# Neoclouds: The GPU Rental Companies Powering the AI Boom
## What Are Neoclouds?
If you’ve been following the artificial intelligence revolution, you’ve probably heard about the enormous demand for computing power. At the heart of this demand are graphics processing units — GPUs — the chips that make training and running AI models possible. For years, the big tech companies like Amazon, Microsoft, and Google controlled most of this computing infrastructure through their cloud platforms. But something new has emerged: **neoclouds**.
Neoclouds are specialist cloud providers that focus almost entirely on renting out GPU clusters for AI work. Unlike the major cloud platforms that offer hundreds of different services, neoclouds zero in on one thing: providing large-scale access to Nvidia GPUs and the infrastructure needed to run AI workloads. Think of them as specialized rental companies for the world’s most expensive and in-demand computer chips.
The two most prominent publicly traded neoclouds are **CoreWeave** (Nasdaq: CRWV) and **Nebius** (Nasdaq: NBIS). Several private companies also operate in this space, including Lambda, Crusoe, Fluidstack, and IREN.
## Where Did Neoclouds Come From?
Here’s where the story gets interesting. Many neoclouds have roots in the cryptocurrency mining industry. CoreWeave, the largest pure-play neocloud, actually started out as an Ethereum miner. Bitcoin miners, who already own massive data centers with powerful hardware and access to cheap electricity, have been repurposing their facilities for AI computing at an astonishing rate.
According to industry data from Bernstein, Bitcoin miners signed roughly 19 AI hosting deals totaling more than $135 billion between mid-2024 and mid-2026. That’s a remarkable pivot for an industry that was once seen as purely digital. These miners already had the land, electrical grid connections, and permits that would take years for a new company to secure — making them natural partners for AI companies desperate for computing capacity.
## How Does the Neocloud Business Work?
The neocloud model depends on three things: **chips, power, and capital**.
– **Chips**: Neoclouds buy Nvidia GPUs in massive quantities and connect them into clusters optimized for AI. Being able to secure the latest Nvidia hardware before anyone else is a major competitive advantage.
– **Power and Space**: AI computing clusters consume enormous amounts of electricity — we’re talking hundreds of megawatts per facility. They also require liquid cooling systems to prevent the chips from overheating. Neoclouds either build their own data centers or lease space from existing facilities, including repurposed crypto mining sites.
– **Capital**: GPUs are expensive and lose value relatively quickly as newer models come out. Neoclouds typically finance their chip purchases through debt, often using the GPUs themselves as collateral and backing their loans with long-term customer contracts.
Customers — which range from AI startups to major tech companies — sign agreements to reserve GPU capacity for periods of several years, or they pay hourly rates for on-demand access. The neocloud profits from the difference between what it charges customers and what it costs to buy, power, and finance the hardware.
## Why Are Neoclouds Important?
The big cloud providers like Amazon Web Services and Google Cloud have their own AI products competing for the same limited supply of GPUs. Neoclouds offer a critical advantage: they can deliver dedicated, large-scale GPU clusters faster and more flexibly than the hyperscalers, which juggle many different priorities.
Some of the biggest technology companies on Earth are themselves customers of neoclouds. Microsoft, Amazon, Google, and Anthropic (the AI company behind Claude) all lease GPU capacity from neoclouds when their own data centers can’t keep up with demand. Nvidia, the dominant GPU manufacturer, is both the primary supplier to neoclouds and an investor in CoreWeave.
## The Crypto Connection
It’s worth understanding that neoclouds are **not** cryptocurrency projects. They don’t issue tokens or operate blockchains. They are conventional companies whose shares trade on traditional stock exchanges. However, their deep ties to the crypto mining industry are a defining feature of the sector. Companies like IREN (a Bitcoin miner turned AI host) and TeraWulf represent this crossover, and several Bitcoin miners have entered into multi-billion-dollar agreements to host AI computing infrastructure.
For crypto enthusiasts, the closest equivalent to a neocloud in the decentralized world would be DePIN projects (Decentralized Physical Infrastructure Networks) like Render, Akash, and io.net. These use crypto tokens to coordinate GPU resources across a peer-to-peer network rather than operating company-owned data centers.
## Market Trouble in October 2026
The neocloud sector faced significant headwinds in October 2026. Shares of both CoreWeave and Nebius experienced sharp declines. CoreWeave’s valuation dropped from a peak of around $90 billion to roughly $34 billion, a loss of about two-thirds of its value in a short period. Nebius fell 27%, and IREN dropped 32%.
