GCP Compute Instance
a2-highgpu-8g
The a2-highgpu-8g is a GPU-accelerated GCP Compute instance with 96 vCPUs and 680 GB of memory, 8 A100 GPUs, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $29.3871/hr in us-central1. Spot pricing starts at $17.5792/hr in us-west4.
Price History
Pricing by Region
| Region | On-Demand ↑ | Spot | Reserved 1Y |
|---|---|---|---|
| us-central1 | $29.387 | $17.632 | $18.514 |
| us-east1 | $29.387 | $17.632 | $18.514 |
| us-west1 | $29.387 | $17.632 | $18.514 |
| europe-west4 | $29.984 | $17.986 | $18.890 |
| us-west4 | $31.475 | $17.579 | $19.656 |
| asia-southeast1 | $32.110 | $19.218 | $20.056 |
| me-west1 | $32.326 | $19.395 | $20.365 |
| asia-northeast1 | $32.397 | $19.389 | $20.237 |
| asia-northeast3 | $32.397 | $19.389 | $20.237 |
| us-west3 | $34.601 | $20.708 | $21.952 |
Spot Interruption Risk
Frequently Asked Questions
- What is the a2-highgpu-8g?
- The a2-highgpu-8g is a GPU-accelerated GCP Compute instance with 96 vCPUs and 680 GB of memory, 8 A100 GPUs, powered by Intel. Ideal for deep learning training, high-performance computing, and GPU-intensive inference workloads. On-demand pricing starts at $29.3871/hr in us-central1. Spot pricing starts at $17.5792/hr in us-west4.
- What is the cheapest region for a2-highgpu-8g?
- us-central1 at $29.387/hr on-demand.
- How much does a2-highgpu-8g cost per month?
- Starting at $21452.57/mo on-demand in us-central1, or $12832.82/mo with spot pricing.
