CoreWeave Brings Multi-Rack Nvidia Vera Rubin NVL72 Cluster Online — To Give Agentic AI Customers Greater Scale, Faster Iteration

• CoreWeave’s multi-rack deployment connects hundreds of Nvidia Rubin GPUs into one cluster.
• Each Vera Rubin NVL72 rack combines 72 Rubin GPUs and 36 Vera CPUs.
• CoreWeave added cross-region write acceleration and a new lower-cost Archive storage tier.

CoreWeave Inc. (CRWV) stock is in focus on Wednesday after the company said it brought up a multi-rack Nvidia Vera Rubin NVL72 cluster on CoreWeave Cloud, connecting hundreds of Nvidia Rubin graphics processing units (GPUs) into a single system for agentic AI workloads.

The multi-rack setup allows training and inference workloads to run across hundreds of Rubin GPUs, providing capacity for larger models, more demanding inference and reinforcement learning at scale, according to CoreWeave.

Hundreds Of Rubin GPUs Connected As One Cluster

A single Nvidia Vera Rubin NVL72 rack combines 72 Rubin GPUs with 36 Vera CPUs, alongside Nvidia networking and data-processing technology. CoreWeave said multiple racks are connected via Nvidia Spectrum-X Ethernet to operate as a single scale-out cluster.

CoreWeave was the first AI cloud provider to validate and bring up a Vera Rubin NVL72, according to Chen Goldberg, Executive Vice President of Product and Engineering.

“With multi-rack Vera Rubin, we are connecting hundreds of Rubin GPUs as a single scale-out cluster,” Goldberg said, adding that the system gives customers building agentic AI greater scale and faster iteration.

CoreWeave's infrastructure automates rack setup, firmware updates, validation, power and cooling. It also tests full racks before moving them into production.

CoreWeave Adds New Storage Features

Alongside the compute expansion, CoreWeave introduced cross-region write acceleration for its AI Object Storage platform. This allows AI workloads to write data in the region where they run, while CoreWeave copies it to another region in the background. This can reduce pauses for customers running training workloads across multiple regions, according to the company.

The AI cloud-computing company also introduced an Archive tier, a lower-cost storage option for data customers want to retain longer. This tier has no retrieval fee, early-deletion fee, or fee for reading data from the Archive tier.

Its Local Object Transport Accelerator, or LOTA, can also cache data closer to AI workloads, with CoreWeave saying it reduces read latency by 8x compared with reading from a traditional storage cluster.

What Retail Investors Think About CRWV Hunt Stock

Retail sentiment around CRWV stock continues to remain in 'bearish' territory on Stocktwits. 

So far this year, CoreWeave shares have risen around 2%. In comparison, ARK Innovation ETF (ARKK), which holds CRWV stock, has risen over 6% while the iShares AI Innovation and Tech Active ETF (BAI) has risen more than 25% during this period.

See Also: Salesforce Gets Fresh Target Hike From Mizuho After Dreamforce — Analyst Says 2026 Setup Looks ‘Much Better’ Than 2025

For updates and corrections, email newsroom[at]stocktwits[dot]com.