CoreWeave Breaks New Ground with NVIDIA Blackwell Cloud Instances
In a groundbreaking development for the field of cloud computing and artificial intelligence, CoreWeave has unveiled the availability of NVIDIA GB200 NVL72-based instances, introducing the NVIDIA Blackwell platform to a wider audience for the first time. This momentous launch promises to revolutionize AI performance and scalability in the cloud, setting a new standard for cutting-edge computing solutions.
Unleashing the Power of NVIDIA GB200 NVL72
At the heart of this innovation lies the NVIDIA GB200 NVL72 platform, a state-of-the-art, liquid-cooled system that boasts a 72-GPU NVLink domain. This unique configuration transforms individual GPUs into a unified, powerhouse unit, significantly boosting computational capabilities. With groundbreaking features like the fifth-generation NVLink delivering 130TB/s of GPU bandwidth and the second-generation Transformer Engine supporting FP4 for enhanced AI performance, the NVIDIA Blackwell platform offers unparalleled speed and accuracy.
CoreWeave’s Tailored Cloud Services for Blackwell
CoreWeave has tailored its cloud services to maximize the potential of the NVIDIA Blackwell platform. The CoreWeave Kubernetes Service streamlines workload scheduling, while Slurm on Kubernetes (SUNK) intelligently distributes workloads across the system. Moreover, CoreWeave’s Observability Platform provides real-time insights into NVLink performance and GPU utilization, ensuring optimal efficiency and performance for AI applications.
Empowering Enterprises with Full-Stack Accelerated Computing
NVIDIA’s full-stack AI platform, combined with CoreWeave’s Blackwell-powered infrastructure, equips enterprises with the tools needed to develop scalable AI models and agents. From customizable workflows with NVIDIA Blueprints to secure model deployment via NVIDIA NIM and model training with NVIDIA NeMo, the NVIDIA AI Enterprise software platform offers a comprehensive suite of solutions for AI development on CoreWeave’s cutting-edge infrastructure.
Pioneering the Next Generation of AI in the Cloud
The introduction of NVIDIA GB200 NVL72-based instances on CoreWeave’s platform signals a joint commitment to pushing the boundaries of computing technology. By providing access to high-performance cloud instances optimized for AI reasoning models, CoreWeave and NVIDIA are paving the way for the future of AI development. Enterprises looking to leverage these powerful cloud-based solutions can now do so through the CoreWeave Kubernetes Service in the US-WEST-01 region.
For enterprises seeking to harness the full potential of NVIDIA Blackwell in the cloud, CoreWeave’s official channels offer detailed information and provisioning options. This collaboration between CoreWeave and NVIDIA marks a significant milestone in advancing AI capabilities in the cloud, opening up new possibilities for innovation and growth in the field of artificial intelligence.
As the boundaries of AI and cloud computing continue to expand, CoreWeave and NVIDIA are at the forefront of driving innovation and empowering enterprises to unlock the full potential of AI technologies in the digital age.
