Neocloud — History & Future
Where the neocloud category came from, the inflection points that built it, the current shape of the industry, and where it's heading. The context every other guide in this topic builds on.
Chapters
00Start HereWhat 'neocloud' means; scope of this guide.
01Defining 'Neocloud'The term, its origin, who counts; the analyst frameworks (ClusterMAX, SemiAnalysis tiers).
02Pre-AI GPU Cloud (2010s)Lambda's hardware shop, early Coreweave, gaming-rental, crypto miners; the pre-history.
03The 2022–2023 InflectionChatGPT, the H100 shortage, capital flooding in, Crusoe's pivot, CoreWeave's rebrand.
04The 2024 Shake-OutTier emergence, ClusterMAX, the divergence between Platinum and the rest.
05The 2025–2026 Enterprise EraHyperscaler customer commitments; Microsoft, Meta, OpenAI as neocloud customers.
06Business Model TaxonomyMarketplace / on-demand / reserved-capacity / managed-inference / energy-arbitrage.
07Hyperscaler vs Neocloud DynamicsCo-opetition; why hyperscalers buy capacity from neoclouds even while building their own.
08NVIDIA's RoleStrategic investor, supplier, customer, kingmaker; the GPU allocation question.
09Future ScenariosConsolidation, futures-market emergence, GPU-price normalization, geopolitics.
10Investment LandscapePublic neoclouds (CoreWeave, Nebius), private funding climate, M&A signals.