4/7
Providers w/ Spot
$0.07
Lowest Spot /hr
97%
Max Savings vs On-Demand
Hardware Specifications
Turing
16 GB GDDR6
VRAM
65 TFLOPS
FP16 Performance
8.1 TFLOPS
FP32 Performance
70W
Thermal Design Power
320 GB/s
Memory Bandwidth
Turing
GPU Architecture
Provider Spot Price On-Demand Savings Availability
Amazon Web Services $0.51/hr $2.26/hr 77% Available
Google Cloud Platform $0.17/hr $0.54/hr 69% Available
Microsoft Azure $0.10/hr $0.53/hr 80% Available
Lambda
CoreWeave
RunPod
Vast.ai $0.07/hr $0.15/hr 53% 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 T4 prices so you don't have to check each provider manually.