AI Dynamics

Global AI News Aggregator

About

GLM 5.2 MoE, NVFP4 467GB, DGX Station memory, offloading works

GLM 5.2 is an MoE, NVFP4 is 467 GB, and the DGX Station comes with 496GB LPDDR5X + 252GB HBM3e GPU memory With the right offloading formula, it should work

→ View original post on X — @theahmadosman