CoreWeave Unveils NVIDIA GB200 Grace Blackwell Systems for Enhanced AI Scalability
CoreWeave, known as the AI Hyperscaler™, has made a significant leap in AI technology with the introduction of the NVIDIA GB200 NVL72 rack-scale systems. These innovative systems are now available to key clients including Cohere, IBM, and Mistral AI, enhancing their capabilities in AI model development and deployment. This launch underscores CoreWeave’s commitment to leading the charge in AI cloud solutions that meet the growing demands of enterprises seeking to harness the power of artificial intelligence.
The NVIDIA GB200 systems are not just hardware; they represent a state-of-the-art resource for AI innovators, equipped with advanced networking capabilities and the NVIDIA Grace Blackwell Superchips specifically designed for reasoning and agentic AI applications. CoreWeave has consistently positioned itself as a pioneer in this field, enabling customers to achieve significant breakthroughs in AI technology. According to Michael Intrator, Co-Founder and CEO of CoreWeave, the launch serves as a testament to their engineering excellence and commitment to rapidly delivering cutting-edge systems. "CoreWeave is built to move faster – and time and again, we've proven it by being first to operationalize the most advanced systems at scale," says Intrator. This ethos of speed and innovation is crucial, as organizations across various sectors are racing to leverage reasoning models to build transformative AI applications that revolutionize both work and leisure activities.
Ian Buck, NVIDIA's Vice President of Hyperscale and HPC, emphasizes the importance of CoreWeave's swift deployment of these systems. He remarked that such advances are essential for turning AI infrastructure and software into the tangible tools that make "AI factories" a real possibility. With this infrastructure in place, organizations can unleash the potential of AI, turning complex models into functional applications with unprecedented efficiency.
CoreWeave continues to break performance records in the field of AI, having recently achieved a noteworthy new industry benchmark in AI inference using the NVIDIA GB200 Grace Blackwell Superchips. Reported in the latest MLPerf v5.0 results, this achievement indicates that CoreWeave’s systems are making significant strides in establishing ML performance standards. This is a reflection of their ongoing commitment to maximizing efficiency and setting new records, showcasing their role at the forefront of AI advancements.
In addition to the groundbreaking GB200 rollout, CoreWeave has previously been among the first to adopt and integrate the NVIDIA H100 and H200 GPUs within their services. Their offerings include a full stack of optimized cloud solutions tailored for the new NVIDIA GB200 NVL72, ensuring performance reliability through services like CoreWeave Kubernetes Service, Slurm on Kubernetes (SUNK), and CoreWeave Mission Control. Furthermore, the NVIDIA Blackwell-accelerated instances have the capability to scale up to an impressive 110,000 Blackwell GPUs, augmented by NVIDIA Quantum-2 InfiniBand networking.
Notably, apart from Cohere, IBM, and Mistral AI, CoreWeave has recently sealed a multi-year contract with OpenAI, the creators behind the popular AI platform, ChatGPT. This partnership further solidifies CoreWeave’s standing as a primary player in the rapidly evolving AI landscape, showcasing their cloud technology as critical for enabling enhanced performance and innovation.
With the cloud-based computing environment continuously evolving, CoreWeave's role is likely to grow even more significant in the upcoming years. Their dedication to providing advanced computing resources at scale is not only transforming how enterprises operate but is also setting the stage for future advancements in AI technology. As they continue to pioneer and develop new solutions, industry leaders have their eyes firmly set on CoreWeave for inspiration and progress in the field of artificial intelligence.
Through their innovative approaches and the launch of the NVIDIA GB200 systems, CoreWeave is empowering businesses to innovate faster and explore new horizons in AI, ultimately leading to a future where AI-driven applications readily enhance everyday life.
The NVIDIA GB200 systems are not just hardware; they represent a state-of-the-art resource for AI innovators, equipped with advanced networking capabilities and the NVIDIA Grace Blackwell Superchips specifically designed for reasoning and agentic AI applications. CoreWeave has consistently positioned itself as a pioneer in this field, enabling customers to achieve significant breakthroughs in AI technology. According to Michael Intrator, Co-Founder and CEO of CoreWeave, the launch serves as a testament to their engineering excellence and commitment to rapidly delivering cutting-edge systems. "CoreWeave is built to move faster – and time and again, we've proven it by being first to operationalize the most advanced systems at scale," says Intrator. This ethos of speed and innovation is crucial, as organizations across various sectors are racing to leverage reasoning models to build transformative AI applications that revolutionize both work and leisure activities.
Ian Buck, NVIDIA's Vice President of Hyperscale and HPC, emphasizes the importance of CoreWeave's swift deployment of these systems. He remarked that such advances are essential for turning AI infrastructure and software into the tangible tools that make "AI factories" a real possibility. With this infrastructure in place, organizations can unleash the potential of AI, turning complex models into functional applications with unprecedented efficiency.
CoreWeave continues to break performance records in the field of AI, having recently achieved a noteworthy new industry benchmark in AI inference using the NVIDIA GB200 Grace Blackwell Superchips. Reported in the latest MLPerf v5.0 results, this achievement indicates that CoreWeave’s systems are making significant strides in establishing ML performance standards. This is a reflection of their ongoing commitment to maximizing efficiency and setting new records, showcasing their role at the forefront of AI advancements.
In addition to the groundbreaking GB200 rollout, CoreWeave has previously been among the first to adopt and integrate the NVIDIA H100 and H200 GPUs within their services. Their offerings include a full stack of optimized cloud solutions tailored for the new NVIDIA GB200 NVL72, ensuring performance reliability through services like CoreWeave Kubernetes Service, Slurm on Kubernetes (SUNK), and CoreWeave Mission Control. Furthermore, the NVIDIA Blackwell-accelerated instances have the capability to scale up to an impressive 110,000 Blackwell GPUs, augmented by NVIDIA Quantum-2 InfiniBand networking.
Notably, apart from Cohere, IBM, and Mistral AI, CoreWeave has recently sealed a multi-year contract with OpenAI, the creators behind the popular AI platform, ChatGPT. This partnership further solidifies CoreWeave’s standing as a primary player in the rapidly evolving AI landscape, showcasing their cloud technology as critical for enabling enhanced performance and innovation.
With the cloud-based computing environment continuously evolving, CoreWeave's role is likely to grow even more significant in the upcoming years. Their dedication to providing advanced computing resources at scale is not only transforming how enterprises operate but is also setting the stage for future advancements in AI technology. As they continue to pioneer and develop new solutions, industry leaders have their eyes firmly set on CoreWeave for inspiration and progress in the field of artificial intelligence.
Through their innovative approaches and the launch of the NVIDIA GB200 systems, CoreWeave is empowering businesses to innovate faster and explore new horizons in AI, ultimately leading to a future where AI-driven applications readily enhance everyday life.
Topics Consumer Technology）
【About Using Articles】
You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.
【About Links】
Links are free to use.