reshaping the mobile core network market, pushing buyers toward cloud-native software, automated operations, and edge-aware architectures that unlock services most people associate with next-generation connectivity. An industry tracker estimates that about 70 mobile network operators have now deployed 5G Standalone across 39 countries and territories. In 2025 alone, five fresh national Standalone networks went live, including Orange in France, Romania, and Slovakia, Vodafone in Spain, and O2 in Czechia. Regionally, there are five public Standalone networks in North America, 26 in Europe, seven in the Middle East and Africa, 13 in Northeast Asia, 13 in Southeast Asia, and six in Latin America. These numbers matter because Standalone brings the 5G Core that enables network slicing, lower latency paths, and more predictable performance. That, in turn, changes how enterprises source connectivity for applications such as private networks, fixed wireless access, computer vision, logistics tracking, and real-time control. If you are a telecom buyer, a CIO considering FWA for branch connectivity, or an operations executive automating a plant, Standalone is the inflection point that moves 5G from fast mobile broadband to a programmable infrastructure platform.
What changed in 2025
Three things converged. First, operators had enough 5G-capable devices and traffic to justify moving voice and data onto the 5G Core at scale. Second, Standalone software matured. Vendors stabilized cloud-native network functions, improved observability, and hardened CI or CD pipelines that keep cores patched and upgraded with less risk. Third, enterprise demand shifted. Organizations that piloted private LTE and early 5G Non-Standalone use cases began asking for deterministic performance, traffic isolation, and APIs to integrate connectivity into their applications. The 2025 launches in France, Romania, Slovakia, Spain, and Czechia show how Western and Central Europe are pushing to close the gap with Asia, where several countries have been running Standalone nationally for longer. In parallel, device firmware updates enabled voice over New Radio in more handsets. Voice is not glamorous, but moving it to Standalone removes the last major Non-Standalone anchor in many footprints and simplifies network engineering. The net result is a mobile core market that is buying cloud-native 5G Core stacks, user plane functions that can run closer to the edge, and policy or charging systems designed for granular service tiers rather than one-size-fits-all plans.
5G Standalone, in plain language
There are two big pieces to understand. First, 5G Non-Standalone uses a 4G core for control functions and anchors many sessions through that 4G infrastructure. Speeds can be high, but the system is not built for the most advanced 5G features. Second, 5G Standalone uses a new cloud-native 5G Core based on a service-based architecture. That core is designed for ultra-reliable low latency, better security primitives, and automated scaling. The main building blocks are the Access and Mobility Management Function, Session Management Function, User Plane Function, Policy Control Function, Network Exposure Function, and associated charging and authentication services. Instead of large, monolithic virtual machines, these functions deploy as microservices that can scale horizontally, fail independently, and be updated more frequently. The architectural difference is what enables network slicing, quality-of-service tiers that can be enforced per application, and the ability to steer traffic to local user plane anchors near an enterprise site or a city edge zone. With Standalone, the core and the radio are designed to work together for predictable latency, not just peak throughput.
Where the money is going in the mobile core
As Standalone adoption grows, spending concentrates in four areas. First is the converged 5G Core. Operators want a single control and policy fabric that can anchor 5G, interwork with 4G for roaming and legacy coverage, and expose clean APIs to both internal engineering teams and enterprise developers. Second is the user plane. Moving packets is where the cost and performance rubber meets the road. Buyers are evaluating data plane acceleration, smart-NIC support, and the ability to spin up user plane instances in regional clouds or on bare metal near city centers. That requires policy engines that are more granular than consumer plans, plus online charging systems that can meter small slices and on-demand features. Fourth is assurance. Cloud-native cores introduce a new observability challenge. Operators are investing in real-time telemetry, synthetic testing, and AI-assisted operations that can detect anomalies before customers notice.
The state of deployments, region by region
The headline numbers provide a useful map for planning. North America counts five public Standalone networks. The region is prioritizing spectrum refarming and FWA scale, so architectures often feature large, centralized cores combined with regional user plane breakouts to keep latency in check for gaming and video conferencing. Europe is the busiest, with 26 Standalone networks. The 2025 launches by Orange in France, Romania, and Slovakia, Vodafone in Spain, and O2 in Czechia underline a broader push to even out coverage and prepare for enterprise slicing. Beyond these examples, many European networks are taking a cloud-first approach to the core, but with data residency controls that keep customer data within national borders. The Middle East and Africa have seven Standalone networks. Here, Standalone often rides alongside ambitious smart city and industrial zone projects. Government digitization and greenfield builds are catalysts, while rural coverage and backhaul economics shape FWA strategies. Northeast Asia has 13 Standalone deployments and remains a bellwether on advanced features, including voice over New Radio at scale and high device penetration for Standalone-capable handsets. Operators in this region are furthest along in moving significant traffic to Standalone and in experimenting with consumer service differentiation that goes beyond speed tiers. Southeast Asia also counts 13 deployments, reflecting a mix of tourist corridor coverage, industrial parks, and national modernization programs. Hybrid cloud cores are common, and regional data centers are being linked with metro edges to support low-latency gaming and video. Latin America has six Standalone networks. These tend to emphasize FWA as an alternative to copper and cable in urban edges and fast-growing suburbs. Operators are pacing core investments with device availability and spectrum auctions, with an eye toward enterprise connectivity for logistics, mining, and agriculture.
