SK hynix begins mass production of 192GB SOCAMM2 memory aimed at Nvidia’s Vera Rubin

The announcement

SK hynix said it has started mass production of a 192GB SOCAMM2 — a next-generation memory module built on its sixth-generation 10-nanometer-class LPDDR5X technology. It has been reported that the module was designed particularly for use with Nvidia’s Vera Rubin AI platform, adapting mobile-oriented low-power DRAM to fit server environments. Big claim? Yes. Bold play? Absolutely.

The technical leap

It has been reported that the SOCAMM2 delivers more than double the bandwidth and over 75% better power efficiency compared with conventional RDIMMs, positioning LPDDR5X as a contender where RDIMM and HBM have traditionally ruled. SK hynix says the 192GB capacity is intended as primary memory for next-gen AI servers — think very large models and very hungry workloads. It’s like putting a phone’s efficient engine under a data-center hood; designed for thrift and speed.

Why it matters

SK hynix says the module will help ease memory bottlenecks in training and inference for large language models with hundreds of billions of parameters, significantly improving system performance. It has been reported that Kim Joo-sun, president and head of AI infrastructure at SK hynix, said, “By supplying the 192GB SOCAMM2, SK hynix has established a new standard for AI memory performance.” If true, this could be a real relief for engineers wrestling with bandwidth and power limits — and another sign that the AI-infrastructure arms race is moving beyond GPUs to the memory that feeds them.

The bigger picture

This move underscores a broader industry trend: memory innovation is now front-and-center in AI scale-up, not just raw compute. Will LPDDR5X displace established server memory standards? Not overnight. But expect more vendors to test mobile-rooted tech in datacenter roles as model sizes and energy bills keep climbing.

Sources: koreaherald.com