William Blair Bullish on Both CoreWeave and Nebius, Projects New Cloud Market to Reach $300 Billion by 2031
nashnova research
William Blair initiated coverage on CoreWeave and Nebius simultaneously, rating both outperform — projecting the neocloud market will reach $200–300 billion by 2031. This marks the first time a single research note has endorsed the entire neocloud thesis.
What is "neocloud," and why does it matter now?
Neocloud refers to cloud platforms built specifically for AI training and inference — unlike traditional clouds (AWS, Azure), the entire stack from hardware to software is designed around GPUs, not general-purpose servers.
William Blair projects neocloud revenue will grow from roughly $25 billion in 2026 to $200–300 billion by 2031, with about 20 gigawatts of data-center capacity coming online over that period.
This means → analysts see neocloud not as a supplement to traditional cloud, but as a standalone market with an exceptionally high compound growth rate.
CoreWeave vs. Nebius — who has the edge?
CoreWeave is the first mover: largest revenue base, deepest order backlog, tightest relationship with Nvidia — but higher customer concentration and heavier debt.
Nebius has nearly tripled in market cap this year, now exceeding CoreWeave. It carries a cleaner balance sheet, more differentiated cloud architecture, and what analysts call greater near-term upside from software and higher-layer AI services.
In plain terms = CoreWeave wins on scale and early advantage; Nebius wins on financial health and value-added services — analysts believe both paths can work.
What are the bears worried about?
The bear case, led by veteran short-seller Jim Chanos, centers on three risks: debt-fueled capex, a possible cyclical pullback in AI demand, and traditional cloud giants muscling in.
William Blair's counter: neocloud firms can leverage strong contracts with hyperscalers to convert sub-investment-grade credit into investment-grade financing, lowering borrowing costs.
This means → the bull-bear debate isn't about whether AI demand exists — it's about whether neocloud firms can build deep enough moats before the giants arrive.
How do the leaders ultimately win?
Analysts expect today's front-runners to be the final winners, likely acquiring smaller neocloud players to extend their scale advantage.
They also see margin improvement through expanding service categories — moving from pure GPU rental toward higher-layer software offerings to boost retention and profitability.
This reflects a deeper signal: the neocloud endgame may not be about who has the most GPUs, but who turns a hardware business into a platform business first.
How is the market voting right now?
ARK Invest bought 254,550 shares of CoreWeave on Tuesday — Cathie Wood doubling down on the first mover.
Post-announcement, CoreWeave dipped 0.1% in premarket while Nebius rose roughly 2% — the market's short-term preference tilts toward the cleaner balance sheet.
Whether the valuation divergence between the two firms is validated by the 2031 market-size forecast is the key watch point for the next phase of this sector.
市场有风险，内容仅供研究参考，不构成投资建议。