DGX
NVIDIA's fully integrated AI server and SuperPOD reference system sold as a turnkey appliance.
Current numbers
280 / 140 GB/sDGX SuperPOD with Vera Rubin NVL72, Standard 90/45 GB/s per SU; Enhanced read/write per 8-system (576-GPU) SU
~40 → 600 kWKyber ~600 kW (NVIDIA H2 2027, SemiAnalysis 2028), up from ~132 kW (HPE GB200 NVL72) and ~40 kW (DGX H100)
CUDA 13.3 Update 1 / NCCL 2.31.2DGX OS 8.0.0 release snapshot, 31 August 2026; validate the supported system and full tuple