AI News
AI News AgentBusinessAnthropic3 min read

Microsoft and NVIDIA expand Anthropic’s AI on Azure

Anthropic will expand Claude on Microsoft Azure with NVIDIA chips and purchase up to $30 billion in computing capacity. Microsoft and NVIDIA will invest up to $15 billion in the company, while Claude comes to Microsoft Foundry and more Copilot products.

Anthropic will bring its Claude AI model to Microsoft Azure using NVIDIA infrastructure, while Microsoft and NVIDIA commit to investing up to $15 billion in the company. The agreement gives companies more options for using AI without depending on a single provider.

Anthropic has agreed to purchase $30 billion in Azure computing capacity and contract additional capacity of up to one gigawatt. That capacity will be needed to train and run models serving many users at the same time.

What the agreement includes

The partnership brings together three major moves:

  • Claude will expand on Azure using NVIDIA systems.
  • Anthropic and NVIDIA will collaborate on model and chip design to improve performance, efficiency and cost.
  • NVIDIA will invest up to $10 billion in Anthropic, while Microsoft will contribute up to $5 billion.

Anthropic will initially use systems based on the NVIDIA Grace Blackwell and NVIDIA Vera Rubin architectures. The company will also work with NVIDIA to adapt its models to future hardware generations and make those chips better suited to Claude’s needs.

The key word is capacity. Building and offering advanced AI models requires enormous amounts of specialized processors. Through this agreement, Anthropic secures access to that infrastructure, while NVIDIA gains a direct collaboration to steer its next products toward one of the largest AI users.

Claude comes to more Microsoft products

Microsoft Foundry customers will be able to access several Anthropic models, including Claude Sonnet 4.5, Claude Opus 4.1 and Claude Haiku 4.5. Foundry is Microsoft’s platform for letting companies choose and deploy AI models for their own applications.

In practice, a company will be able to use Claude to analyze documents, generate code or create an internal assistant within its Azure environment. It will also be able to compare its results with other models available on the platform instead of being limited to a single AI family.

Microsoft will also maintain access to Claude across its Copilot product family, including GitHub Copilot, Microsoft 365 Copilot and Copilot Studio. That means the partnership is not limited to servers. It also extends to tools used by millions of people to program, write documents and automate tasks.

What changes for you

If you use Microsoft products, you may find more Claude-based features or more options for choosing which model processes a task. For businesses, the main change will be the ability to purchase Claude through Azure, using their usual security, billing and administration systems.

The agreement also strengthens competition among the major AI providers. According to Anthropic, Claude will be the only advanced model available across the world’s three leading cloud services. Amazon will still be Anthropic’s primary cloud provider and training partner.

The news does not mean Claude will automatically replace other models or that every feature will be available to every user from day one. What changes is who can offer it, with what infrastructure and at what scale. The next point to watch will be how this investment translates into better prices, more capacity and specific features within Azure and Copilot.

Microsoft and NVIDIA expand Anthropic’s AI on Azure | neversleep.ai