Sandisk and SK hynix Unveil First Open High Bandwidth Flash (HBF) Standard at FMS 2026

By Ashwini Kuldeep on Aug 06, 2026 07:35 AM

At the Future of Memory and Storage (FMS) 2026 conference in Santa Clara, California, Sandisk Corporation and SK hynix Inc. jointly introduced the first official High Bandwidth Flash (HBF) standard specifications.

Released through the Open Compute Project (OCP), the new specification is designed to establish HBF as an open industry standard that addresses the growing memory bandwidth and capacity demands of artificial intelligence (AI) inference workloads.

The announcement comes six months after the HBF technology workstream was established under OCP in February 2026, following the initial standardization collaboration between Sandisk and SK hynix announced in August 2025. Throughout the development process, leading technology companies such as Google and Tenstorrent joined the initiative, contributing to technology validation and the creation of the open standard.

HBF: Bridging the Gap Between HBM and SSDs

High Bandwidth Flash is designed to sit between High Bandwidth Memory (HBM) and Solid State Drives (SSDs), providing a new memory tier that combines high-speed data transfer with the high-density storage capabilities of NAND flash. Positioned close to compute processors, HBF aims to lower total cost of ownership (TCO) while supporting large language models (LLMs) and next-generation AI workloads.

Storage Capacity: Supports capacities of up to 512GB using 8-high and 16-high NAND die stack configurations.
Bandwidth: Available in three performance grades (Grade 1 to Grade 3), delivering scalable throughput from 0.4TB/s up to 3.0TB/s.
Host Interconnect: Utilizes Universal Chiplet Interconnect Express (UCIe) for compatibility with GPUs, CPUs, and specialized xPU accelerators.
Technical Standards: Defines interface specifications, electrical characteristics, packaging requirements, reliability standards, and software guidelines for read/write I/O operations.

By making the specification publicly available through OCP, Sandisk and SK hynix aim to accelerate ecosystem development, improve interoperability, and allow HBF to complement existing HBM-based memory architectures.

FMS 2026 Keynotes and Technical Sessions

Both companies are showcasing HBF technology and tiered memory architectures through keynote presentations and panel discussions during FMS 2026.

Keynote Presentations

SK hynix Keynote (August 4): Executive Vice President Kim Chun-sung and Vice President Kang Uk-song presented "Orchestrating Efficient AI Infrastructure through Tiered Memory in the Era of Agentic AI", highlighting multi-tier memory strategies for meeting rapidly growing AI data requirements.
Sandisk Keynote (August 5, 11:40 a.m. PT): Chief Revenue Officer Jim Elliott, Chief Product Officer Khurram Ismail, and Chief Technology Officer Alper Ilkbahar presented "NAND – The Versatile & Scalable Foundation of the AI Era", focusing on system-level NAND optimizations for AI inference workloads.

Joint Panel Discussion (August 6, 9:45 a.m. PT)

A technical panel titled "Breaking the Memory Wall with High Bandwidth Flash", moderated by Thomas Coughlin, President of Coughlin Associates, features:

Rajeev Nagabhirava – Vice President, Sandisk
Lim Eui-cheol – Vice President, SK hynix
Xiaoyu Ma – Senior Staff Engineer, Google DeepMind

The discussion explores HBF architecture, integration strategies, performance and cost advantages, implementation challenges, and the roadmap for industry-wide standardization.

SK hynix Unveils 375-Layer V10 4D NAND

In addition to the HBF announcement, SK hynix showcased its tenth-generation (V10) 375-layer 4D NAND wafer and product prototypes for the first time.

The company says the new 375-layer 4D NAND delivers approximately 2.5x higher performance-per-watt compared to the previous generation, making it well suited for energy-efficient data center deployments.

SK hynix plans to begin mass production of enterprise SSDs (eSSDs) based on the 375-layer process early next year. At FMS 2026, the company is also demonstrating a broader portfolio of memory solutions designed for mobile devices, automotive systems, PCs, robotics, and AI infrastructure.