Anthropic to Deploy 2 Gigawatts of AMD AI GPUs

AMD and Anthropic have announced a strategic partnership that will significantly expand AI computing infrastructure while deepening collaboration on AI software development.

The agreement calls for Anthropic to deploy up to 2 gigawatts of AMD Instinct MI450 Series GPUs in AMD’s Helios rack-scale AI platform, with the first gigawatt scheduled to begin deployment during the first half of 2027.

AMD also announced plans to make a strategic equity investment of up to $5 billion in Anthropic as the companies strengthen their long-term relationship.

Anthropic Expands AI Infrastructure

Anthropic will use AMD Helios systems built with Instinct MI455X GPUs, AMD EPYC “Venice” processors, Pensando networking technology and ROCm software to support the growing demand for training and running its Claude family of AI models.

The deployment represents one of AMD’s largest AI infrastructure wins to date and builds on Anthropic’s existing use of AMD Instinct accelerators.

The companies said Helios was designed specifically for large-scale AI training and inference workloads, combining compute performance, networking and software into a single integrated platform.

Multi-Year Engineering Collaboration

Beyond infrastructure, AMD and Anthropic will work together to improve AI software development through a multi-year engineering partnership.

The companies plan to use Claude to optimize workloads running on AMD Instinct GPUs while accelerating development of AMD’s ROCm software platform.

AMD also intends to expand the use of Claude across its engineering and product development teams to help improve software development efficiency.

AI Infrastructure Demand Continues to Grow

The partnership comes as AI companies race to secure enough computing capacity to support increasingly sophisticated foundation models and growing enterprise demand.

Rather than relying on a single hardware platform, Anthropic said it intends to diversify its infrastructure, allowing workloads to run across different architectures depending on performance requirements.

The strategy reflects a broader industry trend toward building more flexible AI infrastructure capable of supporting rapidly expanding training and inference workloads.

AMD Strengthens Its AI Position

The agreement further strengthens AMD’s position in the rapidly growing AI infrastructure market as the company continues expanding its portfolio of AI accelerators, processors, networking technology and software.

The planned equity investment also signals AMD’s long-term commitment to Anthropic as competition intensifies among companies developing next-generation AI models and cloud infrastructure.

The partnership combines hardware deployment, software optimization and financial investment into a broad strategic relationship focused on future AI growth.

The Bottom Line

Demand for AI computing continues to accelerate as organizations deploy larger models and expand enterprise AI applications. AMD’s partnership with Anthropic positions both companies to scale the infrastructure behind Claude while advancing hardware, software and AI development through one of the industry’s largest strategic collaborations.