# HBM4

> Source: https://aiwiki.ai/wiki/hbm4
> Updated: 2026-06-27
> Categories: AI Hardware
> From AI Wiki (https://aiwiki.ai), a free encyclopedia of artificial intelligence. Quote with attribution.

HBM4 (High Bandwidth Memory generation 4) is the fourth principal generation of [high bandwidth memory](/wiki/high_bandwidth_memory) (HBM), a stacked DRAM standard published by the JEDEC Solid State Technology Association as JESD270-4 on April 17, 2025.[^1] HBM4 doubles the per-stack interface from 1,024 to 2,048 data bits and delivers roughly 2 TB/s of bandwidth per stack at the JEDEC reference pin rate of 8 Gb/s, about double the per-stack bandwidth of HBM3e at the same signaling rate.[^1] It is the memory standard at the center of next-generation [AI chip](/wiki/ai_chip) design: production HBM4 stacks supply the [NVIDIA Vera Rubin](/wiki/nvidia_vera_rubin) platform and the [AMD MI400](/wiki/amd_mi400) family, and the three makers are [SK hynix](/wiki/sk_hynix), Samsung Electronics, and Micron.[^2][^3][^4]

The standard supports stack heights from 4-high through 16-high using 24-gigabit or 32-gigabit DRAM dies for cube capacities up to 64 GB, and it is the first HBM generation in which the base logic die is widely fabricated on an advanced foundry logic process at [TSMC](/wiki/tsmc) rather than purely on a memory vendor's in-house technology, opening room for customer-specific base dies and tighter integration with accelerator silicon.[^1][^5] JEDEC describes HBM4 as "designed to drive a leap forward in efficient, high performance computing for AI and other accelerated applications."[^1] By mid-2026 all three vendors had entered HBM4 mass production, and on June 1, 2026 NVIDIA named SK hynix, Samsung, and Micron as suppliers for the Vera Rubin platform as it entered full production.[^26]

## What is HBM4?

HBM4 is a 3D-stacked DRAM standard in which multiple DRAM dies are stacked vertically on top of a base logic die and connected to the host processor through through-silicon vias and a silicon interposer using a very wide parallel interface.[^1][^6] The defining change in HBM4 is the width of that interface: 2,048 data bits per stack, double the 1,024-bit interface that every prior HBM generation used.[^1] At the JEDEC reference signaling rate of 8 Gb/s, the 2,048-bit interface yields approximately 2.048 TB/s of bandwidth per stack.[^1] HBM4 also doubles the number of independent channels per stack from 16 (in HBM3) to 32, each split into two pseudo-channels, giving 64 pseudo-channels per stack for finer-grained access.[^1]

Mian Quddus, Chairman of the JEDEC Board of Directors, said at the standard's release that "JEDEC members are dedicated to developing the standards needed to support the technology of the future" and that "the HBM Subcommittee's efforts to continuously improve the HBM standard hold the potential to drive significant advancements in a wide variety of applications."[^1]

## Why does HBM4 matter for AI?

The pressure to widen the interface came from the rise of large language model training and inference, where the ratio of arithmetic throughput on accelerators such as NVIDIA's Hopper and Blackwell GPUs to memory bandwidth had been climbing faster than DRAM signaling could sustain.[^8] Industry analysts framed this as the "memory wall" or "bandwidth wall" problem for AI accelerators, noting that effective floating-point utilization on transformer workloads is often bounded by how fast model weights and KV caches can be read from HBM rather than by compute.[^8] Widening the interface from 1,024 to 2,048 bits is the principal lever HBM4 uses to address that gap, because doubling the wires per stack approximately doubles the achievable bandwidth at any given per-pin clock rate.[^1]

The wider interface is feasible because the same advanced packaging techniques that AI accelerators already use, in particular TSMC's CoWoS family of 2.5D interposers, have grown to support far more interconnect routing per stack than the original HBM-era interposers.[^9] HBM4 was designed alongside packaging vendors so that 2,048-bit stacks would fit within reticle-scale interposers used by NVIDIA, AMD, and Google.[^9]

## Background

High Bandwidth Memory was first standardized by JEDEC as JESD235 in late 2013 and was conceived as a way for graphics and high-performance compute parts to escape the bandwidth and power limits of conventional GDDR memory by stacking DRAM dies on top of a base logic die and connecting them to the host processor through a silicon interposer with a very wide interface.[^6] Successive revisions HBM2 (2016), HBM2E (2018-2020), HBM3 (2022), and HBM3e (2023-2024) widened capacity per stack and lifted per-pin signaling speeds while keeping the 1,024-bit per-stack interface intact.[^6][^7] HBM3 set the per-pin reference at 6.4 Gb/s with up to 819 GB/s per stack; HBM3e pushed signaling past 9 Gb/s in production parts and peaked above 1.2 TB/s per stack in 12-high configurations shipping in late 2024.[^7]

