Anthropic has signed a six-year cloud computing agreement worth $35 billion with Lambda, an AI cloud provider backed by Nvidia. The deal gives Anthropic additional infrastructure as demand for its Claude products continues to rise.
The agreement comes as leading AI companies compete for access to advanced computing capacity. Meanwhile, Anthropic has committed billions of dollars to secure infrastructure across several providers. The company recently agreed to a separate six-year, $45 billion deal with Nscale for capacity in West Virginia.
Texas Data Center Supports Expansion
Under the Lambda agreement, Anthropic will use computing capacity at a data center under development in Nueces County, Texas. Hut 8, a former cryptocurrency mining company, is developing the facility. The site is expected to provide about 350 megawatts of capacity.
Nvidia also plays a significant role in the arrangement. The chipmaker backs Lambda and supplies its AI processors for the infrastructure. Furthermore, Nvidia will hold the lease on the Texas data center, according to reports cited by Reuters and the Wall Street Journal.
However, Anthropic, Nvidia, Hut 8 and Lambda had not publicly commented on the agreement at the time of the reports. Therefore, some details of the structure remain based on sources familiar with the transaction.
Anthropic Accelerates AI Infrastructure Spending
The Lambda agreement adds to Anthropic’s rapidly expanding computing commitments. In addition to Nscale, the company has secured substantial infrastructure from other technology providers. Consequently, its spending reflects the growing cost of training and operating frontier AI systems.
Anthropic has also expanded its hardware strategy beyond a single chip supplier. The company runs Claude across AWS Trainium, Google TPUs and Nvidia GPUs. This approach allows Anthropic to match workloads with different architectures while improving infrastructure resilience.
Moreover, Nvidia has deepened its relationship with Anthropic. Nvidia previously announced that Anthropic would run and scale its AI workloads on Nvidia infrastructure, initially adopting 1 gigawatt of compute capacity.
As AI demand accelerates, therefore, access to power, chips and data centers has become a central competitive factor. Anthropic’s latest agreement highlights how AI developers are securing infrastructure years ahead to support continued model and product growth.