What 5G Standalone unlocks for enterprises
Standalone is not just an operator milestone. It changes what an enterprise can reliably ask a network to do. Three categories stand out. First, deterministic performance. With Non-Standalone, performance can vary because sessions are anchored on a 4G core and traffic competes across a generalized best effort path. Standalone allows an operator to define and enforce quality for a specific application flow and to anchor user plane functions near the workload. For a factory using machine vision or a stadium running computer-assisted officiating, that predictability is the difference between a demo and a dependable production system. Second, isolation and security. Slices can be logically segregated so a retailer’s point-of-sale traffic does not mingle with guest Wi-Fi and employee devices. Exposure APIs allow enterprises to control policies in near real time, such as raising the priority of a robotic flow during a shift change. Third, mobility for fixed services. It sounds contradictory, but FWA is where Standalone quietly shines for business connectivity. A Standalone core with strong policy enforcement can make FWA behave more like a managed Ethernet tail, with set latency and throughput targets, while preserving the flexibility and speed of cellular deployment.
FWA, now with real service levels
FWA matured during the Non-Standalone phase because consumers loved the simplicity of plug-and-play broadband. The next phase is business FWA with defined service levels. Standalone can pin traffic to a local user plane, apply application-aware policies, and leverage radio features that prioritize specific flows during congestion. For a branch office, pop-up retail site, or construction trailer, the operator can carve a slice that ensures the point-of-sale system and video security do not suffer when the site hosts a community event. The combination of deterministic policies and localized user plane anchors is why many operators see FWA and Standalone as commercially linked. The same capabilities that support branch connectivity also reduce jitter for cloud gaming and improve upstream performance for live creators, which helps consumer markets.
Edge computing and the 5G Core
The mobile core market’s other growth lever is the integration of multi-access edge computing with Standalone. MEC is not simply a box in a shelter. It is a topology decision that brings compute into or near the operator domain so application packets can complete their round trip within a local metro. Standalone’s user plane placement allows an operator to steer flows into those zones. The practical question for buyers is placement and density. A metro might justify two or three edge zones because of population and workload density, while rural regions may share a regional edge to balance cost and performance. When a manufacturer or a video production firm asks for a 15 millisecond round trip budget, the calculus is straightforward. Determine how many fiber hops you can tolerate, pick user plane hosts within that distance, put the workload next to it, and monitor the path with synthetic testing. Standalone supplies the control hooks that make that design possible and repeatable.
Security and trust with a cloud-native core
Standalone is more secure by design than legacy architectures, but it changes the threat model. Service-based interfaces expose APIs that are easier to instrument and to protect with modern authentication. Subscriber privacy improves through identifiers that are not reused in the clear over the air. Signaling elements that cross public and partner networks are encapsulated and authenticated in a way that reduces spoofing. The shift to containers, however, means the core’s attack surface now includes the Kubernetes control plane, the container registry, and the CI or CD pipeline. That is why operators are embedding software supply chain controls, image signing, and runtime protection into the core’s standard operating procedures. Enterprises that integrate with exposure APIs should expect rate limiting, token management, and detailed audit logs. For sensitive applications like remote health services or financial transactions, those controls will be part of the compliance package customers evaluate.
Architecture choices buyers must make
Every Standalone program faces four fundamental decisions. First, single vendor core versus best of breed. A single stack simplifies integration and accountability, which can accelerate time to market. A multi-vendor architecture may deliver specific strengths in user plane performance, policy control, or observability. The right answer depends on the operator’s engineering culture and the maturity of its automation. Second, where to run the core. Some cores will run entirely on private infrastructure. Others will split control plane functions into public cloud regions while anchoring the user plane in operator data centers. Hybrid designs can work well, but they require consistent automation, one source of truth for configuration, and careful data-residency controls. Third, how to place and scale the user plane. The user plane is the performance lever. Buyers are evaluating DPU or smart-NIC acceleration, NUMA-aware CPU pinning, and node pools optimized for packet processing rather than general compute. Fourth, how to assure the service. Observability cannot stop at the core. It must extend into the radio access network, transport, and edge workloads. Teams need golden signals for each slice, active tests that mimic real traffic, and dashboards that detect drift from baselines.
