NVIDIA and SK hynix announced a multiyear technology partnership on June 7, 2026, to advance next-generation memory for AI factories and support NVIDIA’s AI infrastructure roadmap. The companies said the agreement will expand advanced memory supply, support semiconductor simulation and help develop autonomous manufacturing systems.
The NVIDIA announcement places advanced memory supply at the centre of global AI infrastructure expansion. The partnership links SK hynix memory codevelopment with NVIDIA platforms used in AI supercomputing, personal AI systems and robotic computing, while also applying NVIDIA tools to semiconductor design and factory operations.
NVIDIA And SK hynix Set Multiyear Partnership
NVIDIA and SK hynix said the agreement is a multiyear technology partnership focused on next-generation memory for the global AI factory buildout. The announcement was issued by NVIDIA on June 7, 2026, with SK hynix named as the memory partner.
NVIDIA said the partnership is intended to support long-term memory supply for AI factories and future computing platforms. The companies said codevelopment efforts will focus on aligning advanced memory technologies with NVIDIA’s AI infrastructure roadmap.
Agreement Supports AI Factory Memory Supply
The NVIDIA release said the agreement supports supply for advanced memory because development cycles, fabrication requirements and capital investment are extended. Additionally, NVIDIA said the partnership is intended to help memory supply keep pace with its AI infrastructure roadmap.
| Indicator | Recent Movement | Context |
|---|---|---|
| Partnership term | Multiyear agreement announced | NVIDIA said the agreement with SK hynix supports next-generation memory codevelopment for AI factories. |
| Memory supply | Expanded supply planning | NVIDIA said advanced memory supply must address extended development cycles, fabrication complexity and capital investment. |
| AI infrastructure | Roadmap alignment | NVIDIA said the partnership aligns SK hynix memory codevelopment with its AI infrastructure roadmap. |
| Market scope | New platform categories named | NVIDIA said SK hynix will work across AI infrastructure, personal AI and physical AI markets. |
Partnership Aligns With NVIDIA Infrastructure Roadmap
NVIDIA said SK hynix will diversify into markets created by NVIDIA across AI infrastructure, personal AI and physical AI. The named platforms include NVIDIA Vera Rubin AI supercomputers, NVIDIA Vera CPUs, NVIDIA RTX Spark-powered PCs and NVIDIA Jetson Thor robotic computing platforms.
According to NVIDIA, the partnership extends across data centre, personal AI and robotics platforms. The announcement links SK hynix memory development to multiple NVIDIA product categories rather than a single hardware line.
Target Platforms Include Vera Rubin And Jetson Thor
The NVIDIA release identified Vera Rubin AI supercomputers and Vera CPUs as infrastructure targets. Meanwhile, RTX Spark-powered PCs and Jetson Thor robotic computing platforms were named as personal AI and physical AI targets.
- AI infrastructure: NVIDIA named Vera Rubin AI supercomputers and Vera CPUs as systems covered by the memory codevelopment work.
- Personal AI: NVIDIA said RTX Spark-powered PCs are among the markets where SK hynix will codevelop memory.
- Physical AI: NVIDIA said Jetson Thor robotic computing platforms are included in the partnership scope.
- Supply context: Reuters reported that NVIDIA’s wider South Korea partnerships come amid sustained AI infrastructure demand and memory supply pressure.
Companies Expand Semiconductor Simulation Work
NVIDIA said the companies will apply AI to semiconductor chip design and manufacturing. Specifically, the release said SK hynix is using NVIDIA CUDA-X libraries and AI to accelerate semiconductor simulation, including technology computer-aided design and computational lithography workflows.
NVIDIA said the collaboration extends beyond memory development into semiconductor design and manufacturing. The announcement identified CUDA-X, PhysicsNeMo, TCAD workflows and in-house simulation codes as technologies being used to accelerate engineering and design processes.
AI Tools Support TCAD And Lithography Workflows
NVIDIA said SK hynix is also using CUDA-X and the NVIDIA PhysicsNeMo framework to accelerate in-house simulation codes and AI physics workflows. Additionally, NVIDIA said the work could support three-way collaboration among chipmakers, NVIDIA and electronic design automation software vendors.
NVIDIA described the work as part of ongoing development activity and did not publish independent performance results for the simulation tools referenced in the announcement.
SK hynix Advances Fab Digital Twins
NVIDIA said SK hynix is developing fab digital twins as a foundation for autonomous fab operations. The release said teams can use NVIDIA Omniverse libraries, OpenUSD pipelines and scene optimization technologies to build 3D factory scenes for visualizing, simulating and optimizing semiconductor manufacturing environments.
NVIDIA said SK hynix is using Omniverse, OpenUSD and scene optimization technologies to develop digital representations of semiconductor manufacturing facilities. The work is intended to support simulation, planning and operational analysis across fab environments.
Omniverse And cuOpt Support Factory Optimization
NVIDIA said the digital twins can support operational optimization, including movement of autonomous mobile robots and other fab assets. The release also named NVIDIA cuOpt, an open source GPU-accelerated decision optimization engine, and NVIDIA Metropolis as tools connected to this work.
However, NVIDIA framed this as development and exploration rather than a completed production result. The announcement said the companies are exploring ways to connect digital twins with legacy software and agentic AI workflows, enabling AI systems to reason over fab data, automate tasks and improve manufacturing decision-making.
Partnership Links Memory, Design And Manufacturing
NVIDIA and SK hynix said the agreement combines memory development, semiconductor design workflows and manufacturing technologies within a single partnership framework. The companies described advanced memory as a critical component of future AI infrastructure and said collaboration will extend from hardware supply to simulation and factory operations.
The announcement also highlighted ongoing work in semiconductor modelling and digital manufacturing, including AI-assisted simulation and development of factory digital twins.
Stakeholder Comments
Chey Tae-won, Chairman of SK Group, said SK hynix and NVIDIA had been building toward the partnership for years. He said the companies are codeveloping next-generation memory for AI factories and applying AI to semiconductor design and manufacturing.
The comments emphasised the partnership’s dual focus on memory development and semiconductor manufacturing, reflecting the broader scope outlined in NVIDIA’s announcement.
NVIDIA and SK hynix said the partnership combines next-generation memory development, semiconductor simulation and digital manufacturing technologies. The agreement supports NVIDIA’s AI infrastructure roadmap while expanding collaboration on memory supply, engineering workflows and factory operations.
The companies said the work will span AI computing platforms, semiconductor design processes and digital manufacturing systems as development continues under the multiyear agreement.
Sources: NVIDIA, Reuters, and SK hynix Newsroom.
Prepared by Ivan Alexander Golden, Founder of THX News, an independent news organization delivering timely insights from global official sources.
Research combines AI-assisted analysis with human-edited accuracy and context.
