Live Floor
$0.1089
Vast · CA
The NVIDIA V100 is based on the Volta architecture with 16 GB of HBM2 VRAM.
Best live route via Vast · CA
Live Floor
$0.1089
Vast · CA
Average Price
$0.1791
Across 9 live offers
Reliability
98.4%
Based on 9 scored hosts
Last Day Delta
-$0.0269
Latest daily low: 2026-08-29
30-Day Price Trend
Available Offers
Sorted by cheapest hourly rate
| Provider | Price/hr | VRAM | GPUs | Reliability | Region | |
|---|---|---|---|---|---|---|
| Vast | $0.1089 | 16 GB | 1x | 98.4% | CA | View Deal |
| Vast | $0.1533 | 32 GB | 1x | 99.2% | BG | View Deal |
| Vast | $0.1618 | 32 GB | 1x | 99.7% | CA | View Deal |
| Vast | $0.1674 | 32 GB | 1x | 99.4% | BG | View Deal |
| Vast | $0.1815 | 32 GB | 1x | 99.7% | US | View Deal |
| Vast | $0.1889 | 32 GB | 1x | 99.4% | US | View Deal |
| RunPod | $0.1900 | 16 GB | 1x | 90.0% | -- | View Deal |
| Vast | $0.2102 | 32 GB | 1x | 99.7% | US | View Deal |
| Vast | $0.2502 | 32 GB | 1x | 100.0% | US | View Deal |
GPU Overview
The NVIDIA V100 is based on the Volta architecture with 16 GB of HBM2 VRAM. It was a landmark datacenter GPU that introduced Tensor Cores, and it continues to serve as a scalable accelerator in legacy AI infrastructure and cloud instances.
Its 16 GB of VRAM supports inference on mid-sized models and training of smaller architectures. The V100 remains a practical choice for budget-conscious datacenter deployments where established ML pipelines run efficiently on Volta-generation hardware, particularly in cloud environments offering competitive per-hour pricing.