NVIDIA announced 'Vera,' a CPU for AI factories (large-scale compute). All five sources are NVIDIA official—a single-vendor hardware announcement, announcement-led but touching a core component of the compute base. As a CPU to support GPU-centric AI infrastructure, the focus is power efficiency and data-supply bandwidth; the through-line is less models or software than optimizing the whole system for running AI at scale. It reads as a vertical-integration move—locking down not just the GPUs that compute but the surrounding CPU, memory, and interconnect in-house. But this is mainly a product announcement—actual performance and efficiency, adoption timing, and supply outlook remain to confirm.
NVIDIA × Infrastructure & Hardware
NVIDIA launches Vera CPU for AI
NVIDIA launches Vera CPU for AI
✎ Story body
▲ Official & Press
Official
NVIDIA Vera CPU Sets a New Standard for Agentic Workloads in AI Factories
NVIDIA's Vera CPU sets a new standard for agentic AI workloads
Press
Microsoft、NVIDIAのSoC搭載でAI特化のミニPC「Surface RTX Spark Dev Box」披露
Microsoft unveils Surface RTX Spark Dev Box, an AI dev mini PC
Official
Build Personal AI Agents on Windows PCs with New Tools from Microsoft and NVIDIA
NVIDIA, Microsoft unveil new tools to build personal AI agents on Windows PCs
Official
Deploy Self-Evolving Agents for Faster, More Secure Research with a Hermes Agent and NVIDIA NemoClaw
NVIDIA on building self-evolving agents for faster, secure research
Official
Deploy Agentic-Ready AI at the Edge with Memory Efficiency in NVIDIA JetPack 7.2
NVIDIA boosts edge agent AI memory efficiency in JetPack 7.2
Press
NVIDIAの“狐”は工場自律管理AIエージェント、台湾メーカーが導入効果を確認
NVIDIA unveils FOX, a blueprint for autonomous factory-management AI agents
Official
Run Local AI Agents with Faster Models and Multi-Node Clustering on NVIDIA DGX Spark
NVIDIA runs local AI agents faster with multi-node DGX Spark clustering
Press
NVIDIAの「NemoClaw」でエッジAIを統合管理、アドバンテックが「WEDA」を発表
Advantech unveils WEDA for integrated edge AI management with NVIDIA NemoClaw
Press
New Server Hopes to Break Through AI’s “Memory Wall”
Majestic Labs' 128TB Prometheus server targets AI's memory wall
Official
How to Post-Train Autonomous Vehicle Models in Closed-Loop with NVIDIA Alpamayo
NVIDIA details closed-loop post-training for AV models via Alpamayo
Official
Welcome NVIDIA Cosmos 3: The First Open Omni-model for Physical AI Reasoning and Action
NVIDIA releases Cosmos 3, first open omni-model for Physical AI
Official
Develop Physical AI Reasoning, World, and Action Models with NVIDIA Cosmos 3
Building Physical AI reasoning, world and action models with Cosmos 3
Official
Advancing AI Infrastructure for Agentic AI with NVIDIA DOCA In-Silicon Security
NVIDIA boosts agentic AI infrastructure with DOCA in-silicon security
Press
インテルがロボット開発の課題を解決、オープンなロボティクスライブラリで
Intel adds open-source OpenVINO Physical AI Framework to robotics SDK
Official
NVIDIA DSX OS Delivers Open, Modular Software for Operating AI Factories at Scale
NVIDIA DSX OS delivers open, modular software to run AI factories at scale