Education & Research Software & Algorithm Provider Cloud & Data
NVIDIA Expands NVLink Fusion with Next-Generation NVHBM High-Bandwidth Memory Technology
Original from NvidiaNews: NVIDIA NVLink Fusion Expands With NVHBM Custom High-Bandwidth Memory

NVIDIA Expands NVLink Fusion with Next-Generation NVHBM High-Bandwidth Memory Technology

NVIDIA has expanded its NVLink Fusion technology by introducing NVIDIA NVHBM, a next-generation high-bandwidth memory designed to enhance AI infrastructure. This innovation aims to meet the increasing demands of AI agents and trillion-parameter workloads by integrating memory performance with compute, storage, and networking in a unified system.

The introduction of NVHBM is significant as it offers up to 30% greater memory bandwidth and 15% lower power consumption compared to traditional HBM architectures. By integrating the memory controller into the HBM base die, NVIDIA frees up to 25% more area on the XPU compute die, allowing for more efficient designs. This development is expected to accelerate the deployment of custom AI chips in collaboration with leading memory partners.

Amazon's Annapurna Labs will be the first to implement NVHBM technology, enhancing performance for AI workloads with its next-generation Trainium chips. This collaboration signifies a strategic partnership that aims to optimize AWS infrastructure designs. No further timeline was disclosed at the time of publication.

Editor's Note

The integration of high-bandwidth memory technologies like NVHBM is crucial for the evolving landscape of AI infrastructure. As companies seek to enhance performance and efficiency, the collaboration between NVIDIA and Amazon's Annapurna Labs highlights the importance of strategic partnerships in driving innovation. This move could reshape how AI workloads are managed and executed in hyperscale environments.

RobotToday Initiative

Robotics needs a service framework.

RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.

Share

Related Articles/News

BeyondMimic Advances from Motion Tracking to Versatile Humanoid Control Using Guided Diffusion

BeyondMimic Advances from Motion Tracking to Versatile Humanoid Control Using Guided Diffusion

BeyondMimic has transitioned from a focus on motion tracking to developing versatile humanoid control through guided diffusion techniques. This advancement signifies a major leap in robotics, allowing for more sophisticated and adaptable humanoid robots that can respond to complex environments and tasks. The importance of this development lies in its pote...

Research Article
NVIDIA Reports Strong Financial Results for Second Quarter of Fiscal 2027

NVIDIA Reports Strong Financial Results for Second Quarter of Fiscal 2027

NVIDIA (NASDAQ: NVDA) announced its financial results for the second quarter ending July 26, 2026, reporting a revenue of $96.2 billion. This figure represents an 18% increase from the previous quarter and a remarkable 106% rise compared to the same period last year. The company's gross margins stood at 75.0%, with GAAP earnings per diluted share at $2.46...

NVIDIA Unveils Jetson Orin Nano 2, Transforming Entry-Level Edge AI Solutions

NVIDIA Unveils Jetson Orin Nano 2, Transforming Entry-Level Edge AI Solutions

NVIDIA has launched the Jetson Orin Nano™ 2, a new robotic computer aimed at redefining entry-level edge AI, making cutting-edge generative AI performance accessible to millions of developers. As AI models become smaller and more efficient, edge devices are evolving into autonomous systems capable of understanding context, parsing language and images, and...

Edge AI Robotics AI Performance
AWS and NVIDIA Expand Collaboration with 2 Million Additional GPUs for AI Infrastructure

AWS and NVIDIA Expand Collaboration with 2 Million Additional GPUs for AI Infrastructure

Amazon Web Services (AWS) and NVIDIA have announced a significant expansion of their partnership to address the increasing global demand for AI infrastructure. The companies plan to deploy an additional 2 million NVIDIA GPUs across AWS’s global infrastructure, enhancing their collaboration in AI factories, CPUs, networking, and robotics. This move aims to...

NVIDIA Launches Jetson Orin Nano 2, Enhancing Edge AI for Robotics with Improved Performance

NVIDIA Launches Jetson Orin Nano 2, Enhancing Edge AI for Robotics with Improved Performance

NVIDIA Corp has unveiled the Jetson Orin Nano 2, an entry-level computer designed for edge AI applications. This new system offers doubled inference performance compared to its predecessor while consuming 40% less power in 15-watt mode. With 78 trillion operations per second of AI compute and 8GB of memory, it is tailored for real-time reasoning in smart ...

Artificial Intelligence Artificial Intelligence / Cognition Development Tools / SDKs / Libraries
Aetina Enhances Edge AI with Support for NVIDIA Jetson Orin Nano 2 Platform

Aetina Enhances Edge AI with Support for NVIDIA Jetson Orin Nano 2 Platform

Aetina Corporation has announced its support for the NVIDIA Jetson Orin Nano 2, enhancing its edge AI capabilities. This new platform allows for higher-performance generative AI and multimodal inference, which are crucial for compact Physical AI systems. The introduction of the Jetson Orin Nano 2 marks a significant advancement in entry-level edge AI robo...

