Samsung Electronics vs SK Hynix
Relationship
Nvidia, the buyer, certified Samsung and SK hynix (among others) as qualified, interchangeable HBM4 suppliers for the same Vera Rubin platform — a direct buyer-choice claim.
2 of 9 capabilities — Shares AI compute hardware and memory.
Ludbee capability tags · from the product recordsShared product type — Both ship hardware.
Ludbee product recordsSourced competitor — “NVIDIA (NASDAQ:NVDA) CEO Jensen Huang confirmed on June 5, 2026 that Samsung Electronics (LON:0593xq), SK Hynix Inc (KS:000660), and Micron Technology Inc (NASDAQ:MU) have all passed certification to supply HBM4 high-bandwidth memory for t…”
finance.yahoo.com · checked 2026-09-19Buy high-bandwidth memory — Rivals on this job — Source the stacked memory an accelerator design packages beside its compute dies.
Ludbee needs vocabulary · the scope on the sourced edgeAligned comparison
Capability overlap
Shared · 2
Not verified for SK Hynix · 7
Recorded for Samsung Electronics. SK Hynix’s product records say nothing either way — a missing record is not a missing capability.
Not verified for Samsung Electronics · 2
Recorded for SK Hynix. Samsung Electronics’s product records say nothing either way — a missing record is not a missing capability.
Products, side by side
Hand-checked pairing
Samsung Electronics
Hardware
DRAM-based CXL memory expansion module (MD220 and MD310 variants) enabling memory capacity and bandwidth pooling for AI, HPC and in-memory database workloads.
Enterprise NVMe and SATA SSD line (PM9D3a, PM9A3, PM893a, PM893, PM897a, PM897) that Samsung markets for reducing AI training and inference data-loading time.
Graphics DRAM delivering up to 40 Gbps per pin and up to 24Gb density via PAM3 signaling, for graphics, AI inference acceleration and autonomous-driving sensor fusion workloads.
Third-generation High Bandwidth Memory delivering up to 819 GB/s bandwidth and 24GB capacity for AI accelerators and datacenter systems.
HBM3E in a 12-layer stack offering up to 1,180 GB/s bandwidth and 36GB capacity, in mass production for AI, datacenter and supercomputer applications.
HBM4 with 2,048 I/O pins, up to 3,300 GB/s bandwidth and 1c DRAM on a 4nm base die, positioned for AI accelerators and datacenter systems.
SK Hynix
Hardware
CXL 2.0 memory module that expands and pools DDR5 capacity and bandwidth beyond the CPU's own channels for AI and high-performance computing servers, with a 96GB version through customer validation and a 128GB 1bnm version following.
Fourth-generation high-bandwidth memory that vertically stacks up to 12 DRAM layers for 36GB at 9.6Gbps, sold to GPU and AI accelerator makers for AI server packages.
Fifth-generation high-bandwidth memory with 2,048 I/O terminals and over 10Gbps operating speed, roughly doubling HBM3E bandwidth for next-generation AI accelerators.
PCIe Gen5 enterprise SSD series in E1.S form factor with in-house ASIC and V8 4D NAND, power-optimised for AI servers and NVIDIA RVL-qualified for GB200 -- one model shipping today, PEB110 (1,920-7,680GB, 20W, mass production).
PCIe Gen5 enterprise SSD series with in-house ASIC and V7 4D NAND for high-performance datacenter workloads, supporting up to 16TB in U.2/3 and the newer E3.S form factor -- two models shipping today, PS1010 and PS1030.
192GB LPDDR5X-based low-power memory module built on the 1cnm process and designed as main memory for AI servers, delivering over double the bandwidth of conventional RDIMM at over 75% better power efficiency.
No counterpart
Samsung Electronics sells these in a stack layer with no product recorded for SK Hynix yet — nothing on the other side to compare them against.
Application
Samsung's voice assistant for hands-free control of Galaxy devices, SmartThings smart-home devices and Quick Commands routines, listed among Galaxy AI's features on Samsung's Galaxy AI page.
On-device and cloud AI features built into Galaxy phones and tablets, covering translation, writing and photo editing.
On-screen AI assistant built into Samsung Smart TVs that answers questions and assists viewers through the television interface.
Infrastructure service
A foundry service bundling advanced process nodes (5nm to 2nm GAA), 2.5D/3D chiplet packaging, custom-HBM logic-memory integration and co-packaged optics for customers designing AI SoCs and chiplet platforms.