Press Releases

Primemas, a fabless semiconductor innovator headquartered in Silicon Valley, is set to begin mass production of JBOM (Just a Bunch of Memory) in the second half of 2025 — a CXL-based solution that expands server memory capacity from the conventional 2–4 TB limit to hundreds of terabytes per CPU server. Designed to tackle the memory bottleneck that prevents expensive GPUs from being fully utilized in AI data centers, JBOM delivers a scalable memory pool architecture that directly translates to higher GPU utilization and AI workload performance. The company has secured mass supply agreements with Micron and is co-developing next-generation CXL memory solutions with Samsung Electronics. Alibaba's Beluga research validates the approach, demonstrating a 7.35x throughput improvement with CXL-based memory expansion. With a roadmap extending beyond 240 TB by 2026 and $72 million raised in its Series B, Primemas is positioning itself at the forefront of the memory-centric computing era.