Nebius Group has announced plans to position itself as a strategic provider of AI compute capacity to hyperscale cloud operators. The company reports that hyperscalers are committing over $12 billion this year to expand their data center footprints, and Nebius has secured contracts worth $2.5 billion in aggregate to supply custom-built GPU clusters. Management highlights that these agreements include multi-year commitments for both hardware leases and ongoing systems integration services, with delivery of the first 50 petaflops of compute capacity slated for Q4.

Over the past month, Nebius shares have climbed nearly 30%, contributing to a 97% rally in the last six months as demand for AI cloud services accelerates. The company opened three new data centers in Northern Virginia, Texas, and Frankfurt in the last quarter, bringing its global facility count to eight. Strategic alliances with two of the world’s top five cloud providers are expected to drive a 45% increase in recurring revenue next fiscal year, based on internal forecasts. Nebius also reports a 60% year-over-year rise in capacity utilization, underscoring robust customer uptake.

Despite the bullish momentum, Nebius faces significant execution and capital-structure risks. The rapid expansion plan calls for $4 billion in incremental capital expenditures over the next 18 months, which could lead to further equity raises. Analysts warn that any delays in data center build-outs or supply-chain disruptions for advanced AI chips could push back revenue recognition. Additionally, if the company issues new shares to fund growth, existing shareholders could see meaningful dilution, a factor that investors must weigh against the strong topline trajectory.