Showing 73–84 of 175 systems
SKU: 478897
~72 TB HBM3e (512× 141 GB) · 128× Intel Xeon 6 (64 nodes) · 128 TB DDR5 ECC · 8 PB parallel NVMe · Spine-leaf InfiniBand (NDR/XDR)
Mid AI supercluster
Fits Frontier multi-rack
SKU: 457680
~98 TB HBM3 (512× 192 GB) · 128× AMD EPYC (64 nodes) · 128 TB DDR5 ECC · 8 PB parallel NVMe · Infinity Fabric + InfiniBand (NDR)
Large AMD Instinct AI supercomputer
Fits HPC + frontier AI
SKU: 148193
36,096 GB HBM3e (256× 141 GB) · 128× AMD EPYC 9005 (12,288 cores) · 64 TB DDR5 ECC · 4 PB NVMe · Non-blocking InfiniBand NDR
CPU+GPU HPC cluster
Fits HPC + large-model AI
SKU: 145727
64× 800G OSFP (XDR) · 102 Tb/s · InfiniBand XDR · Sub-600 ns port-to-port
Fits GPU scale-out fabric
SKU: 209450
11,520 GB HBM3e (64× 180 GB) · 16× Intel Xeon 6 · 16 TB DDR5 ECC · 480 TB NVMe · NVLink + InfiniBand NDR 400G
Full-rack Blackwell training cluster
Fits Trillion-scale
SKU: 118513
~1.3 PB HBM3e (4,608× 288 GB) · 2,304× NVIDIA Grace (ARM) · Grace LPDDR5X coherent memory · 64 PB parallel NVMe · Spine-leaf InfiniBand (NDR/XDR)
Flagship sovereign AI factory
Fits Frontier multi-rack
SKU: 411269
9,024 GB HBM3e (64× 141 GB) · 32× Intel Xeon 6 · 8 TB DDR5 ECC · 960 TB NVMe · InfiniBand NDR 400G
Self-contained AI micro data center
Fits 70B+ local
SKU: 571974
9,024 GB HBM3e (64× 141 GB) · 32× Intel Xeon 6 (16 blades) · 16 TB DDR5 ECC · 320 TB NVMe · InfiniBand NDR 400G
Dense multi-node training & inference
Fits 70B+ training
SKU: 949018
9,024 GB HBM3e (64× 141 GB) · 16× Intel Xeon 6 · 16 TB DDR5 ECC · 480 TB NVMe · NVLink + InfiniBand NDR 400G
Full-rack training cluster
Fits Trillion-scale
SKU: 900430
HBM4 (high capacity per GPU) · AMD EPYC (Venice-class) host nodes · Configurable per node · Parallel filesystem + NVMe scratch tier · Open Ethernet scale-up + scale-out fabric
Sovereign HPC + AI rack
Fits HPC + frontier AI
SKU: 117261
700 kW · L2L CDU (in-row) · Up to 1,100 L/min · Up to 6 dense racks
Fits Liquid-cooled GPU racks
SKU: 900072
~31 TB HBM4 (432 GB/GPU) · AMD EPYC (Venice-class) host nodes · Configurable per node · Configurable parallel-FS / NVMe tier · All-Ethernet open scale-up + scale-out fabric
AMD rack-scale frontier training
Fits Trillion-parameter
Showing 73–84 of 175 systems 