European Cloud Providers Back NVIDIA Vera CPUs for Agentic AI Infrastructure

Why agents put pressure on CPUs

NVIDIA argues that agentic computing changes what data centers need. In its words, agents do more than generate answers: they retrieve information, run software tools, execute code and evaluate results. According to the company, each of those steps depends on fast CPU processing.

NVIDIA says Vera accelerates that work, which helps agents respond faster and lets data centers make better use of their computing capacity. The company also claims Vera is designed to sustain its speed while running thousands of tasks simultaneously, so organizations can expand services without trading responsiveness for scale.

Cloud providers and server makers

AI cloud providers Nebius, Nscale and Verda are building with Vera to support the growing computing needs of businesses and developers. NVIDIA says high-performance, energy-efficient CPUs help these providers scale the data processing and software execution behind their services, alongside the GPUs that run models.

The company also points to the wider sovereign AI ecosystem, which includes Aleria, a sovereign AI infrastructure and platform provider. Aleria supports demand for infrastructure that keeps data and operations under customers' control.

On the hardware side, Bull, Dell Technologies, HPE, Lenovo and Supermicro are building Vera-based systems. NVIDIA says this gives customers more deployment options for high-performance computing and AI data centers. Vera can power standalone CPU servers or work alongside NVIDIA GPUs in accelerated systems.

Early results from data and software companies

A group of data processing companies reported strong results on Vera in early testing, according to NVIDIA. The list includes ClickHouse, Cloudera, Duck Labs, Elastic, Polars, Redpanda, SAP and Starburst. The workloads covered analytical processing, extract-transform-and-load jobs and semantic search. NVIDIA credits Vera's per-core performance, LPDDR5X memory bandwidth and NVIDIA Scalable Coherency Fabric for the gains.

Software and developer-platform companies are also adopting the chip for environments where AI agents work alongside people:

• Atlassian is looking to Vera to accelerate the data processing that builds semantic search indexes for its AI-powered products.
• Inherent has tested Vera and is looking to adopt it for environments where agents work alongside engineers on novel scientific research.
• Nous Research uses Vera to authenticate agent identities and validate every action before it executes, keeping deployments secure as agent fleets grow.

Sandboxes for AI agents

Baseten-Blaxel, Daytona, E2B, Modal, Prime Intellect and Vercel provide cloud-based sandboxes, which NVIDIA describes as computers for AI agents. They are building on Vera to power the infrastructure agents rely on to execute code, run tools and test results.

NVIDIA says every agent needs a secure, isolated sandbox that can be provisioned instantly. Because sandboxes spin up millions of times a day, the company notes, the speed of each startup cycle compounds at scale. It says Vera's fast per-core performance and high-bandwidth memory make it well suited to this pattern.

What it means for the market

The announcement shows NVIDIA positioning its CPU line as a core part of agent infrastructure, not only a companion to GPUs. For European customers, the named cloud providers and the sovereign AI angle point to local capacity that keeps data and operations under their control. Server vendors offering Vera systems widen the purchasing options. Many of the data processing results come from early testing, and the source gives no pricing, availability dates or benchmark figures.