## When was HBM4 standardized?

JEDEC published JESD270-4 "High Bandwidth Memory (HBM4) DRAM" on April 17, 2025 through its JC-42 memory committee.[^1] The press release credited contributions from AMD, Cadence, Google, Meta, Micron, NVIDIA, Samsung, SK hynix, and Synopsys.[^1] JEDEC framed HBM4 as an evolutionary step beyond HBM3 that retains backwards compatibility at the controller interface so that a single memory controller can in principle drive either HBM3 or HBM4 stacks.[^1]

The standard fixes a reference per-pin signaling rate of 8 Gb/s, with implementations free to clock individual stacks higher within the electrical envelope.[^1] At 8 Gb/s the 2,048-bit interface yields approximately 2.048 TB/s of bandwidth per stack, roughly double the per-stack bandwidth of HBM3e at the same signaling rate.[^1] HBM4 doubles the number of independent channels per stack from 16 (in HBM3) to 32, each of which is further split into two pseudo-channels, giving 64 pseudo-channels per stack for finer-grained access.[^1]

JEDEC also defined a wider range of supply voltages than prior HBM generations. JESD270-4 lists four allowed VDDQ values (0.7 V, 0.75 V, 0.8 V, and 0.9 V) and two VDDC core voltages (1.0 V or 1.05 V), giving vendors latitude to trade power against headroom.[^1] The standard supports 4-high, 8-high, 12-high, and 16-high stack configurations using 24-gigabit or 32-gigabit dies, putting the top cube density at 64 GB (32-gigabit dies stacked 16-high).[^1] To accommodate taller stacks within existing thermal and mechanical budgets, JEDEC kept a 775-micrometer nominal package thickness for 12-high and 16-high stacks; TrendForce noted this relaxed the immediate pressure on memory makers to adopt hybrid bonding for HBM4.[^9]

Reliability features in JESD270-4 include Directed Refresh Management (DRFM), a mechanism intended to harden DRAM rows against row-hammer style disturbance attacks, plus broader RAS (reliability, availability, serviceability) provisions.[^1] These are aimed at hyperscale deployments where soft error rates aggregated over thousands of accelerators matter for system reliability.[^1]

## Specifications

The table below compares the principal JEDEC reference parameters for HBM3, HBM3e, and HBM4. HBM3e is not a separately numbered standard but an extension of HBM3 (JESD238) with higher signaling rates published in JEDEC's HBM3 update; production HBM3e parts from SK hynix, Micron, and Samsung commonly run at 9.2-9.6 Gb/s and 36 GB per 12-high stack.[^7][^10]

| Parameter | HBM3 | HBM3e | HBM4 |
|---|---|---|---|
| JEDEC document | JESD238 (2022) | JESD238 update (2023-2024) | JESD270-4 (April 2025)[^1] |
| Interface width per stack | 1,024 bits | 1,024 bits | 2,048 bits[^1] |
| Reference per-pin rate | 6.4 Gb/s | up to 9.2-9.6 Gb/s (vendor) | 8 Gb/s (JEDEC reference)[^1][^7] |
| Peak per-stack bandwidth at reference rate | ~819 GB/s | ~1.2 TB/s (12-high parts) | ~2.0 TB/s[^1][^7] |
| Channels per stack | 16 | 16 | 32 (64 pseudo-channels)[^1] |
| Supported stack heights | up to 12-high | up to 12-high | 4, 8, 12, 16-high[^1] |
| Per-die capacity | 16 Gb / 24 Gb | 24 Gb / 32 Gb | 24 Gb / 32 Gb[^1] |
| Top cube capacity | 24 GB (12-high, 16 Gb dies) | 36 GB (12-high, 24 Gb dies) | 64 GB (16-high, 32 Gb dies)[^1][^7] |
| VDDQ | 0.4 V (HBM3) | vendor extended | 0.7 / 0.75 / 0.8 / 0.9 V[^1] |
| VDDC core voltage | 1.1 V | 1.1 V | 1.0 V or 1.05 V[^1] |

Production HBM4 parts have already exceeded the JEDEC reference signaling rate. SK hynix, in its September 12, 2025 development announcement, said its first generation HBM4 product targets operating speeds above 10 Gb/s while improving power efficiency by more than 40 percent compared with its HBM3e parts.[^11] Micron's June 2025 sampling announcement quoted "over 11 Gb/s" pin speeds and bandwidth of greater than 2.8 TB/s per stack for the 36 GB 12-high product line.[^4] Samsung's qualification samples sent to NVIDIA and AMD in late 2025 ran at 11.7 Gb/s in customer testing, above NVIDIA's reported 10 Gb/s qualification floor.[^12][^13] NVIDIA's Rubin program ultimately pushed required pin speeds above 11 Gb/s (around 11.7 Gb/s), which forced all three vendors to revise their HBM4 designs to hold signal integrity at the higher data rate.[^26][^28]

