Medium–AI inference and Retrieval Augmented Generation (RAG)
- 16 NVIDIA H200NVL GPUs per server or RTX Pro 6000
- 2X HPE ProLiant Compute DL380a Gen12 worker nodes
- 3X HPE ProLiant Compute DL325 Gen 11 control plane nodes
- 109TB file storage based on Alletra MP
- 2X NVIDIA SNA700M switches (400 GbE data network)
- 2X Aruba 6300M (Management & ILO)
- HPE AI Essentials with NVIDIA AI Enterprise Software
- 3-or-5-year subscription
- Up to 13 kW
Large–AI inference, Retrieval Augmented Generation (RAG), and Model Fine Tuning
- 16 NVIDIA H200NVL GPUs per server or RTX Pro 6000
- 2X HPE ProLiant Compute DL380a Gen 12 worker nodes
- 3X HPE ProLiant Compute DL325 Gen 11 control plane nodes
- 217TB file storage based on Alletra MP
- 4X NVIDIA SN4700M switches (400 GbE data network)
- 2X Aruba 6300M (Management & ILO)
- HPE AI Essentials with NVIDIA AI Enterprise Software
- 3-or-5-year subscription
- Up to 17.4 kW