Rise of the Neocloud: Top 10 Providers and Adoption Trends

With the tech world once again in the midst of a major shift, a new group of contenders has entered the market, offering capabilities the traditional hyperscalers can’t match. Often dubbed "neoclouds", these new providers range from developer-focused platforms to GPU-centric infrastructure clouds.

Can these new clouds compete with 100 billion dollar incumbents in a space that's already changing at breakneck speed? What advantages are they bringing to the table that can pry major enterprise deals from the grip of the Big Three hyperscalers?

Let's take a look at the top neocloud providers and then dive into the mechanics of this change!

Top 10 Neocloud Providers

Driving Adoption: Price, Performance, Capacity, and Beyond

Across the industry, a clear trend is taking shape. Organizations are stepping back from the all-in-one-cloud mindset and embracing providers that better fit their needs.

Among these priorities, several consistent drivers stand out:

Cost Efficiency

The cloud has become so expensive that many organizations have found bills to be unsustainably high or worse… unpredictable. In a recent study, 81% of IT leaders were directed by execs to cut or freeze hyperscale cloud spending. A major complaint is just the sheer complexity of the billing process and its opacity. In fact, 42% of businesses struggle to predict monthly cloud costs , and 28% have been hit with shockingly large bills. Many neocloud providers offer a much more straightforward billing model, with flat pricing that is often far below the cost of hyperscalers. You may have even seen some of these success stories as 37 signals highly publicized their move from the cloud to on-prem with an estimated $7 million saved by 2028, and Ahrefs saved a whopping $400 million over 3 years by using colocated servers.

Performance and Capacity

One advantage of being new to the game is having the best hardware and the most capacity. Neocloud providers often offer specialized or newer hardware that the hyperscalers cannot supply quickly or cost-efficiently. The clearest case is AI and GPU-intensive workloads: hyperscalers have faced GPU shortages and long wait times for top-end GPUs, whereas neoclouds like CoreWeave, Lambda Labs, and RunPod deliver immediate access to cutting-edge GPUs. This performance and capacity-driven adoption is further highlighted by firms like OpenAI going to CoreWeave to augment shortcomings in Azure.

Similarly, high-performance bare-metal cloud (PheonixNAP, OVHCloud metal, Latitude, Servers.com ) attracts customers needing consistent high I/O or low latency (think trading platforms and gaming networks) where dedicated hardware outperforms virtualized cloud VMs. In cases like this, it's all about how much performance these organizations can squeeze out of every dollar spent, and it's hard to compete against bare metal.

Flexibility and Control

Many neocloud providers offer a simplified, developer-centric experience. This is a welcome respite from the growing complexity of the hyperscalers, with their ever-expanding catalog of services and configurations. Many organizations feel like they need dedicated consultants just to get things right. Providers like Linode and Vultr focus on core compute/storage primitives without the steep learning curve, which can be appealing to teams who want to get up and running more quickly without needing provider-centric expertise.

Having more control also means freedom from certain cloud limitations. With bare metal, for example, customers can run any software stack and avoid virtualized compute. And with the neocloud providers, many customers are finding themselves supported by a much more hands-on team that’s ready to help when things go wrong. In short, many feel the neocloud experience renews the simplicity and autonomy of owning infrastructure, while still delivering it "as-a-service".

Geographic and Regulatory

There’s a growing wave of momentum that’s taken shape over the past 24 months. Compliance and regulation are accelerating in the wake of political and social change across what was once seen as a global system that “worked.” Now, many organizations are looking to move their data away from U.S.-owned hyperscalers to more “sovereign” locations as they scramble to meet the moment head-on. The neoclouds are rising to meet this shift as well, offering streamlined pathways away from the hyperscalers and, in some cases, leveraging the Cycle platform to build new private clouds.

In Asia, providers like Alibaba Cloud or Tencent Cloud (not profiled above, but significant in their regions) fill the gap for companies operating in China or Southeast Asia, where the U.S. hyperscalers have limited presence or face data rules. Thus, geography and regulation drive a more distributed cloud market. The result is companies mixing and matching providers that best meet local requirements, rather than relying solely on a distant hyperscaler region.

Outside of these 4 core drivers is the need for added resilience. While many enterprises are waking up to the risks of being all-in on a single cloud, newer startups and nimble SMBs are already a few years ahead. Many are running workloads across several different providers, with even more planning to repatriate data and workloads from the hyperscalers into their own multi or hybrid clouds.

The Shift in Cloud Purchasing

Where is this shift towards the neoclouds really coming from, or is it even there at all?

Investment and Growth

S&P Global reported that over $10 billion was invested in neocloud providers last year. Coreweave itself has seen such a dramatic rise in valuation, including not only an IPO but a 300% rise in value since the offering. Lambda Labs claims to already serve 50,000 customers. All this suggests that cloud buyers (from startups to enterprises) are diversifying their spend. Indeed, the big three’s collective market share has stopped growing as fast, hovering around ~63% the past couple years, implying the “other” providers and self-managed infrastructure are capturing the rest.

Hyperscalar Dependence

When Azure moved to secure additional GPU capacity from Coreweave in 2023-24, it recognized that specialized providers had an edge in quickly supplying AI hardware, and the hyperscalers weren't afraid to join in on the opportunity. AWS has responded to pressure by releasing wave after wave of billing tools to try and simplify FinOps on the platform, and GCP has promised a new sovereign cloud in the EU… possibly a little too late.

Publication of Cloud Exits

The award for most publicized cloud exit definitely goes to DHH and the team at 37 Signals. He recently shared a video on LinkedIn of them fully closing their AWS account… a post that happened to directly coincide with the major AWS outage in US-East-1 on October 19th, 2025. The enterprise has taken notice, and over 80% of enterprises plan to repatriate workloads in the coming years.

It All Comes Down to Value

The rise of neoclouds reflects a more value-driven cloud market where decisions hinge on cost, performance, and flexibility. Instead of asking, “What’s the best way to do this on our existing provider?” companies are now asking, “Who offers the best balance for this workload?” And more often than ever, the answer isn’t a hyperscaler.

• Price and Simplicity: For SMBs and general-purpose workloads, affordability and predictability prevail. Providers like DigitalOcean, Linode, Vultr, and Hetzner succeed by offering lean, transparent services that eliminate waste and complexity. With most cloud users overspending on unused capacity, right-sized, no-surprise pricing is compelling.

• Capacity and Performance: For AI, analytics, and HPC workloads, access to high-performance resources is critical. Neoclouds like CoreWeave and Lambda Labs deliver both cost efficiency and availability, offering the GPUs and compute power that hyperscalers can’t always supply on demand. In today’s AI economy, capacity has become the new currency.

• Optimized Mix: Many organizations now blend providers to achieve the best cost-to-performance ratio. Case studies such as Klink AI and Feedonomics show that migrating from hyperscalers to specialized neoclouds can yield several times better performance per dollar while maintaining throughput and scalability.

This shift may very well be the end of the "one-size-fits-all" cloud era. At Cycle, more than half of our customers are consuming compute from multiple providers. Many of our users have been quick to adopt the new virtual providers feature , moving workloads back on-prem or to a bare metal neocloud. We welcome this next phase of cloud computing and can't wait to see what you'll be building!