Nvidia Corporation Stock Price, News & Analysis

NVIDIA (NVDA) and SK Group are partnering to build an AI factory in Korea featuring more than 50,000 NVIDIA GPUs, with phase one targeted for completion by late 2027. The facility will serve SK subsidiaries and external customers via a GPU-as-a-service model.

SK hynix will use NVIDIA CUDA-X, PhysicsNeMo, Omniverse libraries and RTX PRO servers to accelerate chip design, create fab digital twins and progress toward robotic self-optimizing fabs. SK Telecom will deploy an industrial cloud with over 2,000 RTX PRO 6000 Blackwell GPUs. SK aims to deploy AI agents for more than 40,000 employees.

NVIDIA (NVDA) and Hyundai Motor Group announced a deepened collaboration on Oct. 31, 2025 to build an NVIDIA Blackwell-powered AI factory in Korea that integrates in-vehicle AI, autonomous driving, smart factories and robotics.

Key commitments include deploying 50,000 NVIDIA Blackwell GPUs, an approximately $3 billion investment, establishment of an NVIDIA AI Technology Center and Hyundai Physical AI Application Center, use of Omniverse/Cosmos digital twins and RTX PRO servers, development of proprietary LLMs with Nemotron and NeMo, and vehicle compute using DRIVE AGX Thor on safety-certified DriveOS.

NVIDIA (NVDA) announced partnerships with the South Korea government, Samsung, SK Group, Hyundai Motor Group, NAVER Cloud and others to build national AI infrastructure deploying over 260,000 NVIDIA GPUs for sovereign clouds, AI factories, physical AI and LLM development.

Key facts: MSIT plans up to 50,000 GPUs for sovereign AI; initial 13,000 GPUs deployed by NAVER Cloud, NHN Cloud and Kakao; NAVER Cloud plans >60,000 GPUs; Hyundai collaboration includes an approximately $3 billion investment. Projects cover AI factories, digital twins, physical AI, sovereign LLMs, quantum-GPU research and startup support.

NVIDIA (NVDA) will host a webcast conference call on Wednesday, November 19, 2025 at 2:00 p.m. PT (5:00 p.m. ET) to discuss third-quarter fiscal 2026 results for the period ended October 26, 2025. The event will be available in listen-only mode at investor.nvidia.com and will include prepared remarks followed by a Q&A limited to financial analysts and institutional investors.

Written commentary from Colette Kress, executive vice president and chief financial officer, will be posted to investor.nvidia.com immediately after results are publicly announced at approximately 1:20 p.m. PT. The webcast will be recorded and replayed until the company’s next quarterly results call.

NVIDIA (NVDA) and Oracle will build the U.S. Department of Energy’s largest AI supercomputer at Argonne National Laboratory.

The Solstice system will include 100,000 NVIDIA Blackwell GPUs, an Equinox system will include 10,000 Blackwell GPUs (Equinox expected in H1 2026), and both systems together will deliver 2,200 exaflops of AI performance. Systems will run NVIDIA Megatron-Core and TensorRT stacks to support training and inference for agentic AI scientific workflows and connect to DOE experimental facilities.

The build uses a DOE public-private partnership model that includes industry investments and shared use cases.

NVIDIA (NVDA) and U.S. telecom partners unveiled an AI-native wireless stack for 6G on Oct. 28, 2025, built on the NVIDIA AI Aerial platform to enable integrated sensing, spectrum agility and edge AI across hardware, software and architecture.

The coalition—Booz Allen, Cisco, MITRE, ODC and T-Mobile—completed a working stack in six months, performed the first user-to-user phone call on the system, and demonstrated multimodal ISAC sensing and AI-managed spectrum functions designed for public safety, industrial monitoring and secure networks.

NVIDIA (NVDA) announced a partnership with Uber to scale a global Level 4 autonomous ride-hailing network using the new DRIVE AGX Hyperion 10 platform and DRIVE AV software, targeting 100,000 vehicles starting in 2027. Hyperion 10 is a production reference compute and sensor architecture featuring two DRIVE AGX Thor SoCs (>2,000 FP4 teraflops / 1,000 TOPS INT8), a 14-camera multimodal sensor suite plus lidar and radar, and OTA upgradability. NVIDIA will also supply a 1,700-hour, 25-country multimodal AV dataset and launch the Halos Certified Program and Halos AI System Inspection Lab (ANSI-accredited) for physical AI safety and cybersecurity.

NVIDIA (NVDA) announced expanded Omniverse “Mega” Blueprint capabilities on Oct. 28, 2025 to support factory‑scale digital twins, robotics simulation and a three‑computer architecture for collaborative robots.

Key highlights: Siemens will be first to support the industrial Mega Blueprint in its Xcelerator platform; FANUC and Foxconn Fii have begun providing OpenUSD 3D robot models; Foxconn is using Omniverse to design a 242,287‑square‑foot Houston facility; the release cites $1.2 trillion of 2025 U.S. production investments and multiple manufacturers (Belden, Caterpillar, Lucid, Toyota, TSMC, Wistron) using Omniverse for digital twins and automation.

Palantir and NVIDIA (NVDA) announced a collaboration to integrate NVIDIA GPU-accelerated computing, CUDA-X libraries, Nemotron and NeMo models into Palantir Ontology at the core of Palantir AI Platform (AIP).

The integrated stack aims to enable operational AI for enterprises and governments, supporting domain-specific automations, AI agents, real-time decision-making and accelerated supply-chain optimization. Lowe’s is named as an early adopter building a digital replica of its global supply chain. The companies also plan support for NVIDIA Blackwell architecture and NVIDIA AI Factory deployments.