CoreWeave Inc., a cloud compute provider that specializes in large-scale graphics processing unit-accelerated workloads, said today that it secured $221 million in funding that it will use to expand its operations.

The Series B round was led by Magnetar Capital and included participation from the chipmaker Nvidia Corp., former GitHub Chief Executive Nat Friedman and ex-Apple executive Daniel Gross. The raise brings the company’s valuation to $2 billion, according to TechCrunch, and the total raised so far to $371 million.

The company’s goal is timely given the explosion of interest in generative artificial intelligence that started with OpenAI LP’s late 2022 release of the chatbot ChatGPT. “CoreWeave is uniquely positioned to power the seemingly overnight boom in AI technology with our ability to innovate and iterate more quickly than the hyperscalers,” said co-founder and Chief Executive Michael Intrator.

The company said it will use the funding to build out its specialized cloud infrastructure and support the expansion of its data centers in the United States by opening two more this year. This will bring the total number of North America-based data centers up to five.

Founded in 2017, CoreWeave initially homed in on GPUs that were purchased from insolvent cryptocurrency mining firms, which are used largely in the mining of Ethereum. The company chose Nvidia’s hardware primarily thanks to its larger memory and initially used its trove of GPUs for crypto applications but changed strategy in the past few years to general-purpose computing and generative AI technologies, which also require heavy use of GPUs.

This focus on GPU-accelerated technology, Intrator says, allows CoreWeave to outperform other cloud providers when it comes to more specialized use cases, especially AI-specific needs. Generative AI technology, such as ChatGPT chatbot and Stable Diffusion’s art-generating AI, require massive amounts of nearly identical tasks to be run at scale. GPUs are good at doing this in parallel, which provides greatly increased speed and power.

CoreWeave delivers access to Nvidia’s most recent innovations in high-performance data center GPU technology including the Nvidia H100 Tensor Core, which is part of the Nvidia HGX H100 platform, via cloud clusters. In addition to access to the H100, CoreWeave customers can also provision access to other Nvidia technology including A100s, A40s and RTX A6000s as well.

The cloud infrastructure can be tapped for numerous use cases including visual effects rendering, machine learning and AI, large-scale batch processing and pixel streaming, which the company claims can be processed up to 35 times faster at 80% less cost than on a generalized public cloud.

A current customer of CoreWeave includes Anlatan, the developer of NovelAI, a monthly subscription service that provides AI-assisted authorship and storytelling that can generate text based on users’ respective writing styles. Anlatan has tapped into the HGX H100 platform served by CoreWeave in order to power its service.

“We are entirely focused on AI innovation and AI-first products,” said Anlatan CEO Eren Doğan. “Nvidia H100 GPUs are the most top-notch, state-of-the-art machine learning accelerators. This gives us a significant competitive advantage within the machine learning industry.”

In addition to the funding round, the company is also deepening its collaboration with Nvidia. “Nvidia consistently pushes the boundaries of what’s possible in the field of technology,” said Intrator. “Their vision and guidance will be invaluable as we continue to scale our organization.”

Image: CoreWeave