## Who makes HBM4?

Three memory makers produce HBM4: SK hynix, Samsung Electronics, and Micron. SK hynix was first to ship samples and first to declare development complete; Samsung was the first vendor reported to pass NVIDIA's HBM4 quality certification; and Micron reached high-volume production of its 36 GB 12-high HBM4 in March 2026.[^3][^11][^16][^18][^26][^27]

### SK hynix

SK hynix has consistently been the first HBM vendor to ship each new generation and HBM4 is no exception. The company announced on March 18, 2025 that it had shipped the industry's first 12-high HBM4 samples to major customers, using its Advanced MR-MUF (Mass Reflow Molded Underfill) process and reaching 36 GB per cube with bandwidth above 2 TB/s, roughly 60 percent faster than its HBM3e.[^3] SK hynix said it expected to complete preparations for mass production within the second half of 2025.[^3]

On September 12, 2025, SK hynix announced that it had "completed development" of HBM4 and was ready to begin mass production.[^11] The company said it had "completed the world-first development of its high-performance AI memory solution HBM4 and finished preparations for its mass production."[^11] The product runs at over 10 Gb/s per pin, exceeding the JEDEC standard of 8 Gb/s, and improves power efficiency by more than 40 percent over its prior HBM3e generation.[^11] SK hynix also said HBM4 could improve AI service performance by up to 69 percent in customer testing.[^11] The September release describes manufacturing using SK hynix's 1bnm (fifth-generation 10 nm class) DRAM process combined with Advanced MR-MUF stacking.[^11]

TrendForce reported in mid-December 2025 that SK hynix was supplying around 20,000 to 30,000 final HBM4 samples to NVIDIA for final testing and validation, with HBM4 shipments beginning in Q4 2025 and a full ramp through 2026.[^14] In late January 2026 TrendForce additionally reported that SK hynix was expected to supply roughly two-thirds of NVIDIA's HBM4 demand in 2026.[^15]

By February 2026 SK hynix had moved into HBM4 mass production for the Rubin generation, having put its production system in place the previous September and cleared NVIDIA's validation on large volumes of paid samples.[^15][^26] Reporting at the time put SK hynix at roughly two-thirds (about 60 to 70 percent) of NVIDIA's 2026 HBM4 allocation, a share estimate attributed to industry analysts including UBS rather than to either company.[^15][^28] NVIDIA's Rubin design uses eight 12-high HBM4 stacks per GPU, and the push to data rates above 11 Gb/s required SK hynix to refine its 2.5D packaging and circuit design to keep stacks stable at the higher speed.[^26][^28]

### Samsung Electronics

Samsung's HBM4 effort started later than SK hynix's. The Korea Economic Daily reported on December 3, 2025 that Samsung had cleared a critical internal qualification step for its HBM4 chips and was moving toward mass production.[^16] Samsung delivered engineering samples to NVIDIA in late 2025 with final qualification targeted for early 2026.[^16][^12]

By late January 2026 TrendForce reported that Samsung had cleared NVIDIA's and AMD's final HBM4 qualification tests at 11.7 Gb/s per pin and was set to begin mass production and shipments in February 2026.[^13] Samsung's HBM4 is fabricated on its sixth-generation 1c DRAM process and pairs the memory dies with its own 4 nm logic base die rather than depending on TSMC for the base die.[^13] Samsung also said it planned to roughly double its overall HBM production capacity in 2026, with HBM4 taking a growing share.[^17]

By mid-2026 Samsung had become the first vendor to pass NVIDIA's HBM4 quality certification and had begun mass production and supply for the Rubin platform.[^26][^27] Analyst estimates put Samsung at roughly 25 to 30 percent of NVIDIA's 2026 HBM4 allocation, behind SK hynix.[^26][^28] Samsung also began shipping early samples of HBM4E, the seventh-generation extension aimed at the Vera Rubin Ultra platform expected in late 2027.[^26][^27]

### Micron

Micron sampled HBM4 to multiple major AI customers on June 10, 2025.[^4] The samples are 36 GB 12-high stacks built on Micron's 1-beta (1β) DRAM process, with a 2,048-bit interface delivering bandwidth above 2.0 TB/s per stack at the initial product clock, then increasing to over 2.8 TB/s at pin speeds above 11 Gb/s for the production-class part.[^4] Micron quotes more than 60 percent better performance and over 20 percent better power efficiency than its HBM3e.[^4]

