Amazon Web Services
EC2 accelerated computing instances (P5, P4, G6, G5).
9
GPU models
38
offers
5
regions
Regions
DEIEJPSGUS
Offers by GPU
NVIDIA T4 16GB16 GB · Turing
NVIDIA L4 24GB24 GB · Ada Lovelace
NVIDIA A10 24GB24 GB · Ampere
NVIDIA L40S 48GB48 GB · Ada Lovelace
NVIDIA Tesla V100 SXM2 16GB16 GB · Volta
NVIDIA A100 SXM4 40GB40 GB · Ampere
| vCPU / RAM | |||||
|---|---|---|---|---|---|
NVIDIA A100 SXM4 40GB 40 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $1.652/GPU/hr | $0.0413/GB/hr | Launch |
NVIDIA A100 SXM4 40GB 40 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $2.386/GPU/hr | $0.0596/GB/hr | Launch |
NVIDIA A100 SXM4 40GB 40 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $3.67/GPU/hr | $0.0918/GB/hr | Launch |
NVIDIA H100 SXM5 80GB80 GB · Hopper
| vCPU / RAM | |||||
|---|---|---|---|---|---|
NVIDIA H100 SXM5 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $1.769/GPU/hr | $0.0221/GB/hr | Launch |
NVIDIA H100 SXM5 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $2.555/GPU/hr | $0.0319/GB/hr | Launch |
NVIDIA H100 SXM5 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $3.93/GPU/hr | $0.0491/GB/hr | Launch |
NVIDIA A100 SXM4 80GB80 GB · Ampere
| vCPU / RAM | |||||
|---|---|---|---|---|---|
NVIDIA A100 SXM4 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $1.845/GPU/hr | $0.0231/GB/hr | Launch |
NVIDIA A100 SXM4 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $2.665/GPU/hr | $0.0333/GB/hr | Launch |
NVIDIA A100 SXM4 80GB 80 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $4.10/GPU/hr | $0.0513/GB/hr | Launch |
NVIDIA H200 SXM 141GB141 GB · Hopper
| vCPU / RAM | |||||
|---|---|---|---|---|---|
NVIDIA H200 SXM 141GB 141 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $2.025/GPU/hr | $0.0144/GB/hr | Launch |
NVIDIA H200 SXM 141GB 141 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $2.925/GPU/hr | $0.0207/GB/hr | Launch |
NVIDIA H200 SXM 141GB 141 GB · 8× GPU | 🇺🇸🇮🇪🇯🇵🇸🇬🇩🇪 | 96 vCPU · 768 GB RAM | $4.50/GPU/hr | $0.0319/GB/hr | Launch |