Nvidia GPUs drive record stock growth as demand surges

Meta: Meta develops AI technologies and is expanding its capabilities by forming a cloud unit to offer AI computing power. It manages extensive GPU resources for its own operations and potential external rentals. The news mentions Meta's move as part of broader industry efforts to monetize excess capacity. Amazon: Amazon operates one of the largest hyperscale cloud platforms, offering infrastructure services including GPU access for AI applications. It serves major AI labs and enterprises seeking reliable, established providers with full-stack capabilities. The news positions Amazon as a top option amid competition from neoclouds and direct hardware purchases. Google: Google provides cloud infrastructure services that include access to Nvidia GPUs for AI training and inference. It competes in the hyperscaler space while also utilizing third-party capacity when needed. The news describes Google as a key destination for companies running generative AI workloads. Nvidia: Nvidia designs and supplies graphics processing units and AI accelerators that power demanding computing tasks. Its chips have become the primary hardware choice for generative AI workloads across industries. The news highlights how surging demand is driving Nvidia's growth through diverse access channels for customers. Oracle: Oracle delivers cloud infrastructure and allows clients to bring their own GPUs for operation within its data centers. It targets organizations seeking flexibility without building their own facilities. The news covers Oracle's approach as an alternative for companies with capital but limited power or space resources. SpaceX: SpaceX operates advanced computing infrastructure and has begun renting excess GPU capacity through targeted deals. It leverages its investments to generate returns while supporting AI companies. The news discusses SpaceX's arrangements with hyperscalers and AI labs as an emerging rental option. Dropbox: Dropbox develops collaboration software and maintains its own GPU-equipped data centers to support AI-enhanced features. It draws on established supply chain relationships for hardware access. The news references Dropbox as an enterprise example of balancing cloud and on-premises AI infrastructure. Reactor: Reactor is a video generation startup that relies on GPUs from multiple providers to deliver low-latency results for user-created content. It evaluates options based on performance, service, and pricing. The news uses Reactor as an example of companies balancing hyperscalers and neoclouds for specific needs. Everpure: Everpure provides data center storage hardware and software and has acquired its own GPUs to run open-weight AI models internally. It values multiple supply sources amid fluctuating market conditions. The news cites Everpure's strategy as part of the shift toward on-premises or direct hardware ownership. CoreWeave: CoreWeave is a neocloud provider specializing in GPU infrastructure for AI and high-performance computing. It serves customers needing large-scale capacity beyond what some hyperscalers can immediately supply. The news identifies CoreWeave as an example of emerging specialized providers gaining traction. Microsoft: Microsoft runs Azure, a major hyperscale cloud platform that provides GPU resources for AI and enterprise workloads. It partners with leading AI companies and emphasizes trust and security for business customers. The news notes Microsoft's role in handling significant AI computing demand alongside other hyperscalers. GPU Demand: Demand for Nvidia GPUs in AI continues to outpace supply, leading companies to combine multiple access methods including hyperscalers and specialized providers. Infrastructure Options: Businesses are turning to neoclouds, bring-your-own-hardware models, and on-premises deployments to address immediate capacity needs and gain flexibility. Market Diversification: Nvidia's customer relationships are expanding beyond a small group of large cloud providers as more organizations seek computing capacity.