Penguin Solutions Introduces Industry's First Production-Ready CXL-Based KV Cache Server

Penguin Solutions today announced its MemoryAI KV cache server, the industry's first production-ready KV cache server utilizing CXL memory technology.
New Electronics

Marvell introduces new CXL switch to tackle AI memory bottlenecks

Marvell introduces the Structera S 30260, a next‑generation CXL switch designed to enable rack‑level memory pooling.

Marvell Launches Next-generation CXL Switch, Enabling Memory Pooling to Break Through the AI "Memory Wall"

Marvell Technology, Inc. (NASDAQ: MRVL), a leader in data infrastructure semiconductor solutions, today announced Marvell® ...

Sandisk: This Nvidia GTC 2026 Announcement Could Be A Game Changer

Sandisk stock is up 158% YTD. Explore AI data center NAND demand, BiCS8 QLC SSD ramp, and Nvidia GTC 2026 memory hierarchy ...
Guru3D

Micron samples 256GB SOCAMM2 LPDDR5X modules for denser AI servers

Micron has announced it is shipping customer samples of a 256GB SOCAMM2 module built around low-power DRAM for data center platforms. The module targets a growing pain point in modern server design: ...

Micron squeezes 64 32GB LPDDR5X chips into one module

Micron squeezes 64 32GB LPDDR5x chips into one module ...

Micron Sets New Benchmark With the World's First High-Capacity 256GB LPDRAM SOCAMM2 for Data Center Infrastructure

News highlights: 1/3 the power consumption and 1/3 smaller footprint versus standard RDIMMs — enabled by the industry's first monolithic 32Gb ...
Electronics Weekly

Micron sampling 256GB LPDRAM module

Micron is sampling a 256GB LPDRAM module — 256GB SOCAMM2 – based on a 32Gb LPDDR5X die.

Micron Technology, Inc.: Micron Sets New Benchmark With the World's First High-Capacity 256GB LPDRAM SOCAMM2 for Data Center Infrastructure

News highlights: 1/3 the power consumption and 1/3 smaller footprint versus standard RDIMMs enabled by the industry's first monolithic 32Gb LPDDR5X die2.3 times faster time to first token for long-con ...

Alibaba's new open source Qwen3.5-Medium models offer Sonnet 4.5 performance on local computers

This leap is made possible by near-lossless accuracy under 4-bit weight and KV cache quantization, allowing developers to process massive datasets without server-grade infrastructure.