CoreWeave Inc., a startup providing cloud computing infrastructure and artificial intelligence, said today that it has secured $7.5 billion in debt financing.
This was one of the biggest rounds of private debt fundraising to date, led by Blackstone Inc. with participation from Carlyle Group and BlackRock Inc. It comes soon after the $1.1 billion Series C equity funding round the company completed earlier this month, which valued the business at $19 billion.
CoreWeave, a provider of GPUs for AI model training and deployment, has established itself as a key player in the AI revolution. Given the explosive growth in demand for computing capacity, it is the main supplier of Nvidia Corp.’s powerful AI processing chips. In order to prepare AI models for production and operate them there, such as OpenAI’s ChatGPT, which debuted in 2022 and sparked internet controversy because to its capacity for human-like dialogue, GPUs are necessary.
Speaking about the money, CoreWeave Chief Executive Michael Intrator told the Wall Street Journal, “It is a monumental financing in nominal terms, but it’s also a monumental financing in terms of the extent to which it drives the company forward.”
By the end of 2022, the corporation will have 14 data centers under operation; by the end of this year, that number will have doubled. The company offers a large number of Nvidia H100 GPUs in its centers, which are specifically designed to handle AI workloads including costly training and real-world applications. In contrast to AI-optimized cards, the company also provides access to A40 GPUs, which have RT-cores for graphics processing power and ray tracing, which are essential for rendering 3D special effects.
According to Intrator, a large portion of the additional funds would be used to expand the company’s presence in new data centers and buy the pricey GPUs needed to staff them in order to meet the growing demand for both artificial intelligence and graphics processing.
More progress has been made in artificial intelligence recently, pushing models beyond their 2022 starting point. Before OpenAI’s recent announcement, ChatGPT was limited to text-to-text interactions. However, with the addition of sophisticated multimodal features, it can now hold voice conversations and comprehend video. Google LLC revealed a comparable set of characteristics for its Gemini AI model with Project Astra the next day.
The complexity of AI models will continue to rise as a result of these developments, increasing the demands placed on the infrastructure needed to support them. Consequently, investors are swarming to the cloud compute startups that will drive the upcoming surge in artificial intelligence capabilities.
“I don’t think anyone would have expected another hyperscaler was going to come into the market this quickly,” Intrator added.