GCP Compute Instance

a3-highgpu-1g

The a3-highgpu-1g is a GPU-accelerated GCP Compute instance with 26 vCPUs and 234 GB of memory, 1 H100 GPU, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $10.9791/hr in us-central1. Spot pricing starts at $1.0881/hr in europe-north1.

Price History

Pricing by Region

Region On-Demand ↑Spot Reserved 1Y
us-central1$10.979$6.898$7.974
us-east4$10.979$6.587$7.638
us-west1$10.979$6.898$7.974
asia-south1$11.418$6.505$7.926
europe-north1$12.089$1.088$8.392
europe-west1$12.089$6.257$8.392
us-east5$12.366$6.120$8.555
us-west4$12.366$5.842$8.603
europe-west2$12.516$1.431$8.688
asia-east1$12.713$7.628$8.825
europe-west8$12.736$1.267$8.841
europe-west9$12.736$3.998$8.841
europe-west3$12.955$7.773$9.018
asia-south2$13.187$3.646$9.154
australia-southeast1$13.724$3.265$9.574
europe-west4$13.977$7.635$9.957
asia-northeast3$14.099$5.161$9.787
asia-southeast1$14.194$8.517$9.853
australia-southeast2$14.273$1.295$9.908
asia-northeast1$15.724$2.844$10.942

Spot Interruption Risk

Frequently Asked Questions

What is the a3-highgpu-1g?
The a3-highgpu-1g is a GPU-accelerated GCP Compute instance with 26 vCPUs and 234 GB of memory, 1 H100 GPU, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $10.9791/hr in us-central1. Spot pricing starts at $1.0881/hr in europe-north1.
What is the cheapest region for a3-highgpu-1g?
us-central1 at $10.979/hr on-demand.
How much does a3-highgpu-1g cost per month?
Starting at $8014.71/mo on-demand in us-central1, or $794.28/mo with spot pricing.