On March 16, 2026 Micron announced that its HBM4 36 GB 12-high parts had entered high-volume production designed for NVIDIA's Vera Rubin platform, alongside Micron's PCIe Gen 6 data center SSD and SOCAMM2 module products for the same generation of NVIDIA systems.[^18] Micron also said it was sampling 48 GB 16-high HBM4 stacks to customers, increasing per-placement capacity by a third over the 12-high 36 GB part.[^18] When NVIDIA named its Rubin HBM4 suppliers in June 2026, Micron was the third named supplier alongside SK hynix and Samsung, taking the remaining share of allocation after the two Korean vendors.[^26]

### Custom base die and the TSMC partnership

A defining structural change in HBM4 is the move away from memory-vendor-owned base dies toward base dies fabricated on advanced foundry logic processes. SK hynix and TSMC announced a memorandum of understanding on April 19, 2024 under which TSMC would manufacture the HBM4 base die using TSMC's advanced logic process, with the intent of packing additional functionality (memory controller features, custom interfaces, in-package compute hooks) into the limited base-die area.[^5] The same agreement covered tighter integration between SK hynix HBM stacks and TSMC's CoWoS packaging.[^5]

TrendForce later reported that SK hynix was considering TSMC's 3 nm node for the HBM4E (the post-HBM4 extension expected in 2027) logic die and was likely to use TSMC's 12 nm process for mainstream HBM4 base dies while reserving 3 nm for premium designs targeting NVIDIA's flagship GPUs and Google's TPUs.[^19] Samsung took a different approach and used its own 4 nm logic process for its HBM4 base die, preserving more vertical integration.[^13]

The custom base-die concept has become central to vendor differentiation. The Next Platform, SemiAnalysis, and other analysts describe HBM4 base dies as a new arena for "custom HBM," where hyperscale customers can request specific logic, I/O, or in-memory-compute features fused into the bottom of the stack and then 2.5D-integrated with their accelerator.[^20]

## Which AI accelerators use HBM4?

HBM4 is the memory of choice for the 2026-2027 generation of AI accelerators. NVIDIA's Vera Rubin platform and AMD's Instinct MI400 family are the two highest-profile adopters, and Samsung's HBM4 has additionally passed Google qualification for upcoming TPU derivatives.[^21][^22][^24]

### NVIDIA Vera Rubin

NVIDIA's Vera Rubin platform, announced as the successor to Blackwell at GTC 2025 in March 2025 with full production targeted for 2026, is the highest-profile design win for HBM4.[^21] The Rubin GPU as presented at GTC 2025 carries 288 GB of HBM4 with 13 TB/s of aggregate memory bandwidth on the NVL144 platform, up from 8 TB/s on the prior Blackwell Ultra generation.[^21] Rubin Ultra, scheduled for the second half of 2027, uses HBM4e (the post-JEDEC vendor extension) and carries 1 TB of memory across four reticle-sized GPUs per package with roughly 100 PetaFLOPS of FP4 compute.[^21]

At GTC 2026 in March 2026, Micron, Samsung, and SK hynix all positioned HBM4 supply around the Vera Rubin ramp.[^18][^13][^11] Micron explicitly named NVIDIA Vera Rubin as the target platform for its volume HBM4 36 GB 12-high product.[^18]

On June 1, 2026, at the GTC keynote held during Computex in Taipei, NVIDIA CEO Jensen Huang confirmed that the Vera Rubin platform was in full production and named SK hynix, Samsung, and Micron as its HBM4 suppliers.[^26][^27] NVIDIA did not publish official per-vendor allocation figures; analyst estimates circulating around the announcement placed SK hynix at roughly 60 to 70 percent of Rubin HBM4, Samsung at about 25 to 30 percent, and Micron supplying the remainder.[^26][^28] Huang said NVIDIA and its manufacturing partners, including Foxconn, Quanta, and Wistron, were building Vera Rubin NVL72 rack-scale systems at scale, and that rack assembly time had fallen from roughly two hours per rack to about five minutes.[^26]

### AMD Instinct MI400

AMD disclosed the [AMD MI400](/wiki/amd_mi400) series at its 2025 Financial Analyst Day and provided more detail at CES 2026.[^22][^23] The MI400 family is built on AMD's CDNA 5 architecture and uses HBM4 to reach roughly 432 GB of memory per accelerator with 19.6 TB/s of bandwidth, up from 288 GB and 8 TB/s on the prior MI355X generation.[^22] AMD also said the MI400 family adopts CoWoS-L packaging in place of the CoWoS-S used in earlier Instinct parts to handle the larger 2,048-bit interface and the increased die count.[^22] The MI455X variant targets large-scale AI training, the MI430X targets HPC and government workloads, and the MI440X is positioned as a rack-mounted server combining eight MI400 GPUs with an [AMD EPYC Venice](/wiki/amd_epyc_venice) CPU.[^23]

