DallasTexas · US-TX-1
420cards in catalogA dedicated NVIDIA H100 SXM with 80 GB HBM3, 24 vCPU, 192 GB system RAM, and 3.84 TB local NVMe per card.
Node pricing
The product model allocates each GPU to one account for the complete term. Adding cards to a node applies the catalog volume discount to every card.
| Node | Total VRAM | 7 days | 30 days | Discount |
|---|---|---|---|---|
| 1× H100 SXM | 80 GB | $475 | $1,690 | 0% |
| 2× H100 SXM | 160 GB | $903 | $3,211 | 5% |
| 4× H100 SXM | 320 GB | $1,748 | $6,219 | 8% |
| 8× H100 SXM | 640 GB | $3,420 | $12,168 | 10% |
Regional inventory
Ready-to-run images
Workload fit
A candidate for Hopper-optimized training and inference, particularly when a verified multi-GPU fabric is important to the intended parallelism.
Your catalog configuration pairs 80 GB GPU memory with 192 GB host RAM and 3.84 TB local NVMe per card.
The SXM name alone does not establish the allocated host topology. Compare a representative pilot against the lower-priced 80 GB alternatives.
Use $475 for a 7-day pilot, or $1,690 for a 30-day allocation after validating the workload. Options are additional.