# Nebius Group N.V. > Clarifo company profile — qualitative business description generated from > the company's filings. Financial statements, charts and ratios are > available on Clarifo (https://www.clarifo.com/en/companies/Nebius Group N.V.). ## Overview Nebius Group N.V. is an Amsterdam-headquartered AI cloud company that builds infrastructure and software for training, deploying, and serving artificial intelligence applications. Its business combines full-stack cloud services with large-scale compute capacity delivered from a global network of data center and partner infrastructure, and it also owns adjacent technology businesses such as Avride and TripleTen. ## Products & services • AI cloud infrastructure and compute capacity • GPU-based training and inference environments • Full-stack platform for AI model development and deployment • Dedicated long-term capacity contracts • Autonomous vehicle and delivery robot technology • Edtech reskilling platform for tech careers - **AI cloud infrastructure** (80%) — GPU-backed cloud capacity, data centers, and compute services for AI workloads. - **Software stack and platform services** (10%) — Software, orchestration, and deployment tools used across AI training and inference. - **Long-term capacity contracts** (5%) — Committed customer capacity agreements that reserve infrastructure over multi-year terms. - **Adjacency businesses** (5%) — Other technology businesses including autonomous vehicles, delivery robots, and edtech. - AI cloud infrastructure and compute capacity - GPU-based training and inference environments - Full-stack platform for AI model development and deployment - Dedicated long-term capacity contracts - Autonomous vehicle and delivery robot technology - Edtech reskilling platform for tech careers ## Customers Nebius sells primarily to startups, AI-native companies, and enterprises that need scalable GPU capacity for training and running AI models. It also serves large strategic customers that contract for dedicated compute over multi-year periods, often to secure supply and performance at scale. The adjacent businesses address separate customer groups in mobility automation and technology reskilling. - **AI startups** (primary) — Buy flexible GPU capacity and platform tools to train models and launch AI products. - **Enterprise AI customers** (primary) — Buy reliable inference and deployment capacity for production AI workloads. - **Large strategic customers** (primary) — Contract for dedicated multi-year compute capacity to secure supply at scale. - **Mobility and robotics customers** (secondary) — Use Avride technology for autonomous vehicles and delivery robot applications. - **Career reskilling learners** (secondary) — Use TripleTen programs to learn technical skills for software and data careers. - AI startups buying compute to train and deploy models - Enterprises needing inference and production AI capacity - Large strategic customers signing multi-year capacity deals - Developers and teams building AI products and agents - Learners seeking tech career reskilling through TripleTen ## Geography Nebius is headquartered in Amsterdam and operates with a global footprint built around cloud infrastructure capacity and customer delivery locations. The company serves customers worldwide, and its disclosed agreements indicate multi-location compute deployment rather than dependence on a single market. Geography matters because power, data center access, and customer demand are all tied to where capacity can be built and delivered. - Headquartered in Amsterdam, Netherlands - Global delivery footprint for AI cloud capacity - Multiple deployment locations support customer contracts - Worldwide customer base for AI infrastructure services - Geography affects power, data center, and latency economics ## Strategy Nebius is focused on building a scaled AI cloud platform through long-term capacity contracts, expansion of its compute footprint, and a software stack aimed at both training and inference workloads. The company is also using asset-level financing and infrastructure-backed funding to support capacity buildout while preserving optionality for future customer demand. Adjacent businesses and equity stakes broaden the group beyond core cloud infrastructure. - **Expand AI cloud capacity** (short-term) — More deployed GPU capacity supports larger contracts and broader customer reach. - **Deepen enterprise and AI-native product stack** (medium-term) — A stronger software layer helps win and retain customers beyond raw compute supply. - **Use disciplined financing for infrastructure scale** (short-term) — Asset-level financing can reduce dilution and match funding to contracted assets. - **Diversify through adjacent technology businesses** (long-term) — Non-core businesses can broaden the group’s technology platform and optionality. - Secure long-term customer commitments for GPU capacity - Expand global data center and compute infrastructure - Build inference and agentic AI software capabilities - Use asset-backed financing to fund capacity growth - Develop capital-efficient partnerships alongside owned assets - Maintain optionality through adjacent businesses and stakes ## Risks Nebius depends on large-scale infrastructure investment, customer concentration, and successful execution of complex multi-year deployments, all of which can create volatility if contracts slip or demand changes. The AI cloud market is also capital intensive and technologically fast-moving, so margins, pricing power, and utilization can be pressured by competition, supply constraints, and changes in customer preferences. - **Large customer concentration** [high] — A meaningful share of future capacity is tied to a few strategic contracts. - **Infrastructure buildout execution** [high] — Revenue depends on timely delivery and commissioning of data center capacity. - **Capital and financing risk** [high] — The business requires continued access to large amounts of funding for expansion. - **Vendor and hardware dependence** [medium] — AI cloud offerings rely on high-performance GPU supply and platform ecosystems. - **Competitive pricing pressure** [medium] — Hyperscalers and specialized cloud providers can compress pricing and utilization. - Customer concentration in large multi-year contracts - Execution risk in building and delivering GPU capacity on time - Capital intensity and funding needs for infrastructure growth - Technology and vendor dependence on NVIDIA platform cycles - Competition from hyperscalers and specialized AI cloud providers - Demand uncertainty if AI infrastructure spending slows ## Accounting Nebius’s reporting is heavily influenced by revenue timing under long-term infrastructure contracts, where recognition depends on delivery and service periods rather than contract signing alone. Investors should also watch depreciation of GPU and data center assets, capitalized buildout costs, and fair value or impairment judgments on investments and adjacent holdings, because these can materially affect reported earnings and book value. - **Revenue recognition on capacity contracts** — Affects timing of reported revenue and backlog conversion - **Depreciation of infrastructure assets** — Drives reported margins and earnings through non-cash expense - **Capitalized development and buildout costs** — Affects amortization, impairment risk, and cash flow presentation - **Equity investments and fair value judgments** — Can create volatility in other income and equity carrying values - Revenue recognition under multi-year capacity contracts - Timing differences between contract signing and service delivery - Depreciation of GPU infrastructure and data center assets - Capitalized buildout costs and related asset lives - Fair value and impairment of equity stakes and other holdings --- *Last updated: 2026-09-28T12:13:24.417287+00:00*