CoreWeave Inc. announced a multi-year agreement with Anthropic to support the massive compute requirements necessary for developing and deploying Anthropic’s Claude family of models, its second major artificial intelligence (AI) deal in several days.
Within a 48-hour window, the company also secured a $21 billion expanded agreement with Meta Platforms Inc. With these additions, CoreWeave now provides the infrastructure for nine of the world’s top 10 AI model developers, including industry titans like OpenAI and Google.
Under terms of the agreement, Anthropic will leverage CoreWeave’s specialized GPU-optimized cloud to run workloads at production scale. This provides Anthropic with immediate access to vast arrays of NVIDIA Corp. graphics processing units (GPUs) housed in facilities across the United States. While the financial specifics were undisclosed, the move is a clear response to insatiable demand for the Claude AI suite.
“AI is no longer just about infrastructure; it’s about the platforms that turn models into real-world impact,” Michael Intrator, co-founder and CEO of CoreWeave, said in a statement. “We’re excited to work with Anthropic at the center of where models are put to work. It’s exactly the kind of real-world deployment CoreWeave was built for.”
More important, the deal highlights a shift in the neocloud sector. Traditionally seen as a temporary supplement for companies unable to build their own data centers fast enough, providers like CoreWeave have graduated into a permanent, essential layer of the global AI economy.
Even tech giants spending upwards of $100 billion annually on their own hardware are now turning to CoreWeave’s end-to-end technology stack, which recently earned top Platinum rankings from industry analysts at SemiAnalysis.
For sectors like financial technology, the implications of this infrastructure surge are significant.
As Anthropic scales, the enhanced reasoning capabilities of the Claude models are expected to accelerate innovations in real-time fraud detection, dynamic risk assessment, and automated regulatory compliance. Claude’s focus on Constitutional AI and safety makes it a particularly attractive partner for banks and investment platforms that require high precision and low hallucination risks.
However, the sheer scale of these operations brings new challenges. Anthropic recently revealed a separate 3.5-gigawatt commitment for TPU capacity with Broadcom Inc. and Google. As these energy demands soar, the industry is beginning to weigh the benefits of hyperscale performance against the growing pressures of sustainability and ESG reporting.
With compute scheduled to come online later this year, the CoreWeave-Anthropic alliance signals that the race for AI supremacy is no longer just about who has the best algorithm, but about who has the power to run it.
