Latest

Nvidia Vera CPU Leaked: 1.5x Performance Over x86

Noah

24 May 2026 — 1 min read

NVIDIA is poised to unveil its Vera Rubin CPU at Computex 2026, promising a 1.5x speed boost and 2x overall performance over x86 competitors, according to GF Securities. The Arm-based processor, featuring 88 custom Olympus cores on TSMC 3nm, targets the burgeoning Agentic AI inference market with up to 1.5TB of LPDDR5X memory and a 1.8TB/s NVLink-C2C interconnect. Shipments are forecast to reach 4.2 million units by fiscal 2028 as the server CPU market swells to $211 billion by 2030.

Vera Rubin CPU Specs: 88 Arm Cores, 3nm Node, and KV Cache Optimization

The Vera Rubin CPU is built around 88 Armv9.2 Olympus cores fabricated on TSMC's 3nm process, offering a claimed 4x density advantage over x86 rivals. It supports up to 1.5TB of LPDDR5X memory and features a 1.8TB/s NVLink-C2C interconnect for tight GPU coupling. The chip is specifically optimized for Agentic AI inference workloads and KV cache management, signaling NVIDIA's push to dominate the AI compute stack beyond GPUs.

Shipment Forecasts and Market Growth Driven by Agentic AI

GF Securities projects NVIDIA will ship 1.2 million Vera Rubin units in fiscal 2027, ramping to 4.2 million in fiscal 2028. This aligns with a broader explosion in server CPU demand: the total addressable market for server CPUs is expected to reach $211 billion by 2030, fueled by Agentic AI workloads. Annual CPU unit demand is forecast to grow from 3.7 million in 2026 to 16.3 million in 2028, highlighting the massive shift toward Arm-based architectures in the data center.

NVIDIA's Arm Ecosystem Play Challenges x86 Incumbents

The Vera Rubin launch is a strategic move to solidify NVIDIA's Arm-based CPU ecosystem against entrenched x86 players. By combining the CPU with its NVLink-C2C interconnect and existing GPU dominance, NVIDIA aims to create a tightly integrated platform that can handle end-to-end AI inference pipelines. The forecasted shipment ramp—from 1.2M to 4.2M units in just one year—underscores the urgency with which hyperscalers are expected to adopt the new architecture.

NVIDIA and SK hynix Partner on AI Factory Memory

NVIDIA and SK hynix have entered a multiyear technology partnership to co-develop next-generation memory specifically designed for AI factories, aligning tightly with NVIDIA’s infrastructure roadmap. The collaboration spans memory for Vera Rubin supercomputers, Vera CPUs, RTX Spark PCs, and Jetson Thor robots, marking a deep integration between chip design

Goose: AI agent runs code, not just suggests

The Linux Foundation has officially unveiled Goose, an open-source AI agent that marks a paradigm shift from advisory tools to autonomous code execution, capable of installing, editing, testing, and running code without human intervention. Built in Rust and boasting over 70 extensions via the Model Context Protocol, Goose connects to

Accenture: Enterprise AI Reaches Scale in 2026

Enterprise AI is approaching a critical inflection point in 2026, as Accenture predicts a transition from experimental pilots to large-scale deployment, with customer-facing AI solutions and agentic systems moving into production. However, three major barriers—data infrastructure and cloud migration, enterprise knowledge bases, and governance plus workforce transformation—must be

Kimi Work Beta: Smarter Office Agent for Chinese Workplace

Chinese AI startup Kimi has launched Kimi Work, a local universal agent specifically designed for knowledge workers, offering a pragmatic alternative to broad-spectrum AI assistants. Early beta test results reveal strong performance in routine office tasks like report generation and data collection, though advanced capabilities remain behind a paywall. Kimi