### Google TPU and others

Reporting from late 2025 and early 2026 indicates that Samsung's HBM4 has passed Google qualification and will appear in Google's seventh-generation TPU lineage, branded TPU Ironwood and its successors.[^24] Google's Ironwood announcement at Cloud Next described 192 GB of HBM per chip and 7.4 TB/s of memory bandwidth, with public materials referring to HBM3e for the launch generation and HBM4 for upcoming derivatives.[^24] [TPU Ironwood](/wiki/tpu_ironwood) therefore sits at the transition point in Google's TPU memory roadmap from HBM3e to HBM4.[^24]

## Manufacturing and supply

HBM4 production rests on three tightly coupled manufacturing technologies: the DRAM die process at each memory vendor, the base-die logic process at TSMC or the memory vendor's own foundry, and 2.5D advanced packaging using TSMC's CoWoS family.[^5][^9] In 2026 TSMC is the dominant supplier of CoWoS packaging for HBM4-based accelerators, with capacity expansion plans aimed at supporting the Vera Rubin and MI400 ramps.[^9]

CoWoS-L, the variant that uses smaller silicon bridges instead of a monolithic silicon interposer, is the version most associated with HBM4-class designs because it allows package sizes approaching six times the conventional reticle limit, which is needed to host the larger GPU dies plus eight or more HBM4 stacks.[^9][^22] NVIDIA's Rubin platform and AMD's MI400 are both reported to use CoWoS-L; AMD has stated this directly.[^22]

On the memory side, TrendForce reported on December 30, 2025 that Samsung planned to raise its overall HBM capacity by roughly 50 percent in 2026, with much of the new capacity dedicated to HBM4.[^17] SK hynix similarly raised guidance for HBM capacity in 2026 and 2027.[^14][^15] Industry coverage at the start of 2026 framed HBM4 supply as the principal bottleneck for AI accelerator shipments through the rest of the year.[^17][^25]

A January 2026 report from i-Connect007 noted that broad HBM4 mass production for some customers had been delayed to the end of Q1 2026 due to spec upgrades requested by NVIDIA and product strategy changes, although Samsung was on track for a February 2026 production start with NVIDIA and AMD.[^25][^13]

### 2026 production ramp

By the first half of 2026 all three HBM4 vendors were in mass production for the Rubin generation. SK hynix entered HBM4 mass production around February 2026, Samsung began mass production and supply after becoming the first vendor to pass NVIDIA's HBM4 quality certification, and Micron reached high-volume production of its 36 GB 12-high HBM4 in March 2026.[^15][^18][^26][^27] NVIDIA's June 1, 2026 confirmation that Vera Rubin had entered full production, with all three vendors named as suppliers, marked the point at which HBM4 became a volume production part rather than a sampling-and-qualification effort.[^26]

The supply picture stayed concentrated. SK hynix held the largest share of NVIDIA's 2026 HBM4 allocation at an estimated 60 to 70 percent, Samsung an estimated 25 to 30 percent, and Micron the remainder, figures attributed to industry analysts rather than to NVIDIA.[^26][^28] These allocation estimates should be read as analyst projections, not official splits, since NVIDIA has not published per-vendor numbers.[^26] Attention also began shifting to the next extension: Samsung shipped early HBM4E samples ahead of the Vera Rubin Ultra generation due in late 2027, signaling that the HBM4 ramp and the start of HBM4E sampling overlapped through 2026.[^26][^27]

## How is HBM4 different from HBM3e?

HBM4 is best understood as a step change in interface width relative to HBM3e, paired with a comparatively modest increase in per-pin signaling. HBM3e in production typically runs at 9.2-9.6 Gb/s on a 1,024-bit per-stack interface to give roughly 1.2 TB/s per stack and 36 GB per 12-high cube using 24-gigabit dies.[^7] HBM4 at the JEDEC reference of 8 Gb/s on 2,048 bits gives about 2.0 TB/s per stack and 48-64 GB per cube depending on stack height and die density.[^1] Vendors clocking HBM4 above 10-11 Gb/s in production lift per-stack bandwidth to 2.5-2.8 TB/s.[^4][^11][^13]

Capacity per accelerator therefore scales roughly twofold from HBM3e to HBM4 at constant stack count: an eight-stack accelerator goes from around 288 GB (HBM3e, 12-high, 36 GB stacks) to 432 GB (HBM4, 12-high, 54 GB stacks at 32 Gb dies) or higher with 16-high stacks.[^22] The AMD MI400 series claims 432 GB and 19.6 TB/s, and NVIDIA's Rubin claims 288 GB and 13 TB/s using a different stack count and configuration.[^21][^22]