Several factors contributed to this sell-off. Credit markets grew uneasy about the heavy debt loads these companies carry. CoreWeave’s unsecured bonds began yielding around 15%, and the cost of insuring its bonds against default rose significantly in a single month. Meanwhile, research firm Gartner issued a prediction that more than 80% of neocloud providers would cease to operate independently by 2030, citing cost pressures and a lack of a differentiated business model.
## What’s Next for Neoclouds?
Despite the volatility, the fundamental demand for AI computing power remains strong. CoreWeave’s co-founder has stated that nearly all of the company’s 2027 capacity is already under contract, with commitments extending into 2028 and beyond. Whether neoclouds remain independent companies or eventually get acquired by larger tech firms or private equity groups, the infrastructure they provide will continue to be essential to the AI industry.
The biggest question going forward is whether these companies can generate enough revenue from their GPU rental operations to pay down the substantial debts they’ve taken on to build out their capacity. If they can, investors may find renewed confidence in the sector. If not, consolidation and acquisitions seem likely.
—
## Frequently Asked Questions (FAQ)
**Q: What does “neocloud” stand for?**
A: It’s a shortened form of “new cloud” — a term used to describe the new generation of cloud providers that specialize in AI computing infrastructure, particularly GPU rental, as opposed to traditional cloud platforms like AWS or Google Cloud.
**Q: Can I buy neocloud stocks?**
A: Yes. CoreWeave (CRWV) and Nebius (NBIS) trade on the Nasdaq stock exchange and are available through standard brokerage accounts. IREN also trades on Nasdaq. Several Bitcoin miners that have pivoted to AI hosting — such as TeraWulf, Cipher, Hut 8, and Core Scientific — also offer public stock trading and provide exposure to AI hosting operations.
**Q: Are neoclouds the same as crypto projects?**
A: No. Neoclouds are traditional, for-profit companies. They do not issue cryptocurrency tokens or operate on blockchain networks. Their shares are conventional stocks traded on regulated exchanges. However, many of them originated from or are closely connected to the cryptocurrency mining industry.
**Q: Why are GPUs so expensive?**
A: High-end Nvidia GPUs like the H100 and upcoming B200 are in extremely high demand for AI training and inference. The manufacturing process is complex, production capacity is limited, and the global supply chain for advanced semiconductor chips remains tight, all of which drive up prices.
**Q: What’s the risk of investing in neoclouds?**
A: Neoclouds carry significant risk. They operate with heavy debt, GPUs depreciate quickly as newer models arrive, and the AI industry is still evolving rapidly. Gartner’s prediction that most neoclouds may not survive independently adds further uncertainty. These are high-risk, high-volatility investments.
**Q: How does a neocloud make money?**
A: A neocloud earns the difference between what it charges customers for GPU access and what it costs to purchase the GPUs, power the data centers, and finance the equipment through debt. With large-scale operations, small margins per GPU can add up to significant revenue when multiplied across thousands of chips.
**Q: Will neoclouds replace traditional cloud providers?**
A: It’s unlikely that neoclouds will fully replace traditional cloud providers like AWS, Azure, or Google Cloud. More likely, they will coexist, with neoclouds serving as specialized GPU rental platforms and traditional clouds continuing to offer a broad range of services. Some neoclouds are already selling capacity directly to the traditional cloud providers.
**Q: What are DePIN tokens?**
A: DePIN (Decentralized Physical Infrastructure Networks) are crypto-based projects that aim to provide GPU computing through a distributed, peer-to-peer network rather than centralized data centers. Examples include Render (RENDER), Akash, and io.net. They target a similar market to neoclouds but use blockchain technology and tokens instead of traditional company structures.
—
## Conclusion
Neoclouds represent one of the most fascinating developments in the intersection of artificial intelligence and technology infrastructure. Born from the crypto mining industry, these GPU-focused cloud providers have grown from obscure startups into companies with multi-billion-dollar contracts and public stock listings — all within a matter of a few years. They fill a critical gap in the AI ecosystem, providing dedicated computing capacity that the major tech companies themselves increasingly rely on.
The sector is not without its challenges. Heavy debt loads, rapid hardware depreciation, and intense competition have created a volatile investment environment. Predictions of industry consolidation by 2030 add another layer of uncertainty. But the demand for AI computing power is not slowing down, and neoclouds are positioned at the center of that demand. Whether they evolve into enduring independent companies or become acquisition targets for larger players, their role in powering the AI revolution is undeniable.
As with any emerging sector, potential investors should do thorough research and understand the significant risks involved before committing capital.
Thank you for reading
