Entity trail

Vera CPU

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-06 / rss-researcherNVIDIA open-sources cuFile and introduces SCADA to let GPUs pull only the bytes they need from storageAt FMS, NVIDIA open-sourced its cuFile APIs so GPUs — not just CPUs — read and write storage directly in microseconds, and introduced SCADA (scaled, accelerated data access), a framework letting massively parallel GPUs pull only application-necessary data from storage into high-bandwidth memory. It also showed the Vera CPU in the Vera BlueField-4 STX storage processor delivering up to 3.21x higher throughput than an x86 CPU on a two-stage compression-and-encryption pipeline. Over 40 storage and flash vendors are in the Storage-Next initiative, with DDN integrating SCADA into Infinia; the driver is KV-cache and long-context working sets outgrowing system memory.
2026-07-25 / news-researcherAMD Launches Helios Rack-Scale AI System With 72 GPUs, Adds Microsoft as CustomerAMD unveiled its first rack-scale system to directly contest Nvidia's rack-level position, with engineering samples shipping in H2 2026 and mass production targeted for Q2 2027. Microsoft joins Meta, OpenAI, and Oracle as customers; Meta plans to deploy 1 gigawatt in Helios racks by year-end against a long-term commitment of up to 6 GW of AMD GPUs. AMD is already discussing MI500 and Verano CPUs for 2027, with the MI500 generation beginning a shift toward optical scale-up networking.
2026-06-13 / news-researcherChips and Cheese Interviews Intel's Xeon 6 Product Director Kira BoykoA technical interview with Intel's Xeon 6 product director Kira Boyko details architecture and product-direction decisions behind the latest Xeon generation, relevant as CPUs increasingly handle inference-adjacent and orchestration workloads in AI pipelines. The deep-dive format surfaces forward-looking signals on data-center silicon strategy that mainstream coverage misses. Worthwhile for builders weighing CPU vs. accelerator tradeoffs for serving and agent workloads.
2026-06-02 / skill-finderNVIDIA Formally Launches Vera Rubin Data Center Platform at GTC Taipei — Six Co-Designed Chips for Next-Gen AIJensen Huang formally launched the NVIDIA Vera Rubin data center platform at GTC Taipei (co-located with Computex, June 2-5), comprising six co-designed chips: Vera CPU, Rubin GPU, NVLink 6 Switch, ConnectX-9 SuperNIC, BlueField-4 DPU, and Spectrum-6 Ethernet Switch. The platform is the successor to Blackwell/Grace Hopper, designed to slash both AI training time and inference token costs through extreme cross-chip co-design. This positions NVIDIA's next-generation stack for large-scale agentic and frontier model workloads.
2026-06-01 / rss-researcherNVIDIA Vera Rubin Platform Now in Production — 10x Agentic Throughput vs Blackwell, Shipping Fall 2026NVIDIA announced at Computex that Vera Rubin is in full production on TSMC 3nm, combining an 88-core Vera CPU with next-gen Rubin GPU. The platform delivers 10x agentic AI throughput, 10x reduction in inference token cost, and 4x reduction in GPUs needed to train MoE models versus Blackwell. AWS, Google Cloud, Microsoft, OCI, CoreWeave, Lambda, and Nebius will be among the first to deploy Vera Rubin instances in late 2026.
2026-06-01 / thought-leaders-researcherJensen Huang: 'AI Is Now a GDP Generator' — Claims Agentic AI Triples Developer Productivity to $9T OutputAt GTC Taipei, Jensen Huang declared 'AI is now a profit generator. AI is now a GDP generator,' claiming agentic AI enables software developers to triple productivity — generating $9 trillion in output from $3 trillion in salaries. He unveiled Vera Rubin, a multi-rack pod-scale system integrating GPUs, CPUs, storage, and networking for agentic inference with NVLink 72 and advanced liquid cooling. NVIDIA's 150 Taiwan supply chain partners enabled rapid production. This frames agentic AI as an economic multiplier, not just a cost cutter.
2026-06-01 / thought-leaders-researcherJensen Huang: NVIDIA Vera CPU Unveiled — First Processor Purpose-Built for AI Agents, Anthropic and OpenAI Are Launch CustomersJensen Huang announced Vera, a standalone data center CPU with 88 custom Olympus cores, 227 billion transistors, and 1.2TB/s LPDDR5X memory bandwidth, designed specifically for agentic AI workloads. NVIDIA claims 1.8x faster task completion than x86 CPUs for agent inference. Initial customers include Anthropic, OpenAI, and SpaceX, with mass shipments beginning fall 2026. This is the first CPU designed around agent workloads rather than general compute, marking a new hardware category.
2026-05-27 / news-researcherNVIDIA Vera CPU First Benchmarks: Custom 88-Core Olympus ARM Chip Outperforms AMD EPYC 9575F and Intel Xeon 6980PPhoronix published first independent benchmarks of NVIDIA's Vera CPU with 88 custom Olympus ARM cores. Results: 10% faster than AMD EPYC 9575F (5.0GHz), 1.55x Intel Xeon 6980P, and 1.63x NVIDIA's own Grace. Olympus features a 10-wide instruction front-end (matching Apple M silicon), 1.2 TB/s memory bandwidth, neural branch predictor evaluating two branches per cycle, and claimed 1.5x IPC improvement over previous gen. NVIDIA's boldest move into the general-purpose CPU market.
2026-05-23 / thought-leaders-researcherJensen Huang: Vera Rubin Opens 'Brand New $200B Market' — NVIDIA Already at $20B in CPU RevenueDuring NVIDIA's earnings call, Jensen Huang declared Vera Rubin the company's 'major new growth driver,' calling it 'even more successful than Grace Blackwell.' Vera Rubin, focused on powering agentic AI, opens a $200 billion total addressable market for NVIDIA. The company has already sold $20 billion worth of standalone Vera CPUs this year alone, putting NVIDIA on track to become the world's largest CPU supplier — a market it wasn't even in two years ago.
2026-05-19 / news-researcherNVIDIA Hand-Delivers First Vera CPUs to Anthropic, OpenAI, SpaceXAI, and Oracle — 88 Custom Cores for Agentic AINVIDIA VP Ian Buck hand-delivered the first Vera CPUs to Anthropic, OpenAI, SpaceXAI, and Oracle Cloud Infrastructure over the weekend. Vera is NVIDIA's first custom CPU, featuring 88 Olympus cores, 1.2 TB/s memory bandwidth, and 50% faster per-core performance than predecessors — purpose-built for agentic AI workloads. Oracle plans to deploy hundreds of thousands of Vera CPUs, and Jensen Huang positioned the standalone CPU as NVIDIA's next multi-billion-dollar business line.
2026-05-10 / news-researcherNVIDIA Vera Rubin Platform Enters Full Production — Cloud Deployments Begin H2 2026NVIDIA's next-generation Vera Rubin platform (comprising seven chips, five rack-scale systems, and one supercomputer purpose-built for agentic AI) has entered full production. AWS, Google Cloud, Microsoft Azure, and OCI will be the first to deploy Vera Rubin-based instances in H2 2026. The platform includes the new Vera CPU and BlueField-4 STX storage architecture, positioning it as the successor to the current Blackwell generation.
2026-04-25 / news-researcherNVIDIA Vera Rubin Platform in Full Production — Seven New Chips for Training and Agentic InferenceNVIDIA announced the Vera Rubin platform with seven chips now in full production: Vera CPU, Rubin GPU, NVLink 6 Switch, ConnectX-9 SuperNIC, BlueField-4 DPU, Spectrum-6 switch, and the newly integrated Groq 3 LPU. Designed for massive-scale pretraining through real-time agentic inference, Rubin-based products will be available from AWS, Google Cloud, Microsoft, and OCI in H2 2026.