Power efficiency improves through a combination of the lower VDDQ voltages allowed by JEDEC (0.7-0.9 V on HBM4 versus 0.4 V VDDQ in the HBM3 reference but with significantly higher VDDQ in vendor HBM3e extensions) and the larger interface that reduces required signaling rate per pin to hit a given bandwidth.[^1] SK hynix and Micron each quote roughly 20-40 percent better power efficiency than their respective HBM3e products at iso-bandwidth.[^4][^11]

A second important difference is the base-die strategy. HBM3 and HBM3e base dies were fabricated by the memory vendor on a relatively mature DRAM-compatible process. HBM4 base dies are commonly fabricated on advanced foundry logic processes (TSMC 12 nm or 3 nm; Samsung 4 nm) to support more sophisticated controller logic, configurable I/O, and the option of in-stack compute.[^5][^13][^19] This shift makes HBM4 a more flexible building block for custom accelerator designs.

The table below summarizes the headline differences between the two generations at the system level.

| Difference | HBM3e | HBM4 |
|---|---|---|
| Interface width per stack | 1,024 bits | 2,048 bits[^1] |
| Per-stack bandwidth (production) | ~1.2 TB/s | 2.5-2.8 TB/s[^4][^7][^11] |
| Top cube capacity | 36 GB (12-high) | 64 GB (16-high)[^1][^7] |
| Base die | memory-vendor DRAM process | advanced foundry logic (TSMC 12/3 nm; Samsung 4 nm)[^5][^13][^19] |
| Lead accelerators | NVIDIA Blackwell, AMD MI350 | NVIDIA Vera Rubin, AMD MI400[^21][^22] |

## What are the limitations of HBM4?

HBM4 inherits the basic constraints of stacked HBM. Per-cube capacity, while higher than HBM3e, is still capped at 64 GB for the 16-high 32-gigabit configuration, far less than the hundreds of gigabytes of DDR5 a server CPU can attach.[^1] HBM stacks must be 2.5D-integrated next to the accelerator on a CoWoS-class interposer; TSMC's CoWoS capacity has historically been the principal bottleneck for AI accelerator volumes, and HBM4's larger package sizes have intensified pressure on CoWoS-L supply.[^9]

Thermal management is a growing concern as 12-high and 16-high stacks dissipate more power per square millimeter of stack footprint. JEDEC's decision to keep 775-micrometer package thickness for taller stacks reduced the urgency of moving to hybrid bonding for HBM4, but analysts expect HBM4E and HBM5 to require hybrid bonding to keep going.[^9]

Cost remains high. HBM accounts for a large fraction of accelerator bill-of-materials at the HBM3e generation already, and HBM4's wider interface, taller stacks, and advanced-node base die push that share higher.[^4][^11] Reports from late 2025 and early 2026 describe a multi-quarter shortage of HBM4 supply relative to demand from NVIDIA, AMD, and hyperscale TPU customers, with Samsung's late ramp and the dependency on TSMC for both base dies and CoWoS-L packaging as principal risk factors.[^14][^17][^25] The higher signaling NVIDIA demanded for Rubin, above 11 Gb/s rather than the JEDEC 8 Gb/s reference, also forced all three vendors to retool circuits and 2.5D packaging, which contributed to the qualification delays seen in late 2025 and early 2026.[^25][^26][^28]

Finally, HBM4 still relies on a 2.5D side-by-side topology rather than true 3D integration of memory directly on top of the compute die. This bounds total bandwidth by the cross-section of the interposer and by the I/O ring around the compute die, which is why HBM4's 2,048-bit interface required parallel advances in CoWoS-L and large reticle packaging.[^9] Future generations are widely expected to move portions of HBM stacks onto the compute die through 3D hybrid bonding, with HBM4E and HBM5 forming intermediate steps.[^19][^24]

## See also

- [high bandwidth memory](/wiki/high_bandwidth_memory)
- [tsmc](/wiki/tsmc)
- [nvidia vera rubin](/wiki/nvidia_vera_rubin)
- [amd mi400](/wiki/amd_mi400)
- [tpu ironwood](/wiki/tpu_ironwood)
- [nvidia blackwell](/wiki/nvidia_blackwell)
- [nvidia h200](/wiki/nvidia_h200)
- [amd instinct mi355x](/wiki/amd_instinct_mi355x)
- [ai chip](/wiki/ai_chip)

