Sources familiar with the matter indicate that CoreWeave, known for offering cloud computing services built on Nvidia's graphics processing units (GPUs), is set to supply computing capacity to Google's cloud unit. Subsequently, Alphabet's Google will then resell this capacity to OpenAI, a major player in the AI landscape, to meet the burgeoning requirements for its services, including the widely recognized ChatGPT. Google is also expected to contribute some of its own computing resources to OpenAI as part of this arrangement.
This intricate partnership, initially reported by Reuters, sheds light on the evolving dynamics between established hyperscalers like Amazon.com, Microsoft, and Google, and emerging "neocloud" companies such as CoreWeave. Hyperscalers are large-scale cloud service providers with extensive data center infrastructure. The insatiable appetite for computing power has not only spurred significant investment but also transformed competitive rivals into strategic partners.
CoreWeave, which has already garnered backing from OpenAI and Nvidia, brought Google on board as a customer in the first quarter of this year. While CoreWeave, Google, and OpenAI have not publicly commented on the specific details of this new arrangement, CoreWeave's existing relationship with OpenAI is well-documented.
CoreWeave, which went public in March, has been a significant infrastructure provider for OpenAI. It holds a five-year contract valued at $11.9 billion with OpenAI to furnish dedicated computing capacity for OpenAI's model training and inference. OpenAI further solidified its commitment to CoreWeave by taking a $350 million equity stake in the company in March. This partnership was subsequently expanded last month through an additional agreement worth up to $4 billion, extending through April 2029, a clear indicator of OpenAI's rapidly increasing need for high-performance computing resources.
Industry insiders suggest that the addition of Google Cloud as a new customer could be a strategic move for CoreWeave, enabling it to diversify its revenue streams. Furthermore, having a reputable and financially robust partner like Google could facilitate more favorable financing terms for CoreWeave, supporting its ambitious plans for data center expansion across the country.
For Google's cloud unit, which generated $43 billion in sales last year, this partnership offers an opportunity to capitalize on the growth of OpenAI, despite OpenAI also being a significant competitor in areas like search and chatbots. It positions Google as a neutral provider of computing resources, competing with peers such as Amazon and Microsoft in this crucial segment of the AI ecosystem.
The timing of CoreWeave's deal with Google coincides with Microsoft's re-evaluation of its data center strategy, including pulling back from certain data center leases. Microsoft, which was once CoreWeave's largest customer, accounting for approximately 62% of its 2024 revenue, is also reportedly renegotiating with OpenAI to revise the terms of their multibillion-dollar investment, including the future equity stake Microsoft will hold in OpenAI.
Backed by Nvidia, CoreWeave has rapidly established itself as a prominent provider of GPU-based cloud infrastructure, riding the wave of the AI boom. While its initial public debut in March was met with some caution due to concerns about its highly leveraged capital structure and fluctuating GPU demand, the company's stock has since experienced a remarkable surge. Following its IPO price of $40 per share, CoreWeave's stock has gained over 270%, reaching a record high of $166.63 in June, signaling a strong market response to its strategic positioning in the AI computing landscape.