LLM Icon
Eight GPU LLM Server
Machine Learning and AI recommended system banner

Eight GPU Large Language Model Server

Powerful 4U rackmount server supporting up to eight NVIDIA GPUs for training, fine-tuning, and inference with AI large language models.


Eight GPU 4U server supporting NVIDIA RTX Ada, L40S, and H100 NVL graphics cards

  • Up to 752GB of VRAM across eight GPUs
  • Great for 150B parameter fp16 inference and fine-tuning smaller models
  • Requires four 200-240V power connections on separate circuits

Not sure what you need?

and one of our experts will reply within 1 business day to help configure the right computer for your workflow. If you don’t see what you are looking for here, check out our other systems for more options.

System Core


NVIDIA Mellanox Dual 100GbE QSFP28 PCI-E Card   Limited Supply [add $1150.94]
Up to one PCI-E x16 and one x8 card may be selected for additional high-speed networking capability.


Internal M.2 slot supports PCIe Gen3 x4
These drive bays are not connected by default – if you need to use them, please talk to a consultant.

Chassis & Cooling

This system requires four 200-240V power connections. For redundancy to be functional, the total power consumption of the system must be lower than the maximum output of two of the PSU modules.



Select this NVLink kit only if you picked 2 or more NVIDIA H100 NVL GPUs.

Additional Information

Help us help you! We review each configuration to ensure you’re getting the right hardware. Any info you can provide about your workflow and software will help us provide you with a better experience.

System Cost


per unit

Typically ships in 1-2 weeks

Contact us for lead times

Contact us for quotes for more than 100 units