AWS EC2 Instance

g4dn.metal

The g4dn.metal is a GPU-accelerated AWS EC2 instance with 96 vCPUs and 384 GB of memory, 8 T4 GPUs, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $7.8240/hr in us-east-1. Spot pricing starts at $0.9596/hr in me-south-1.

Price History

Pricing by Region

Region On-Demand ↑Spot Reserved 1Y
us-east-1$7.824$4.424$5.396
us-east-2$7.824$4.048$5.396
us-west-2$7.824$4.722$5.396
eu-north-1$8.300$1.915$5.725
ap-south-1$8.612$2.605$5.940
ca-central-1$8.687$4.381$5.992
eu-west-1$8.724$4.292$6.519
eu-west-3$9.148$3.397$6.310
eu-west-2$9.154$1.201$6.840
eu-south-1$9.160$2.734$6.844
us-west-1$9.389$2.101$7.232
me-south-1$9.596$0.960$7.171
ap-northeast-2$9.624$4.331$6.628
eu-central-1$9.780$4.397$7.308
ap-southeast-2$10.174$4.931$7.018
af-south-1$10.381$10.381$7.757
ap-northeast-1$10.562$5.407$7.893
ap-northeast-3$10.562$4.096$7.893
ap-southeast-1$10.948$3.545$7.551
ap-east-1$12.048$1.261$8.311
sa-east-1$13.298$1.330$9.930

Spot Interruption Risk

Frequently Asked Questions

What is the g4dn.metal?
The g4dn.metal is a GPU-accelerated AWS EC2 instance with 96 vCPUs and 384 GB of memory, 8 T4 GPUs, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $7.8240/hr in us-east-1. Spot pricing starts at $0.9596/hr in me-south-1.
What is the cheapest region for g4dn.metal?
us-east-1 at $7.824/hr on-demand.
How much does g4dn.metal cost per month?
Starting at $5711.52/mo on-demand in us-east-1, or $700.51/mo with spot pricing.