Economics, TCO, and the energy bill
Standalone is a technology upgrade, but the decision is ultimately financial. The cost model covers licenses for core functions, infrastructure for control and user plane clusters, observability, charging, and policy systems, plus the integration and testing required to migrate subscribers and services. On the other side of the ledger, Standalone can reduce operating costs by enabling automation, faster patch cycles, and less manual work when introducing new features. It can also raise average revenue per user through new service tiers and more reliable FWA. Packet processing at high throughput consumes power. That is why user plane acceleration, workload placement, and careful tuning of container density are not just engineering choices, they are financial ones. The best programs pair finance with network engineering early, build a power model that covers typical and peak loads, and use that model to guide hardware selection and placement.
A practical migration playbook for operators
- Define the service boundary. Write down what services you will run on Standalone in the first two quarters, what you will migrate in the next four, and what you will leave on legacy systems until year two. Voice deserves its own plan because moving to voice over New Radio is often the final step that enables full Standalone coverage. 2) Build a golden path environment. Pick one cloud stack, one cluster blueprint, one observability plan, and one way to deploy network functions. Document it, automate it, and defend it from ad-hoc exceptions. 3) Place the user plane with intent. Use real device telemetry to find your latency hot spots. Put user plane instances where they will make a measurable difference, not just where you have space. 4) Harden the software supply chain. Require image signing, scan everything for vulnerabilities, and secure the CI or CD pipeline. 5) Run synthetic tests before you carry customer traffic. Test bad days by pulling nodes and links. 6) Migrate traffic in stages. Start with mobile broadband for a small region. Add FWA tiers. Bring in voice. Then open exposure APIs to a limited group of enterprise customers who can help you refine the product. 7) Measure and publish results. Track attach success rates, call setup success for voice over New Radio, user plane throughput per watt, and slice SLA compliance. Share that data internally and use it to guide investments.
A buyer’s checklist for enterprises
If you are an enterprise IT or OT leader, Standalone can influence your connectivity roadmap for the next three years. Use the following checklist during RFPs and proofs of concept. Coverage and device readiness. Ask for a map of Standalone coverage, including where voice over New Radio is available and where sessions will fall back to Non-Standalone. Confirm that your existing devices support Standalone and voice capabilities through firmware. Policy control and exposure. Deterministic performance. Agree on latency and jitter budgets for your applications. If you need 15 milliseconds round trip, make sure the provider can place the user plane and edge workload close enough to meet that goal. Security posture. Review how your traffic is isolated from other customers, how tokens are managed, what logs you will receive, and how long they are retained. FWA service tiers. If you are replacing a wireline tail, ask for business FWA packages with clear throughput and jitter targets, not just peak speed. Integration and support. Clarify who handles SIM lifecycle, device onboarding, and support escalation. Ask for a runbook that shows how incidents are handled and how your team can trigger priority changes during an outage or a special event. Pricing transparency. Make sure you understand what happens when you scale. Look for rate cards that show the cost per additional slice, per additional policy, and per gigabyte of assured traffic.
The special case of private networks
Standalone matters even if you plan to deploy a private cellular network. Many private LTE systems met their initial goals for coverage and basic mobility, but they did not always deliver deterministic performance for machine vision or control loops. As private 5G matures, it inherits Standalone’s strengths. A private network can either host its own 5G Core or federate with a public Standalone core for roaming and shared services. The design choice depends on what you need to keep on-site and how tightly you want to integrate with public coverage. A manufacturer might host the core on-premises to satisfy data residency requirements, then use a roaming plan for visitors and for assets that leave the premises. A port might use a public Standalone core with local user plane instances and a dedicated slice that covers the entire harbor.
Obstacles, risks, and how to mitigate them
A rigorous test plan and a small, cross-functional tiger team that includes radio, core, transport, and security will shorten the time to root cause. Roaming is the third. As more networks go Standalone, roaming agreements must catch up so sessions do not fall back to legacy behavior that breaks advanced features. Commercial teams should start those conversations early and include clear test criteria. Invest in platform engineering, GitOps practices, and blameless postmortems so teams learn from incidents quickly. Finally, messaging risk is real. Some markets equate Standalone with instant, ubiquitous miracles. Success requires careful expectation management. Publish what has improved, where it works, and what is next on the roadmap.
