A3 family · Accelerator Optimizedx86_64 · Intel/AMD

a3-highgpu-2g

The a3-highgpu-2g machine type has 52 vCPUs and 468 GB of memory. Pricing for this instance starts at $22.12 per hour and $16149.42 monthly in the us-central1 region.

Updated June 16, 2026
vCPUs
52
Memory
468 GB
Network
Up to 200 Gbps
Storage
Local SSD

Save up to $8885/mo on every a3-highgpu-2g

Connect Google Cloud — we'll find the optimal pricing model for your workload.

All pricing options

us-central1 · USD · monthly = 730 hrs
Prices exclude local taxes
TermHourlyMonthlySavings
On-Demand
$22.1225$16149.42 Baseline
1-Year Committed Use Discount (CUD)
$15.3459$11202.52 -31%
3-Year Committed Use Discount (CUD)
$9.7161$7092.75 -56%
Preemptible / Spot
$9.9519$7264.90 -55%
You could save $8884.52/mo on this instance with the right pricing model.
Find my savings

Monthly cost estimator

Estimate your spend based on actual usage.
730 h
1h365h730h (24/7)
1
Estimated monthly
$16149.42
730 hrs × $22.1225/hr × 1

Compute

x86_64 · Intel/AMD
vCPUs i52
Memory i468 GB
Physical processorIntel Sapphire Rapids
Nested VirtualizationNot supported
Sole TenantNot supported
GPU2 GPU
Within A3 family
a3-highgpu-1g
26 vCPU
a3-highgpu-2g
52 vCPU
a3-highgpu-4g
104 vCPU
a3-highgpu-8g
208 vCPU

Lock in this rate across your fleet — typically save 30–40%

Connect your Google Cloud account to apply this pricing logic to every running instance automatically.

Networking

Max egress bandwidth200 Gbps
Tier 1 bandwidthN/A
Enhanced networking iYes (gvnic/virtio)
IPv6 supportYes (Dual-stack VPC)

Storage

Max persistent disks128
Max disk size512 TB
Local SSDsSupported (Scratch Disk)

Compare with another instance

vs.
Speca3-highgpu-2ga3-highgpu-1gΔ
vCPUs5226-50%
Memory468 GB234 GB-50%
Hourly Price$22.1225$11.0612-50%
Monthly Price$16149.42$8074.71-50%
Cost optimization

GCP recommender alerts can cut this bill ~30-40%.

Connect your GCP account. We'll identify every underutilized a3.* instance and show exactly where downsizing, cleaning up orphaned disks, or shifting to CUDs saves money — with zero code changes.