Reliably run training and inference on AI/ML models.

Hardware
–
Intel® Core™ i5-13500 (13th Gen Raptor Lake) with Hyper-Threading
–
6 Performance Cores + 8 Efficiency Cores
–
NVIDIA RTX™ 4000 SFF Ada Generation GPU
–
20 GB GDDR6 ECC GPU Memory
–
64 GB DDR4 RAM
–
2 × 1.92 TB NVMe SSD (Datacenter Edition, Gen3)
–
Software RAID 1 (mirrored storage)
Network
–
1 GBit/s dedicated network port
–
Guaranteed 1 GBit/s bandwidth
–
Unlimited traffic
–
47.15 TBit total network capacity
–
Redundant network infrastructure
–
Juniper routing technology
–
Minimum 99.9% network availability

Hardware
–
Intel® Xeon® Gold 5412U processor
–
Hyper-Threading Technology
–
NVIDIA RTX PRO™ 6000 Blackwell Max-Q GPU
–
96 GB GDDR7 ECC GPU memory
–
256 GB DDR5 ECC Registered RAM
–
Optional RAM upgrade up to 768 GB DDR5 ECC (additional charge)
–
2 × 960 GB NVMe SSD (Datacenter Edition)
–
Software RAID 1 (mirrored storage)
Network
–
1 GBit/s dedicated network port
–
Guaranteed 1 GBit/s bandwidth
–
Unlimited traffic
–
47.15 TBit total network capacity
–
Redundant network infrastructure
–
Juniper routing technology
–
High-speed access to all Internet uplinks

8× AMD Instinct™ MI300X GPUs, 192 GiB each
1,920 GiB system memory
160 vCPUs
1.25 TiB NVMe boot storage
40 TiB NVMe scratch storage
80,000 GiB transfer bandwidth

8× NVIDIA H100 GPUs, 80 GiB each
1,920 GiB system memory
160 vCPUs
1.25 TiB NVMe boot storage
40 TiB NVMe scratch storage
80,000 GiB transfer bandwidth