Memory and Storage Chips

DRAM · HBM · NAND · SRAM · NOR · emerging memory

DRAM makers earned about $155 billion in the second quarter of 2026 and NAND makers about $79 billion, both records, as AI servers took memory faster than fabs could add it. Samsung leads DRAM and NAND, SK hynix leads HBM, Micron is the only US-headquartered maker, and China's CXMT and YMTC are the fastest-growing challengers.

Memory revenue in the second quarter of 2026: DRAM about $155 billion, led by Samsung, SK hynix, Micron and CXMT; NAND about $79 billion, led by Samsung, SK hynix, Micron, Kioxia and SanDisk MEMORY REVENUE, 2Q 2026, $ BILLIONS $0B $40B $80B $120B $160B DRAM SAMSUNG 61 SK HYNIX 39 MICRON 36 $155B NAND SAMSUNG 23 $79B SOUTH KOREA UNITED STATES JAPAN CHINA TAIWAN AND OTHERS DRAM revenue rose 60% in one quarter; the top five NAND makers rose 77%. Samsung leads both; Micron is the only US-headquartered maker in either. SOURCE: TRENDFORCE, AUGUST AND SEPTEMBER 2026. NAND OTHERS DERIVED FROM SAMSUNG SHARE.
DRAM revenue
$155B, 2Q 2026
NAND revenue
~$79B, 2Q 2026
DRAM growth
+60% in a quarter
HBM leader
SK hynix, 50%
US maker
Micron
Hub
Chips

Memory families

Six Memory Families, Three That Matter for AI

DRAM, HBM and NAND carry nearly all memory revenue and all of the AI load. SRAM lives inside logic chips; NOR, MRAM and FRAM serve boot code and safety-critical state in cars and industrial gear.

Memory families
FamilyProductsProcessLeading makersWhere it goes
DRAMDDR5, LPDDR5X, GDDR7, DDR41-alpha to 1-gamma DRAM; EUV on a few layersSamsung, SK hynix, Micron, CXMT, Nanya, WinbondWorking memory in every server, PC, phone and car
HBMHBM3E, HBM4DRAM core dies on a logic base die, stacked with TSVsSK hynix, Samsung, MicronAI accelerators: NVIDIA, AMD, Google, AWS
NAND flashTLC and QLC 3D NAND218 to 321 layers in volume; etch-limitedSamsung, SK hynix and Solidigm, Micron, Kioxia, SanDisk, YMTCSSDs, phones, AI training and inference storage
SRAMEmbedded cache; standalone partsBuilt on the host logic node; standalone at 28 to 130nmTSMC and Samsung (embedded); Renesas, Infineon, ISSI, GSI TechnologyCPU, GPU and NPU caches; network buffers
NOR flashSPI NOR (Winbond W25Q, Macronix MX25)45 to 130nm planarWinbond, Macronix, GigaDevice, Infineon, MicronBoot and firmware in cars, MCUs and IoT
MRAM, FRAM, ReRAMEverspin MRAM; embedded MRAM at GlobalFoundries 22FDX and Samsung 28nm; Fujitsu FRAM; Weebit ReRAMBack-end-of-line integration at 22 to 28nmEverspin, GlobalFoundries, Samsung, TSMC, Fujitsu, Weebit NanoAutomotive MCUs, industrial state retention; Intel Optane (PCM) ended in 2022

Makers

Who Holds DRAM, HBM and NAND

Samsung, SK hynix and Micron are the only makers at scale in all three AI memories. Every one of them is an integrated device maker: memory has no foundry model, so capacity decisions in Suwon, Icheon and Boise reach world supply within months.

Revenue share by memory type, 2Q 2026
Maker (HQ)DRAMHBMNANDPosition
Samsung (South Korea)39.4%33%29.3%Largest DRAM and NAND maker; HBM4 in mass production since February 2026
SK hynix (South Korea)24.9%50%18.1%HBM leader; NAND includes Solidigm
Micron (United States)23.3%18%15.1%Only US-headquartered maker; DRAM output in Manassas, Virginia; Boise and Clay fabs building
Kioxia (Japan)--13.6%NAND only; shares Yokkaichi and Kitakami fabs with SanDisk
SanDisk (United States)--11.4%NAND only; spun off from Western Digital in February 2025
CXMT (China)9.5%Sampling-Commodity DDR4, DDR5 and LPDDR5; HBM3 sampling
YMTC (China)--~14% of bitsThird by NAND bits; on the US Entity List

DRAM and NAND shares: TrendForce (SK hynix, Micron and SanDisk NAND shares derived from its revenue figures). HBM shares: Counterpoint. YMTC bit share: Tom's Hardware reporting.

The HBM diversion

HBM Takes About Three Times the Wafer per Bit

Each HBM stack needs 12 to 16 known-good DRAM dies, through-silicon vias on every die and a stacking step that loses yield at each layer. Micron has put HBM3E at about three times the wafer supply of DDR5 for the same bits. When the three leaders move wafers to HBM for AI accelerators, DDR5 for servers and LPDDR5X for phones and cars tighten, even with no fab running short.

The HBM diversion: the same DRAM wafer yields far fewer bits as HBM than as DDR5, because each stack needs extra dies for through-silicon vias, testing and stacking losses; Micron has put HBM3E at about three times the wafer supply of DDR5 per bit ONE DRAM WAFER, TWO PRODUCTS (SCHEMATIC) 1-BETA DRAM WAFER DDR5 and LPDDR5X EVERY GOOD DIE SOLD AS IS HBM stack TSVS, 12-16 DIES, STACK LOSS 1x wafer per bit About 3x wafer per bit RATIO: MICRON, HBM3E AGAINST DDR5.

HBM then has to be joined to the GPU in TSMC CoWoS packaging, a separate line with its own capacity limit; see CoWoS.

Record

Page Record

Covers
Memory families, makers and shares, and the HBM wafer diversion
Sources
TrendForce DRAM and NAND revenue rankings (August and September 2026); Counterpoint HBM shares (September 2026); Micron investor statements; company releases
Last verified
October 2026