## References

[^1]: JEDEC Solid State Technology Association, "JEDEC and Industry Leaders Collaborate to Release JESD270-4 HBM4 Standard: Advancing Bandwidth, Efficiency, and Capacity for AI and HPC", JEDEC, 2025-04-17. https://www.jedec.org/news/pressreleases/jedec%C2%AE-and-industry-leaders-collaborate-release-jesd270-4-hbm4-standard-advancing. Accessed 2026-05-26.
[^2]: SK hynix, "SK hynix Completes World's First HBM4 Development and Readies Mass Production", PRNewswire, 2025-09-12. https://www.prnewswire.com/news-releases/sk-hynix-completes-worlds-first-hbm4-development-and-readies-mass-production-302554538.html. Accessed 2026-05-26.
[^3]: SK hynix, "SK hynix Ships World's First 12-Layer HBM4 Samples to Customers", PRNewswire, 2025-03-18. https://www.prnewswire.com/news-releases/sk-hynix-ships-worlds-first-12-layer-hbm4-samples-to-customers-302405246.html. Accessed 2026-05-26.
[^4]: Micron Technology, "Micron Ships HBM4 to Key Customers to Power Next-Gen AI Platforms", GlobeNewswire, 2025-06-10. https://www.globenewswire.com/news-release/2025/06/10/3096784/14450/en/Micron-Ships-HBM4-to-Key-Customers-to-Power-Next-Gen-AI-Platforms.html. Accessed 2026-05-26.
[^5]: SK hynix, "SK hynix Partners with TSMC to Strengthen HBM Technological Leadership", SK hynix Newsroom, 2024-04-19. https://news.skhynix.com/sk-hynix-partners-with-tsmc-to-strengthen-hbm-technological-leadership/. Accessed 2026-05-26.
[^6]: JEDEC Solid State Technology Association, "High Bandwidth Memory (HBM) DRAM (JESD235)", JEDEC, 2013. https://www.jedec.org/standards-documents/docs/jesd235. Accessed 2026-05-26.
[^7]: Anton Shilov, "JEDEC finalizes HBM4 memory standard with major bandwidth and efficiency upgrades", Tom's Hardware, 2025-04-17. https://www.tomshardware.com/pc-components/ram/jedec-finalizes-hbm4-memory-standard-with-major-bandwidth-and-efficiency-upgrades. Accessed 2026-05-26.
[^8]: Dylan Patel et al., "Scaling the Memory Wall: The Rise and Roadmap of HBM", SemiAnalysis, 2024. https://semianalysis.com/2024/11/13/the-memory-wall-hbm-roadmap/. Accessed 2026-05-26.
[^9]: TrendForce, "JEDEC Finalizes HBM4 Standard, Potentially Easing Pressure on Hybrid Bonding Adoption", TrendForce, 2025-04-21. https://www.trendforce.com/news/2025/04/21/news-jedec-finalizes-hbm4-standard-potentially-easing-pressure-on-hybrid-bonding-adoption/. Accessed 2026-05-26.
[^10]: SK hynix, "SK hynix Becomes Industry's First to Mass-Produce HBM3E", SK hynix Newsroom, 2024-03-19. https://news.skhynix.com/sk-hynix-becomes-industrys-first-to-mass-produce-hbm3e/. Accessed 2026-05-26.
[^11]: SK hynix, "SK hynix Completes World's First HBM4 Development", SK hynix Newsroom, 2025-09-12. https://news.skhynix.com/sk-hynix-completes-worlds-first-hbm4-development-and-readies-mass-production/. Accessed 2026-05-26.
[^12]: TrendForce, "Samsung Reportedly to Deliver HBM4 Samples to NVIDIA This Month, Eyes Early-2026 Validation", TrendForce, 2025-11-04. https://www.trendforce.com/news/2025/11/04/news-samsung-reportedly-to-deliver-hbm4-samples-to-nvidia-this-month-eyes-early-2026-validation/. Accessed 2026-05-26.
[^13]: TrendForce, "Samsung Reportedly Set to Begin Official HBM4 Shipments to NVIDIA and AMD in February", TrendForce, 2026-01-26. https://www.trendforce.com/news/2026/01/26/news-samsung-reportedly-set-to-begin-official-hbm4-shipments-to-nvidia-and-amd-in-february/. Accessed 2026-05-26.
[^14]: TrendForce, "SK hynix, Samsung Reportedly Deliver Paid HBM4 Samples to NVIDIA Ahead of 1Q26 Contracts", TrendForce, 2025-12-16. https://www.trendforce.com/news/2025/12/16/news-sk-hynix-samsung-reportedly-deliver-paid-hbm4-samples-to-nvidia-ahead-of-1q26-contract-finalization/. Accessed 2026-05-26.