Graph relationships

RELEASED
NVIDIA
NVIDIA announced Vera CPU with custom Olympus cores.
Source finding
RELEASED
NVIDIA
NVIDIA announced Vera CPU at GTC 2026.
Source finding
RELEASED
NVIDIA
NVIDIA released Vera CPU
Source finding
RELEASED
NVIDIA
NVIDIA released Vera CPU with 88 Olympus ARM cores
Source finding
RELEASED
NVIDIA
NVIDIA unveiled Vera data center CPU for AI agents
Source finding
RELEASED
NVIDIA
NVIDIA released Vera CPU, its first custom CPU with 88 cores.
Source finding
BENCHMARKED AGAINST
Intel Xeon 6980P
Vera CPU is 1.55x faster than Intel Xeon 6980P
Source finding
BENCHMARKED AGAINST
AMD EPYC 9575F
Vera CPU is 10% faster than AMD EPYC 9575F
Source finding
USES
SpaceX
SpaceX is a launch customer for Vera CPU
Source finding
USES
OpenAI
OpenAI is a launch customer for Vera CPU
Source finding
USES
Anthropic
Anthropic is a launch customer for Vera CPU
Source finding
RELEASED
NVIDIA -> Vera CPU
NVIDIA announced Vera CPU with custom Olympus cores.
Source finding

Source trail

Announced at FMS The platform Announced at Computex The largest European AI infrastructure investment ever Native integrations Source TechCrunch Chips and Cheese (Hacker News)NVIDIA Blog Phoronix TechCrunch NVIDIA AI Blog

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…