8 node HGX B300 AI pod, 64 GPUs
Allocation required- GPU nodes
- 8 x 8 GPU HGX B300 servers, 64 x B300
- Fabric
- 64 x 800 Gb/s ports, non blocking
- CPU and control
- 2 x CPU nodes for data services
- Chassis
- Multi rack, about 125 kW, air cooled
Clusters and racks
Complete AI pods and clusters, from an eight node HGX B300 pod to a 64 node training cluster with CPU, control plane and storage nodes, plus NVIDIA GB300 NVL72 and Vera Rubin NVL72 racks.
16 node HGX B300 AI pod, 128 GPUs
Allocation required- GPU nodes
- 16 x 8 GPU HGX B300 servers, 128 x B300
- Fabric
- 128 x 800 Gb/s ports, leaf and spine
- CPU and control
- 3 x CPU nodes for data services
- Chassis
- Multi rack, about 245 kW, air or liquid cooled
64 node HGX B300 AI cluster, 512 GPUs
Allocation required- GPU nodes
- 64 x 8 GPU HGX B300 servers, 512 x B300
- Fabric
- 512 x 800 Gb/s ports, leaf and spine
- CPU tier
- 8 x dual socket EPYC 9005 nodes
- Chassis
- Multi rack, about 950 kW, air or liquid cooled
8 node HGX B200 AI pod, 64 GPUs
- GPU nodes
- 8 x 8 GPU HGX B200 servers, 64 x B200
- Fabric
- 64 x 800 Gb/s ConnectX-8 ports
- CPU and control
- 2 x CPU nodes for data services
- Chassis
- Multi rack, about 125 kW, air cooled
8 node HGX H200 AI pod, 64 GPUs
Available on request- GPU nodes
- 8 x 8 GPU HGX H200 servers, 64 x H200 SXM
- Fabric
- 64 x 400 Gb/s ConnectX-7 NDR InfiniBand
- CPU and control
- 2 x CPU nodes for data services
- Chassis
- Multi rack, about 95 kW, air cooled
NVIDIA GB300 NVL72 rack
Allocation required- GPUs
- 72 x NVIDIA Blackwell Ultra, 288 GB HBM3e each
- CPUs
- 36 x NVIDIA Grace, 2,592 Arm Neoverse V2 cores
- NVLink
- 130 TB/s total NVLink bandwidth across all 72 GPUs
- Chassis
- Single rack, power not yet published, liquid cooled
GB300 NVL72 multi rack AI factory
Allocation required- GPU memory
- 20 TB HBM3e per rack as published by NVIDIA
- Scale out fabric
- 72 x 800 Gb/s ConnectX-8 ports per rack into a Quantum-X800
- Chassis
- Multi rack, power not yet published, liquid cooled
NVIDIA Vera Rubin NVL72 rack
Coming soon- GPUs
- 72 x NVIDIA Rubin, 288 GB HBM4 each
- CPUs
- 36 x NVIDIA Vera, 3,168 custom Olympus cores
- NVLink
- Sixth generation, 216 TB/s per rack
- Chassis
- Single rack, power not yet published, liquid cooled
Vera Rubin NVL72 multi rack AI factory
Coming soon- GPU memory
- 20.7 TB HBM4 per rack at 1,400 TB/s
- Performance
- 3,600 PFLOPS sparse NVFP4 inference and 2,520 PFLOPS dense NVFP4 training per rack
- Scale out fabric
- ConnectX-9 SuperNICs and BlueField-4 DPUs into Quantum-X800
- Chassis
- Multi rack, power not yet published, liquid cooled
8 node HGX Rubin NVL8 AI pod
Coming soon- GPU nodes
- 8 x 8 GPU HGX Rubin NVL8 servers, 64 x Rubin
- Fabric
- 64 x 800 Gb/s ports, non blocking
- CPU and control
- x86 or Vera hosts per NVIDIA's published options
- Chassis
- Multi rack, about 125 kW, air or liquid cooled
No systems match these filters.
About clusters and racks
We specify compute, fabric, storage and power together, have the cluster racked, cabled and tested before it ships, and deliver it integrated so that it is deployable as delivered into your own data center or colocation facility.
Not seeing the configuration you need?
Tell us the workload and we specify it: the GPU, the chassis, the CPU, memory and storage around it, and the fabric if there is more than one node.
- (858) 613-1818
- sales@computermodules.com
- Monday to Friday, 8am to 5pm Pacific