AI cloud computing provider CoreWeave has secured a five-year contract worth $11.9 billion with OpenAI. As part of the deal, OpenAI will also acquire approximately $350 million worth of CoreWeave shares. CoreWeave's CEO, Michael Intrator, highlighted the company's proven capabilities in delivering high-performance infrastructure services for leading AI labs, emphasizing their role as a preferred partner in advancing AI innovation.
OpenAI's CEO, Sam Altman, expressed satisfaction with the collaboration, acknowledging the importance of reliable computing power for advanced AI systems. He noted that CoreWeave is a vital addition to OpenAI's infrastructure portfolio, complementing their existing agreements with Microsoft (MSFT) and Oracle, as well as their joint venture with SoftBank.
CoreWeave has filed for an initial public offering (IPO) on NASDAQ under the ticker "CRWV," aiming to raise approximately $4 billion with a valuation target exceeding $35 billion. This move positions CoreWeave as one of the largest IPOs in the AI computing sector in recent years.
CoreWeave has experienced rapid growth since 2020, driven by the surge in AI computing demand. The company provides cloud-based computing services using NVIDIA GPUs, catering to machine learning and deep learning applications. In 2024, CoreWeave's revenue reached $1.9 billion, marking a 700% increase, although it reported a net loss of $863 million due to significant investments in AI infrastructure.
Investors are discussing the sustainability of CoreWeave's growth, particularly if major clients like Microsoft and Meta (META) reduce reliance on external cloud providers. CoreWeave acknowledges competition with cloud giants like Microsoft's Azure and Amazon's AWS. Recently, there were reports of Microsoft dropping some contracts with CoreWeave due to delivery issues, which CoreWeave denied.
CoreWeave is known for its close relationship with NVIDIA (NVDA), having been an early adopter of NVIDIA GPUs. This partnership has enabled CoreWeave to rapidly expand its AI computing capabilities, securing priority access to NVIDIA's advanced GPUs.
