Key Takeaways
- Nvidia’s newly launched Vera CPU has reached production, with initial systems delivered to leading AI organizations such as Anthropic, OpenAI, Oracle Cloud Infrastructure, and xAI.
- Ian Buck, Nvidia’s VP, personally delivered these inaugural units, starting with Anthropic’s headquarters in San Francisco.
- The Vera processor boasts 88 specialized Olympus cores, memory bandwidth of 1.2 TB/s, and achieves 50% superior per-core performance versus conventional CPU architectures.
- Oracle Cloud Infrastructure has announced plans for massive Vera CPU deployment in 2026, becoming the inaugural hyperscale cloud provider to embrace Vera technology.
- xAI is conducting tests with Vera for reinforcement learning applications, with Elon Musk receiving a detailed system walkthrough.
Nvidia (NVDA) has successfully transitioned its Vera CPU from concept to reality, shipping initial production units to prominent players in the artificial intelligence sector.
Ian Buck, Nvidia’s VP of Hyperscale and HPC, conducted personal deliveries of these first production units to Anthropic, OpenAI, Oracle Cloud Infrastructure, and xAI during a two-day tour last week.
The delivery tour kicked off at Anthropic’s headquarters in San Francisco’s SoMa district, where James Bradbury, who oversees compute operations at Anthropic, accepted the hardware. Buck demonstrated the chip’s unique features using an exposed Vera CPU motherboard.
“Compute scaling represents a crucial catalyst for model advancement,” Bradbury commented. “Vera’s arrival as a valuable ecosystem component for agentic workload solutions is genuinely exciting for us.”
The subsequent stop brought Buck to OpenAI’s Mission Bay campus, where Sachin Katti, responsible for compute infrastructure, met him on an exterior terrace. During their discussion, Buck took out a screwdriver to open the system and display its internal architecture.
Musk Receives Technical Briefing
The day’s concluding delivery reached xAI’s Palo Alto location, where Nvidia’s delegation provided Elon Musk with an in-depth technical presentation. Musk inquired about core specifications, memory architecture, and thermal management systems.
xAI is currently assessing Vera’s capabilities for reinforcement learning processes and agent-based simulation frameworks within its training infrastructure.
The following Monday, Buck traveled to Oracle’s AI Customer Excellence Center in Santa Clara, where OCI’s product and customer success teams examined the unpacked system firsthand.
“OCI will implement hundreds of thousands of Nvidia Vera CPUs starting in 2026, as agentic AI requires consistent performance at enormous scale,” stated Karan Batta, OCI’s overall product management leader.
OCI stands as the pioneering cloud service provider committing to hyperscale Vera implementation.
Understanding Vera’s Purpose
Vera represents Nvidia’s inaugural custom CPU design, engineered explicitly for agentic AI workloads — scenarios where models perform actions, execute code, invoke tools, and maintain extended context states rather than simply responding to queries.
While GPUs manage intensive computational tasks, the surrounding orchestration infrastructure — encompassing tool invocations, data transfers, sandboxing, and retrieval operations — depends on CPU processing. Vera was designed specifically for these functions.
The processor integrates 88 proprietary Olympus cores, delivers 1.2 TB/s memory bandwidth, and provides 50% enhanced per-core performance at maximum capacity compared to conventional CPU designs.
“When AI models receive questions, the responses aren’t typically pre-generated,” Buck explained during the Oracle meeting. “The models must frequently generate Python code to produce accurate answers. This explains the surging demand for CPU capabilities.”
Vera additionally functions as the host processor within the Vera Rubin NVL72 architecture, connecting with Rubin GPUs through second-generation NVLink-C2C in a unified memory framework. According to Nvidia, this configuration operates at twice the energy efficiency of conventional setups.
Jensen Huang unveiled Vera during GTC San Jose in March, positioning it as Nvidia’s upcoming multi-billion dollar revenue stream.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
