AIThe Next Webabout 1 hour ago
Startup ditches GPUs to attack AI memory bottleneck
1 min read
Majestic Labs unveiled Prometheus, a server replacing Nvidia GPUs with Ignite AI Processing Units that pool up to 128TB of LPDDR6 memory. The Tel Aviv startup claims one rack matches 25 Nvidia Vera Rubin racks for fast memory at a fraction of the power.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