Metrics that actually matter
Operators and enterprise buyers should track a handful of metrics that predict whether a Standalone rollout is delivering value. Percent of traffic anchored on Standalone, split by consumer, FWA, and enterprise. Voice over New Radio call setup success rate and drop rate. User plane throughput per watt, which indicates whether your infrastructure tuning is paying off. Slice SLA compliance, measured as the percent of intervals that meet latency and jitter targets. Time to deploy a new policy or slice, which reflects the maturity of your automation. Exposure API usage, including the number of enterprise customers integrating policy controls into their applications. Mean time to detect and mean time to recover for incidents in the core and at the edge. Publish these metrics and use them to guide investment, not just to report after the fact.
Country launches that signal a turning point
The 2025 list of new Standalone networks in Europe is significant and instructive. Orange lighting up Standalone in France, Romania, and Slovakia shows how a multi-country operator can roll a common core blueprint across diverse markets. It also demonstrates the advantage of treating automation and observability as first-class investments. Vodafone’s Standalone launch in Spain highlights how a major market can use the new core to expand business FWA and enhance consumer streaming and gaming experiences. O2 bringing Standalone to Czechia continues the Central European momentum, where operators are using Standalone as a lever to compete on enterprise solutions and on overall service quality. These launches tell a broader story. Europe’s footprint of 26 Standalone networks, the largest of any region today, gives enterprises a real chance to standardize their connectivity strategy across the continent. A multinational manufacturer can evaluate similar service tiers and policy controls in several countries, which reduces integration work and accelerates rollouts.
How to decide if Standalone should change your roadmap this year
If you are an operator, the answer is yes, with a focus on measurable outcomes. If you are an enterprise, the answer depends on your application mix. Use these decision cues. You should prioritize Standalone services if your applications suffer from jitter, if you need traffic isolation, if you are replacing slow or expensive wireline tails with FWA, or if you want to integrate connectivity controls into your software. You can pace Standalone adoption if your use cases are tolerant of latency and you mainly need coverage and basic mobility. Even then, pilot Standalone in a subset of locations so you have a real benchmark. The tipping point often comes when a single business unit demonstrates a clear productivity win or a new revenue stream. For example, a distribution center that uses computer vision on a Standalone slice to reduce error rates, or a stadium that sells premium low-latency connectivity to event production teams. Those proofs unlock executive sponsorship to scale.
Frequently asked questions
Do I need Standalone for network slicing. Yes, slicing relies on the 5G Core and its policy enforcement. Non-Standalone can simulate tiers with QoS profiles, but Standalone provides end-to-end control and better telemetry. Will 4G cores go away. Not immediately. Interworking will persist for roaming and for areas where Standalone coverage is still maturing. Over time, more traffic will anchor on Standalone as voice over New Radio spreads and device penetration improves. Is Standalone only for big cities. No. Once it is deployed, rural areas can benefit, especially through FWA and through localized user plane anchors in regional hubs. Does Standalone lower my costs. It can, depending on execution. Automation reduces operational toil, and new service tiers can raise revenue. That said, the user plane consumes power and must be placed carefully. Do I need to move workloads to the edge to benefit. Not always. Many gains come from simply anchoring the user plane closer to the workload and enforcing policies. For very tight latency budgets, edge placement helps, but it is not a universal requirement. How do I test a provider’s claims. Ask for a pilot with synthetic tests that mimic your traffic, then compare results against your metrics under load and during induced failures.
Conclusion
Standalone is now a broad reality rather than a slide in a strategy deck. Around 70 operators in 39 countries and territories have deployed it, with 2025 adding five new European launches from Orange in France, Romania, and Slovakia, Vodafone in Spain, and O2 in Czechia. Regionally, the footprint includes five in North America, 26 in Europe, seven in the Middle East and Africa, 13 in Northeast Asia, 13 in Southeast Asia, and six in Latin America. This distribution lines up with different regional priorities, from advanced consumer services and enterprise slicing to smart city programs and business FWA. For the mobile core market, the implications are direct. Spending is shifting toward cloud-native cores, edge-aware user planes, modern policy and charging, and deep observability. For enterprises, the opportunity is practical rather than abstract. Standalone can bring deterministic performance to applications that stalled on Non-Standalone. It can turn FWA into a credible replacement for some wireline tails. It can expose connectivity controls in a way that lets software orchestrate the network alongside compute and storage. Success will come from clear service definitions, disciplined user plane placement, strong software supply chain practices, and a metrics-driven operations culture. If you anchor your program in those fundamentals, Standalone will not just check a box. It will become the foundation for the next decade of connected applications.