CoreWeave announced on 10 April a multi-year agreement with Anthropic to provide Nvidia GPU capacity for production-scale Claude workloads across US data centres. The deal, which begins with a phased infrastructure rollout later in 2026 with potential to expand over time, establishes CoreWeave as Anthropic's third major compute provider alongside Amazon Web Services (for Trainium training chips) and Google Cloud (for TPUs coming online from 2027).
The financial terms were not disclosed, but the deal's significance is clear from the context. Anthropic's annualised revenue run rate surpassed $30 billion in early April 2026 — more than tripling from roughly $9 billion at the end of 2025 — and the company needs diversified compute infrastructure to keep pace with enterprise demand for Claude. CoreWeave CEO Michael Intrator framed the partnership in practical terms: 'AI is no longer just about infrastructure, it is about the platforms that turn models into real-world impact.'
The announcement came during an extraordinary 48-hour stretch for CoreWeave. Just one day earlier, Meta committed an additional $21 billion to CoreWeave for dedicated AI cloud capacity running from 2027 through December 2032, bringing Meta's total relationship to approximately $35 billion. Combined with OpenAI's roughly $6.5 billion expansion earlier in 2026, CoreWeave now serves nine of the ten leading AI model providers and holds a contracted backlog of $66 billion — significantly reducing the company's earlier dependence on Microsoft, which previously accounted for 67% of revenue.
For context engineers, this deal has a direct practical implication: more GPU capacity dedicated to Claude inference means better availability, lower latency, and more headroom for the compute-intensive agentic workflows that developers rely on daily — extended Claude Code sessions, multi-agent orchestrations, and MCP server chains. Anthropic is clearly building redundancy across three distinct compute providers (AWS, Google Cloud, CoreWeave) to ensure that no single infrastructure bottleneck can constrain Claude's production capacity as enterprise adoption accelerates.
