Hudson River Trading signs multibillion dollar CoreWeave AI computing deal

Hudson River Trading signed a multiyear agreement worth billions of dollars with CoreWeave to expand the computing infrastructure used for its trading research and models.

The agreement builds on an existing partnership between the companies and represents a material expansion of their relationship, CoreWeave Chief Revenue Officer Jon Jones told Bloomberg. He declined to disclose specific financial terms.

As part of the deal, HRT will become one of the first CoreWeave customers to receive wide scale access to Nvidia’s newest Vera Rubin AI chips.

HRT, one of the largest nonbank trading firms in the US, has been increasing its use of specialized cloud providers such as CoreWeave as its research requires greater computing capacity. Additional processing power allows researchers to test more models and evaluate trading strategies across a wider range of markets and time horizons.

Kevin Lee, head of research and development at HRT, said greater access to computing resources lets the firm’s researchers run more experiments and potentially identify new trading opportunities. He also said CoreWeave had performed well in delivering infrastructure on schedule and operating Nvidia hardware efficiently.

The deal further expands CoreWeave’s customer base as the company works to reduce its historical dependence on Microsoft and increase its exposure to financial services customers. HRT became a CoreWeave customer in March.

CoreWeave highlighted its growing relationship with HRT in its second quarter earnings report last week as it pushes further into financial services alongside its existing technology and AI customers.

Jane Street, one of HRT’s largest competitors, has also expanded its relationship with CoreWeave. The trading firm previously agreed to spend $6 billion on CoreWeave data center capacity and invested $1 billion in the cloud infrastructure provider.

CoreWeave has also raised billions of dollars in debt to finance additional computing capacity tied to demand from customers including HRT and Jane Street.

Jones said financial firms are increasingly adopting the same advanced computing infrastructure used by developers of frontier AI models as trading companies expand their use of artificial intelligence and large scale computing.