Ian Finder, head of Nvidia's data center CPU business, refuted claims that the Vera CPU was developed with the specific intention of being an "agentic CPU." He described the idea that Nvidia predicted the "agentic wave" four years ago, when Vera's development started, as "absurd."
Finder clarified that while the company possesses strong "design intuition," it does not have a "magic crystal ball" to foresee future technological shifts like the agentic moment.
Instead of a specific agentic design, Vera's development was characterized by its co-design with Nvidia's accelerators. This approach, which began with the Grace data center CPU, assumes that data centers and AI workloads are accelerated.
The primary goal of this co-design is to create a CPU that can efficiently support and feed the accelerator, starting from a "conceptually simple space" of what kind of CPU is needed for this purpose.
Finder used the term "opinionated" to describe aspects of both Vera and Grace. Grace introduced an "opinionated memory system" with Nvidia's Scalable Coherency Fabric.
Vera extends this memory system and incorporates an "opinionated microarchitecture" through Nvidia's Olympus core. This indicates a deliberate and specific design choice rather than a broad, general-purpose approach.
Nvidia's strategy in this CPU space is described as starting from a "clean sheet of paper," contrasting with the "legacy constraints" faced by established CPU manufacturers like Intel and AMD. These competitors have large customer bases with specific existing needs.
Nvidia's objective with Vera is not to directly take market share from AMD and Intel in their traditional segments, but rather to capture market share within the CPU sector that is expanding due to the growth of AI agents.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's briefNew every morning, and the back catalogue is archived by date.
Nvidia's head of data center CPU business, Ian Finder, stated it is "absurd" to suggest the Vera CPU was designed with foreknowledge of the "agentic moment" in AI. Instead, Vera's development focused on co-design with Nvidia's accelerators, building on the approach used for the Grace CPU. This strategy aims to support accelerated data center and AI workloads by ensuring the CPU efficiently feeds the accelerator.