[^15]: TrendForce, "SK hynix Reportedly to Supply About Two-Thirds of NVIDIA HBM4; Samsung Targets Early Delivery", TrendForce, 2026-01-28. https://www.trendforce.com/news/2026/01/28/news-sk-hynix-reportedly-to-supply-about-two-thirds-of-nvidia-hbm4-samsung-targets-early-delivery/. Accessed 2026-05-26.
[^16]: Lee Min-hyung, "Samsung advances HBM4 toward mass output after clearing key internal test", The Korea Economic Daily, 2025-12-03. https://www.kedglobal.com/korean-chipmakers/newsView/ked202512030011. Accessed 2026-05-26.
[^17]: TrendForce, "Samsung Reportedly Plans 50% HBM Capacity Surge in 2026, Spotlight on HBM4", TrendForce, 2025-12-30. https://www.trendforce.com/news/2025/12/30/news-samsung-reportedly-plans-50-hbm-capacity-surge-in-2026-spotlight-on-hbm4/. Accessed 2026-05-26.
[^18]: Micron Technology, "Micron in High-Volume Production of HBM4 Designed for NVIDIA Vera Rubin, PCIe Gen6 SSD and SOCAMM2", GlobeNewswire, 2026-03-16. https://www.globenewswire.com/news-release/2026/03/16/3256773/14450/en/Micron-in-High-Volume-Production-of-HBM4-Designed-for-NVIDIA-Vera-Rubin-PCIe-Gen6-SSD-and-SOCAMM2.html. Accessed 2026-05-26.
[^19]: TrendForce, "SK hynix Reportedly Weighs TSMC 3nm for HBM4E Logic Dies to Gain Edge over Samsung", TrendForce, 2026-03-20. https://www.trendforce.com/news/2026/03/20/news-sk-hynix-reportedly-weighs-tsmc-3nm-for-hbm4e-logic-dies-to-gain-edge-over-samsung/. Accessed 2026-05-26.
[^20]: TrendForce, "TSMC Unveils Custom C-HBM4E Details: N3P Logic Dies Reportedly Target 2x Efficiency Gain", TrendForce, 2025-12-01. https://www.trendforce.com/news/2025/12/01/news-tsmc-unveils-custom-c-hbm4e-details-n3p-logic-dies-reportedly-target-2x-efficiency-gain/. Accessed 2026-05-26.
[^21]: TrendForce, "GTC 2025: NVIDIA Unveils More on Rubin GPUs, Announces Feynman for 2028 in Roadmap Update", TrendForce, 2025-03-19. https://www.trendforce.com/news/2025/03/19/news-gtc-2025-nvidia-unveils-more-on-rubin-gpus-announces-feynman-for-2028-in-roadmap-update/. Accessed 2026-05-26.
[^22]: Anthony Garreffa, "AMD's next-gen Instinct MI400 GPU confirmed: rocks 432GB of HBM4 at 19.6TB/sec ready for 2026", TweakTown, 2025-06-13. https://www.tweaktown.com/news/105758/amds-next-gen-instinct-mi400-gpu-confirmed-rocks-432gb-of-hbm4-at-19-6tb-sec-ready-for-2026/index.html. Accessed 2026-05-26.
[^23]: Paul Alcorn, "AMD touts Instinct MI430X, MI440X, and MI455X AI accelerators and Helios rack-scale AI architecture at CES", Tom's Hardware, 2026-01-08. https://www.tomshardware.com/tech-industry/artificial-intelligence/amd-touts-instinct-mi430x-mi440x-and-mi455x-ai-accelerators-and-helios-rack-scale-ai-architecture-at-ces-full-mi400-series-family-fulfills-a-broad-range-of-infrastructure-and-customer-requirements. Accessed 2026-05-26.
[^24]: Google, "Ironwood: The first Google TPU for the age of inference", Google Cloud Blog, 2025-04-09. https://blog.google/products/google-cloud/ironwood-tpu-age-of-inference/. Accessed 2026-05-26.
[^25]: I-Connect007, "HBM4 Mass Production Delayed to End of 1Q26 By Spec Upgrades and Nvidia Strategy Adjustments", I-Connect007, 2026-01-19. https://iconnect007.com/article/148351/hbm4-mass-production-delayed-to-end-of-1q26-by-spec-upgrades-and-nvidia-strategy-adjustments/148348/ein. Accessed 2026-05-26.
[^26]: Tech Times, "Nvidia Vera Rubin Enters Full Production: Samsung, SK Hynix, Micron Named HBM4 Suppliers", Tech Times, 2026-06-02. https://www.techtimes.com/articles/317539/20260602/nvidia-vera-rubin-enters-full-production-samsung-sk-hynix-micron-named-hbm4-suppliers.htm. Accessed 2026-06-03.
[^27]: Jie Ye-eun, "Huang confirms Vera Rubin in full production with Korean memory", The Korea Herald, 2026-06-01. https://www.koreaherald.com/article/10761132. Accessed 2026-06-03.
[^28]: Lim Sae-rom, "SK Hynix ready to supply HBM4 memory chips", Data Center Dynamics, 2026. https://www.datacenterdynamics.com/en/news/sk-hynix-ready-to-supply-hbm4-memory-chips/. Accessed 2026-06-03.