NVIDIA Unveils RTX Spark with Major Game Publishers at Gamescom 2023

NVIDIA Unveils RTX Spark with Major Game Publishers at Gamescom 2023

NVIDIA is showcasing RTX Spark at Gamescom in Cologne, Germany, highlighting support from major game publishers like Electronic Arts, Embark, and Ubisoft. These developers are set to bring their blockbuster titles to the RTX Spark platform ahead of its fall launch, enhancing gaming experiences with advanced visual quality and anti-cheat technologies. The ...

Amazon's Valuation Could Surpass Tesla and SpaceX Combined by 2030

Amazon's Valuation Could Surpass Tesla and SpaceX Combined by 2030

Amazon, currently valued at $2.8 trillion, is projected to exceed the combined market cap of Tesla and SpaceX, which stands at $3.3 trillion. Analysts anticipate that by 2030, Amazon's growth, driven by its profitable AWS segment, will lead to a valuation of $215 billion in net income on $1.38 trillion in sales, outpacing the expected earnings of Tesla an...

NVIDIA's NVLink Fusion Enhances AI Factory Performance with Custom XPUs

NVIDIA's NVLink Fusion Enhances AI Factory Performance with Custom XPUs

NVIDIA's NVLink Fusion is designed to optimize AI factories by integrating custom XPUs with advanced infrastructure. This approach addresses the complexities of AI platform development, enabling faster market entry and improved performance metrics such as tokens per second and cost per token. The significance of this innovation lies in its ability to stre...

ROBOTTODAY WEEKLY August 17 – 21, 2026

ROBOTTODAY WEEKLY August 17 – 21, 2026

This week in robotics: Unitree surges 629% in its Shanghai STAR Market debut at a $66 billion valuation, Nevada approves 7,000 robotaxis, Amazon Prime Air expands to nearly 500 US cities, and Beijing's World Robot Conference marks the industry's pivot from spectacle to scale. August 17–21, 2026.

RobotToday Weekly
Amazon Robotics' Bhavana Chandrashekhar to Join Women in Robotics Lunch at RoboBusiness 2026

Amazon Robotics' Bhavana Chandrashekhar to Join Women in Robotics Lunch at RoboBusiness 2026

Bhavana Chandrashekhar, senior manager of applied science at Amazon Robotics, will participate in the Women in Robotics Lunch at RoboBusiness 2026, scheduled for October 20-21 in Santa Clara, California. This event aims to foster connections among women in the robotics industry, which currently sees only 35% of the global STEM workforce and 22% of the AI ...

Events Logistics News
Amazon to Expand Prime Air Drone Delivery to Nearly 500 U.S. Cities by 2026

Amazon to Expand Prime Air Drone Delivery to Nearly 500 U.S. Cities by 2026

Amazon.com Inc. announced plans to expand its Prime Air drone delivery service to nearly 500 cities and towns across the U.S. by the end of 2026, marking a sixfold increase from its current service area. This expansion aims to provide tens of millions of customers with access to drone delivery for groceries, electronics, cosmetics, medications, and househ...

Drones News Retail / Wholesale

Related Suppliers

NVIDIA Robotics

NVIDIA robotics division delivering Isaac simulation and training tools, Jetson edge compute modules, and Omniverse digital twin frameworks for physical AI.

Core Component Supplier Software & Algorithm Provider

Deci AI (Acquired by NVIDIA)

Israeli deep learning optimization startup acquired by NVIDIA in May 2024 for ~$300M; known for AutoNAC neural architecture search technology.

Amazon Robotics

Amazon's robotics division, formerly Kiva Systems (founded 2003, acquired 2012), in North Reading, MA, developing AMRs and automation for Amazon fulfilment.

Nscale

London-based Nscale runs a full-stack AI/GPU cloud platform and on Aug 1, 2026 agreed to acquire Anyscale for $1.65 billion.

ZML

Paris-based open-source project shipping a hardware-agnostic AI inference compiler stack that runs models on NVIDIA, AMD, TPU, and Trainium chips.

ChangXin Memory Technologies 长鑫存储技术有限公司

China's largest DRAM integrated device manufacturer, designing and producing DDR4/DDR5 and LPDDR memory chips from fabs in Hefei.

Cervoz Technology Co., Ltd.

Taiwanese supplier of industrial-grade SSDs, DRAM modules, and embedded peripherals engineered for AMR/AGV, machine vision, and edge robotics applications.

Fireworks AI

Redwood City AI infra startup (2022) serving open-source LLMs like Kimi K3 via API, used by Cursor, Vercel and Notion.

Kinetix AI 超维动力

Shenzhen embodied-AI startup building ultra-human-like humanoid robots powered by a physical world model for whole-body control.

SK hynix Inc. SK海力士

South Korean memory giant (KRX: 000660) and world's primary HBM3E supplier powering NVIDIA H100/H200 and other AI accelerators for robotics and AI compute.

Microway, Inc.

US manufacturer of HPC and AI computing systems — GPU clusters, workstations, and servers — serving scientific and engineering users since 1982.

Suxian Micro 速显微

Suxian Micro is a Tianjin-based fabless chipmaker known for its Tiandu-series display GPUs and an August 2026 TH1100 GPGPU aimed at embodied-AI robotics.

inJoin the RobotToday community on LinkedIn

Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.