CPU Compute

Get more out of your AI infrastructure

‍

<Versatile CPU instances> to support AI workloads

CoreWeave clusters utilize cutting edge CPUs, helping your teams get more out of GPU compute.

Superpower compute-intensive projects with powerful NVIDIA Grace CPUs and the latest AMD and Intel chips.

Use performant CPUs to support GPU-accelerated workloads

Available in high-performance, high-core, and general-purpose configurations
Up to 26% better gen-to-gen performance compared to general-purpose Milan servers
Supporting up to 1.5TB of memory per system
Equipped with NVIDIA Bluefield DPUs

CoreWeave uses powerful and efficient CPUs to enable higher utility and performance from GPU clusters.

Let CPUs handle generalized computing needs, freeing GPUs to focus on high-intensity tasks.

Get better performance

Eliminate the unnecessary overhead of virtualization and get the benefits of bare metal

Work at hyperspeed

Scale across hundreds of thousands of CPU cores in seconds—with boost clock speeds of up to 4.3Ghz+

Use a faster network

Access up to 384 cores at up to 200Gbps speeds for ultra-fast data movement

Let GPUs focus on building LLMs by using the right CPUs as supporting players.

Get the compute you need for data preprocessing, data analytics, and other critical tasks in your instance.