The Thesis
Nebius Group is an AI infrastructure company that builds and operates massive GPU clusters for developers and enterprise AI teams. Revenue reached $0.53 billion in fiscal year 2025, representing a pivot into the hypergrowth phase following its separation from legacy business assets. The successful divestiture of its Russian operations and the shift to a pure-play European AI cloud provider is the structural shift that enables its current trajectory.
The investment case for Nebius Group boils down to four specific things.
We think the price already reflects the growth that is realistically achievable here. While the revenue trajectory is exceptional, the current valuation requires near-perfect execution on both GPU deployment and customer retention. The case only strengthens if Nebius can prove its custom software stack delivers performance advantages that the larger cloud providers cannot easily replicate. For long-term investors, waiting for a more reasonable entry point is the prudent move.
Numbers at a Glance
What does it do?
Nebius Group is a hypergrowth business that earns money by renting high-performance computing power to AI developers and researchers. The company operates specialized data centers packed with thousands of NVIDIA GPUs. Customers pay for access to these clusters via the Nebius AI cloud platform to train large language models and run complex AI inference. By owning the full stack, from the physical data center design to the cloud software, Nebius extracts a fee for the speed and specialized support it provides to the AI industry.
Where does revenue come from?
The vast majority of revenue comes from providing GPU-as-a-Service to AI-focused enterprises and startups. The company also generates revenue from developer tools, data labeling services (Toloka), and its autonomous driving division (Avride). Geographically, the business is centered in Europe, with its primary data center located in Finland and a headquarters in Amsterdam.
Who are its customers?
Nebius Group serves artificial intelligence startups, research labs, and large enterprises that require massive computing power for model training. The customer base is currently concentrated among high-end AI developers who cannot wait for capacity from larger cloud providers. While the company does not disclose a total user count, its Q1 revenue reached $0.40 billion, a significant jump from $0.06 billion just one year ago. This revenue spike indicates a rapid onboarding of large-scale enterprise clients who commit to multi-month or multi-year compute contracts.
What gives it staying power?
The company's staying power comes from its proprietary software stack that optimizes how thousands of GPUs talk to each other. This deep technical integration makes it difficult for customers to move their workloads to a generic cloud provider.
Where is it headed?
Management is betting heavily on becoming the dominant independent AI cloud provider in Europe. The focus is on building "sovereign AI" infrastructure that keeps data within European borders while scaling capacity at a pace that rivals US-based competitors.
Revenue is in a state of hyper-acceleration, growing from $60 million to $400 million in just four quarters. This rapid climb is driven by the immediate monetization of new GPU clusters as they come online.
Cash flow is heavily negative as the company pours billions into purchasing expensive NVIDIA hardware. Free cash flow was negative $3.68 billion in 2025, reflecting a business that is trading current cash for future compute capacity.
The balance sheet remains liquid following the divestiture of legacy assets, but the high debt-to-equity ratio of 1.31x reflects aggressive expansion. Management is using leverage to fund the capital-intensive build-out of its AI infrastructure.
Nebius Group is a high-growth infrastructure play with financials that are currently defined by massive capital spending and explosive revenue scaling.
Revenue growth is scaling at an extraordinary pace, reaching $0.40 billion in the most recent quarter. This growth proves that the demand for specialized AI compute is high enough to immediately absorb every new GPU cluster the company brings online.
Free cash flow remains deep in the red, with a negative $3.68 billion outflow in the most recently completed fiscal year. Investors must monitor whether the revenue from these hardware investments ramps fast enough to bridge the gap before the company needs additional capital.
The AI cloud infrastructure market is roughly $100B today and is growing at a ~35% annual rate, on track to reach $350B by 2028. This is a high-stakes growth industry where pricing power is determined by the availability of the latest hardware and the performance of the underlying software stack. The shift toward specialized AI compute is the structural force moving workloads away from general-purpose cloud servers. Nebius Group is a challenger in this market, carving out a niche as a high-performance, independent alternative to the US hyperscalers.
The market for AI compute is brutally competitive, with the world's largest companies spending tens of billions of dollars to build the same infrastructure. While there is a temporary shortage of GPUs, long-term pricing power will depend on who can offer the most efficient software and lowest latency. Barriers to entry are high due to the extreme capital required to buy hardware, but competition among well-funded players will eventually pressure margins.
Microsoft Azure(MSFT) and AWS are the primary threats, as they can bundle AI compute with their existing software suites to lock in enterprise customers. CoreWeave is the most dangerous direct threat because it operates a similar specialized model and often receives priority GPU allocations from NVIDIA. Google Cloud also presents a risk by developing its own custom chips, which could eventually reduce demand for the NVIDIA-based clusters that Nebius provides.
Nebius is currently gaining share by moving faster than the giants and securing European capacity. The six-fold increase in quarterly revenue over the last year is definitive evidence that Nebius is winning customer workloads today.
The primary source of protection for Nebius is its proprietary software stack that manages how GPUs work together at scale. This deep technical IP creates a performance advantage that is difficult to replicate with off-the-shelf software. The company's 47.9% gross margin in a capital-intensive industry suggests that customers are willing to pay a premium for this specialized performance.
The high gross margins and explosive revenue growth prove that the company has a temporary edge in a supply-constrained market. However, the negative ROIC and massive cash burn indicate that this advantage is not yet structurally protected by switching costs or network effects. The numbers suggest a business that is executing exceptionally well in a gold-rush environment rather than one with a wide, permanent moat.
The moat is currently narrowing as competitors build out similar capacity, making proprietary software optimization the single most important signal to watch.
Delivered 6x year-over-year revenue growth in the most recent quarter.
Successfully completed a complex divestiture of Russian assets to focus on AI.
Founder Arkady Volozh remains CEO and maintains a significant personal stake.
Capital Allocation Track Record
Arkady Volozh has successfully navigated a massive corporate restructuring that most leaders would have failed to manage. The management team has demonstrated high execution by scaling revenue from almost nothing to a $1.6 billion annual run rate in one year. While the capital allocation is aggressive and risky, it is aligned with a clear, singular vision for the AI infrastructure market.
© 2026 ClearThesis.ai · Report generated on May 27, 2026
This is an AI-generated analysis for informational purposes only and does not constitute financial advice. Data and analysis may not reflect recent developments if viewed significantly after the generation date. Always conduct your own due diligence before making any investment decisions.