Pull to refresh
Logo
SK hynix ships first 12-layer HBM4E samples for AI systems

SK hynix ships first 12-layer HBM4E samples for AI systems

New Capabilities

The memory that feeds AI chips is now the bottleneck, and SK hynix is racing Samsung to qualify the next generation

June 18th, 2026: SK hynix ships HBM4E samples

Overview

Updated Jun 18

Modern AI chips can crunch numbers faster than memory can feed them data. On June 18, 2026, SK hynix said it shipped samples of HBM4E, a stacked memory chip built to narrow that gap, to major customers.

The 12-layer part holds 48 gigabytes and moves data at up to 16 gigabits per second per pin. It also draws more than 20% less power than the prior generation. Memory, not just processors, now sets the ceiling on how large an AI data center can grow.

Why it matters

The memory feeding AI processors is now the limit on data-center size. Whoever ships it fastest sets how quickly AI gets cheaper to run.

Questions about this story

Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.

No questions yet — be the first to ask.

Key Indicators

58%
SK hynix HBM revenue share
Its slice of the high-bandwidth memory market in the first quarter of 2026.
48 GB
Capacity per stack
One 12-layer HBM4E chip, up from 36 gigabytes on HBM4.
16 Gb/s
Speed per pin
Peak data rate for each of the chip's input-output pins.
20%+
Power efficiency gain
Reduction in energy use versus the prior HBM4 generation.
~70%
SK hynix share of Nvidia Rubin HBM4
Its reported allocation of memory for Nvidia's next AI platform.

Voices

Curated perspectives — historical figures and your fellow readers.

Ever wondered what historical figures would say about today's headlines?

Sign up to generate historical perspectives on this story.

People Involved

Organizations Involved

Timeline

September 2025 June 2026

5 events Latest: June 18th, 2026 · 3 months ago
Tap a bar to jump to that date
  1. SK hynix ships HBM4E samples

    Latest Qualification

    SK hynix ships 12-layer, 48-gigabyte HBM4E samples running at up to 16 gigabits per second per pin. Its shares hit a record high.

  2. Samsung samples HBM4E first

    Qualification

    Samsung ships the industry's first 12-layer HBM4E samples to global customers, claiming a lead over SK hynix.

  3. HBM4 mass production starts

    Production

    SK hynix begins mass-producing HBM4 for Nvidia's Rubin AI chips, with a reported share near 70% of the order.

  4. Paid HBM4 samples reach Nvidia

    Qualification

    SK hynix and Samsung deliver paid HBM4 samples to Nvidia ahead of supply contracts for the Rubin platform.

  5. SK hynix finishes HBM4 first

    Milestone

    SK hynix says it is the first company to complete development of HBM4 and ready it for mass production.

Scenarios

1

SK hynix wins the majority of Nvidia's next HBM4E order

Likely Resolves by Q2 2027

Discussed by: TrendForce, The Korea Herald

SK hynix passes Nvidia's qualification first and keeps the lead it built on HBM4. Nvidia names it the primary HBM4E supplier for the 2027 Rubin Ultra platform, with more than half the allocation. This is the path most analysts expect if SK hynix's yields hold, given its track record of qualifying each generation ahead of rivals.

2

Samsung's earlier sample turns into the first HBM4E qualification win

Possible Resolves by End of 2026

Discussed by: The Korea Herald, TechTimes

Samsung shipped HBM4E samples three weeks before SK hynix. If it also clears a major customer's qualification first, it could take back share it lost during the HBM3E generation. The trigger would be a public confirmation that a leading AI chipmaker has approved Samsung's HBM4E for production before SK hynix earns the same approval.

3

HBM4E mass production slips into late 2027 on yield problems

Possible Resolves by End of 2026

Discussed by: TrendForce

Stacking 12 layers at higher speeds is hard, and TrendForce has reported yield and capacity limits on HBM4. If HBM4E hits similar trouble, SK hynix may not start mass production on its planned schedule. This resolves YES if SK hynix has not announced the start of HBM4E mass production by the end of 2026.

Historical Context

2 moments from history that rhyme with this story — and how they unfolded.

2017–2019

The memory supercycle and crash (2017–2019)

A shortage of DRAM and NAND sent memory prices soaring, and SK hynix and Samsung posted record profits. Then demand cooled and new supply arrived. Prices fell hard and earnings collapsed within about a year.

Then

Memory makers booked huge profits in 2017 and 2018, then deep declines in 2019.

Now

It confirmed how cyclical memory is: record highs in chip stocks have repeatedly given way to sharp downturns.

Why this matters now

SK hynix shares hit a record on the HBM4E news. The supercycle is a reminder that memory demand and pricing can reverse quickly, even for parts in short supply today.

2023–2024

The HBM3E qualification race (2023–2024)

As the AI boom took off, SK hynix qualified its HBM3E memory with Nvidia first and became its main supplier. Samsung, the larger memory maker overall, spent months trying to pass Nvidia's tests and fell behind in the most profitable memory segment.

Then

SK hynix captured most of Nvidia's HBM orders and saw profits surge. Samsung's delays cost it share and management changes followed.

Now

SK hynix entered the HBM4 era as the market leader, with roughly 58% share by early 2026.

Why this matters now

It shows why passing qualification matters more than shipping the first sample. Samsung sampled HBM4E first this time, but the HBM3E race was decided in the test lab, not at the sample stage.

Sources

(8)