BofA warns Fed risks policy mistake with early rate cuts
LIVINGSTON, N.J. - CoreWeave, Inc. (NASDAQ:CRWV), a $72.7 billion market cap AI infrastructure company that has seen its stock surge 278% in the past six months, announced Wednesday it has become the first cloud platform to make NVIDIA RTX PRO 6000 Blackwell Server Edition instances generally available. According to InvestingPro analysis, the company currently trades above its Fair Value, reflecting investor enthusiasm for its AI capabilities.
The new GPU instances achieve up to 5.6x faster LLM inference and 3.5x faster text-to-video generation than previous generations, according to the company. The hardware is designed for inference of models up to 70B parameters and provides a more cost-efficient alternative to larger GPU clusters. With impressive gross margins of 74.3% and trailing twelve-month revenue of $2.7 billion, CoreWeave has demonstrated strong operational efficiency in the competitive AI infrastructure market.
"CoreWeave is built to move at the speed of innovation, and with the new RTX PRO 6000-based instances, we’re once again first to bring advanced AI and graphics technology to the cloud," said Peter Salanki, Co-Founder and Chief Technology Officer of CoreWeave.
The RTX PRO 6000-based instances feature 8x RTX PRO 6000 GPUs, 128 Intel Emerald Rapids vCPUs, 1TB System RAM, 100 GBps Networking Throughput, and 7.68TB local NVMe storage.
With this addition, CoreWeave now offers multiple NVIDIA Blackwell infrastructure options, including the NVIDIA GB200 NVL72 system and NVIDIA HGX B200 platform.
"The NVIDIA RTX PRO 6000 GPU represents a breakthrough in AI and graphics performance, empowering a variety of industries with advanced, cost-effective solutions," said Dave Salvator, director of accelerated computing products at NVIDIA.
CoreWeave has previously been among the first to offer NVIDIA H200 GPUs and was the first AI cloud provider to make NVIDIA GB200 NVL72 systems generally available, according to the press release statement.
In June, CoreWeave collaborated with NVIDIA and IBM to submit an MLPerf Training v5.0 benchmark using nearly 2,500 NVIDIA GB200 Grace Blackwell Superchips, completing the Llama 3.1 405B model training in 27.3 minutes. For investors interested in deep-dive analysis of high-growth AI companies like CoreWeave, InvestingPro offers comprehensive research reports with detailed financial metrics, valuation analysis, and expert insights on over 1,400 US stocks.
In other recent news, CoreWeave announced its intent to acquire Core Scientific in an all-stock transaction valued at approximately $9 billion. This acquisition is expected to deliver significant cost savings, with projections of eliminating $10 billion in lease overhead over the next 12 years and achieving $500 million in annual run-rate cost savings by 2027. Analysts have reacted to the news with mixed ratings; Mizuho downgraded CoreWeave from Outperform to Neutral while raising the price target to $150.00, and Stifel downgraded it from Buy to Hold, setting a new price target of $115.00. Macquarie maintained a Neutral rating with a $65.00 price target, noting the deal could be 5% accretive to earnings per share based on projected net annual savings.
In addition to the acquisition news, CoreWeave became the first AI cloud provider to deploy NVIDIA GB300 NVL72 systems, aiming to enhance AI development capabilities. The deployment, achieved in collaboration with Dell, Switch, and Vertiv, offers significant improvements in user responsiveness and throughput. Barclays maintained an Equalweight rating with a $100 price target, citing the acquisition as a strategic move to vertically integrate data center operations. The Core Scientific deal is expected to close in the fourth quarter of 2025, pending regulatory approval. These developments reflect CoreWeave’s ongoing efforts to expand its data center footprint and enhance its AI infrastructure capabilities.
This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.