Nvidia Expands AI Infrastructure Push as Startup Modal Adopts Its New Vera CPUs

Nvidia is expanding its presence in the artificial intelligence infrastructure market as cloud-computing startup Modal plans to adopt the company's new Vera central processing units (CPUs) to run AI agents. The move marks a significant development in Nvidia's efforts to compete in a processor market traditionally dominated by Intel and AMD.

According to[ [Reuters' report on Nvidia's agreement with Modal]](https://www.reuters.com/business/nvidia-wins-fast-growing-ai-startup-modal-new-central-processors-2026-10-19/))(, Modal plans to use Vera CPUs to host AI agents that perform extended tasks on behalf of users. The company has historically relied on processors from Intel and AMD for these workloads.

The development highlights how the growing use of AI agents is creating demand for computing infrastructure designed not only to run AI models but also to execute code, process information, coordinate tools, and manage complex tasks.

What Is Nvidia's Vera CPU?

Nvidia's Vera CPU is a processor designed to handle the computing work behind agentic AI, reinforcement learning, and data processing. While graphics processing units (GPUs) are widely used to train and run AI models, CPUs remain essential for managing the surrounding operations that allow AI systems to function efficiently.

These operations include running software environments, executing code, handling data pipelines, coordinating tools, and managing interactions between AI models and external systems.

Nvidia says its Vera platform is designed to improve performance and energy efficiency for these workloads. The company has positioned the processor as part of a broader approach to AI infrastructure in which CPUs and GPUs work together to support increasingly complex AI applications.

More details are available in Nvidia's official overview of the Vera CPU.

Why Modal Is Adopting Vera

Modal provides cloud computing infrastructure that allows developers to run AI applications without having to manage all the underlying computing hardware themselves.

As AI agents become capable of completing longer and more complicated tasks, infrastructure providers need to support more than conventional model inference. Agents may repeatedly call tools, execute programs, retrieve information, and operate within isolated software environments before producing a final result.

These activities can create substantial CPU workloads alongside the GPU computing required by AI models.

Reuters reported that early tests of Nvidia's Vera processors delivered approximately twice the task performance of the Intel and AMD alternatives Modal had been using for the tested workloads. The results helped motivate Modal's decision to incorporate Vera into its infrastructure.

However, this reported improvement relates to early tests and should not be interpreted as proof that Vera is twice as fast for every application. Actual performance depends on the workload, software configuration, and hardware being compared.

Modal is not abandoning its existing suppliers. The company intends to use a mixed hardware strategy, choosing processors according to performance and the requirements of particular workloads.

Nvidia's Strategy Beyond GPUs

Nvidia has become a central supplier of AI computing infrastructure through its GPUs, networking technologies, software ecosystem, and integrated data-centre systems.

Its expansion into CPUs represents another part of that strategy. Rather than concentrating exclusively on the processors that train and run AI models, Nvidia is targeting the wider infrastructure required to operate AI systems at scale.

This approach is particularly relevant as businesses move from conversational AI tools towards agentic systems that can perform multiple steps with limited ongoing human intervention.

Such systems need computing resources to plan tasks, execute code, manage intermediate results, and interact with other software. Improving the performance of these operations could help infrastructure providers increase throughput and use computing resources more efficiently.

Vera also fits into Nvidia's broader AI infrastructure platforms, where CPUs coordinate operations and work alongside GPUs and other specialised components.

What This Means for Intel and AMD

Intel and AMD have long supplied CPUs used in servers, cloud infrastructure, and enterprise computing. Nvidia's effort to win customers for Vera introduces additional competition in a market that has historically been separate from its dominant GPU business.

Modal's planned adoption is notable because the startup previously relied on Intel and AMD processors for the relevant workloads. Its decision provides an early indication that Nvidia may be able to compete for CPU deployments in AI-focused cloud infrastructure.

However, one customer's adoption does not establish a broad shift in market share. Intel and AMD remain important suppliers, and infrastructure companies can continue using multiple processor architectures depending on compatibility, price, availability, and performance.

The longer-term competitive impact will depend on how Vera performs in production, how readily developers adapt their software, and whether its benefits justify the cost of deployment.

The Growing Importance of AI Infrastructure

The AI industry is increasingly focused on the systems that support AI applications, not just the models themselves.

As AI agents become more sophisticated, the computing required to run them can involve repeated cycles of reasoning, tool use, code execution, and data processing. These activities create demand for infrastructure that can handle large numbers of concurrent tasks reliably and efficiently.

For cloud providers, the ability to deliver more useful work from a given amount of computing capacity can affect operating costs, service responsiveness, and the economics of offering AI services.

This creates opportunities for specialised processors, improved memory systems, faster networking, and integrated hardware platforms. It also increases competition among infrastructure suppliers seeking to become essential to the next generation of AI applications.

Modal's decision to adopt Vera is one example of how this shift is influencing hardware choices.

What Happens Next?

The next important test will be how Nvidia's Vera CPUs perform as Modal incorporates them into its infrastructure and serves real customer workloads.

Key factors will include sustained performance, reliability, energy efficiency, software compatibility, and the cost of running AI agents at scale. The results could help determine whether other cloud infrastructure providers expand their use of Vera for similar workloads.

Modal's continued use of processors from multiple suppliers also suggests that the market will remain competitive rather than shifting immediately towards a single hardware provider.

Conclusion

Nvidia's push into the CPU market is expanding its role in the infrastructure that powers modern artificial intelligence. Modal's planned adoption of Vera CPUs highlights the growing importance of processors capable of handling the code execution, orchestration, and data-processing tasks associated with AI agents.

The reported early performance gains offer an encouraging signal for Nvidia, but broader adoption will depend on real-world results and the economics of deployment.

As AI applications evolve from generating responses to completing complex tasks, competition is likely to extend across the entire computing stack. Nvidia's progress with Vera could become an important part of that next phase of AI infrastructure development.