){kind=logo}
){kind=logo}
24 Aug 2026
Primemas Shatters AI Memory Wall at FMS 2026 with CXL Full-Stack Products and Micron Collaboration on the Abaco Project
Primemas Inc., a pioneer in next-generation AI memory system architectures and chiplet-based silicon, showcased its latest CXL memory innovations at the Future of Memory and Storage (FMS) 2026.
At the event, Primemas highlighted its strategic collaboration with Micron Technology on the Abaco Project, a rack-scale compute system with over 100TB of CXL-attached shared and pooled memory developed for the Department of Energy's (DOE) Pacific Northwest National Laboratory (PNNL) to power next-generation, large-scale scientific AI, HPC and data-intensive workloads.
Primemas' CXL add-in cards (PMA14/PMA16), a PCIe-based CXL memory expansion card built on its proprietary Hublet® architecture, ship to Micron in September, where they’ll be installed in the Abaco system racks. Primemas collaborates with Liqid on Abaco's memory pooling solution.
Primemas also unveiled SLiM, a switchless pooled memory solution designed for KV-cache offloading in AI inference workloads.
From SLiM to our roadmap for near-memory computing built on Hublet®, Primemas is redefining how memory scales for AI.
At the event, Primemas highlighted its strategic collaboration with Micron Technology on the Abaco Project, a rack-scale compute system with over 100TB of CXL-attached shared and pooled memory developed for the Department of Energy's (DOE) Pacific Northwest National Laboratory (PNNL) to power next-generation, large-scale scientific AI, HPC and data-intensive workloads.
Primemas' CXL add-in cards (PMA14/PMA16), a PCIe-based CXL memory expansion card built on its proprietary Hublet® architecture, ship to Micron in September, where they’ll be installed in the Abaco system racks. Primemas collaborates with Liqid on Abaco's memory pooling solution.
Primemas also unveiled SLiM, a switchless pooled memory solution designed for KV-cache offloading in AI inference workloads.
From SLiM to our roadmap for near-memory computing built on Hublet®, Primemas is redefining how memory scales for AI.