Claude in Microsoft Foundry is now generally available
Microsoft has announced the general availability of Anthropic's Claude models within Microsoft Foundry on Azure, marking a significant step forward for enterprises looking to operationalize AI. This integration allows businesses to access powerful generative AI capabilities directly within their existing Azure environment, leveraging familiar authentication, billing, and governance mechanisms. The move is designed to simplify the journey from initial AI experimentation to full-scale production deployment.
The availability includes key Claude models such as Claude Opus 4.8 and Claude Haiku 4.5, accessible via the Messages API. These models offer advanced capabilities for complex reasoning, coding, agentic workflows, and tool streaming. Microsoft Foundry acts as the comprehensive platform for building, evaluating, deploying, and scaling these AI agents, providing the necessary enterprise harness to manage AI solutions effectively.
A crucial aspect of this offering is the underlying infrastructure, which utilizes NVIDIA GB300 Blackwell Ultra GPUs. This accelerated computing power, combined with NVIDIA GB300 NVL72 systems and Quantum-X800 InfiniBand networking, ensures that enterprises can run advanced, specialized AI agents with the performance, scale, and security required for demanding production environments. This collaboration builds on a strategic partnership announced in November 2025 between Microsoft, NVIDIA, and Anthropic, aimed at expanding enterprise access to Claude on NVIDIA accelerated computing.
The initiative directly addresses common hurdles in enterprise AI adoption, which often stem not from model quality, but from complexities around procurement, governance, networking, and data management. By hosting Claude models on Azure, Microsoft provides a trusted path for organizations to develop AI applications, orchestrate them with Foundry Agent Service, and ground them in enterprise knowledge using Microsoft IQ, all while adhering to established Azure controls and commitments. This enables teams to concentrate on developing innovative agentic applications rather than managing underlying infrastructure.
Read original source