LIVE — Updated every 30 min
L40S Spot Price Comparison —
Real-Time Pricing Across Providers
Live spot prices for NVIDIA L40S across AWS, GCP, Azure, Lambda Cloud, CoreWeave, RunPod, and Vast.ai. Updated from live provider APIs every 30 minutes.
4/7
Providers w/ Spot
$0.79
Lowest Spot /hr
75%
Max Savings vs On-Demand
Hardware Specifications
Ada Lovelace
48 GB GDDR6X
VRAM
366 TFLOPS
FP16 Performance
91.6 TFLOPS
FP32 Performance
350W
Thermal Design Power
864 GB/s
Memory Bandwidth
Ada Lovelace
GPU Architecture
| Provider | Spot Price | On-Demand | Savings | Availability |
|---|---|---|---|---|
| Amazon Web Services | $1.82/hr | $3.22/hr | 43% | Available |
| GCP | — | — | — | — |
| Azure | — | — | — | — |
| Lambda | — | — | — | — |
| CoreWeave | $1.86/hr | $1.84/hr | -1% | Available |
| RunPod | $0.79/hr | $0.86/hr | 8% | Available |
| Vast.ai | $0.85/hr | $1.19/hr | 29% | Available |
What is GPU Spot Pricing?
GPU spot instances use unused cloud capacity at discounts of 60–90% compared to on-demand pricing. The tradeoff: cloud providers can reclaim instances with little notice (typically 30 seconds warning). For fault-tolerant workloads like batch inference, training, and CI/CD, spot pricing dramatically reduces GPU costs.
RoofRun monitors prices from 7 providers every 30 minutes and surfaces the cheapest L40S prices so you don't have to check each provider manually.