A2 family · Accelerator Optimizedx86_64 · Intel/AMD

a2-ultragpu-1g

The a2-ultragpu-1g machine type has 12 vCPUs and 170 GB of memory. Pricing for this instance starts at $5.07 per hour and $3700.22 monthly in the us-central1 region.

Updated June 26, 2026
vCPUs
12
Memory
170 GB
Network
Up to 24 Gbps
Storage
Local SSD

Save up to $1853/mo on every a2-ultragpu-1g

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
$5.0688$3700.22 Baseline
1-Year Committed Use Discount (CUD)
$5.0688$3700.22 Baseline
3-Year Committed Use Discount (CUD)
$5.0688$3700.22 Baseline
Preemptible / Spot
$2.5303$1847.13 -50%
You could save $1853.09/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
$3700.22
730 hrs × $5.0688/hr × 1

Compute

x86_64 · Intel/AMD
vCPUs i12
Memory i170 GB
Physical processorIntel Cascade Lake
Nested VirtualizationNot supported
Sole TenantNot supported
GPU1 GPU
Within A2 family
a2-highgpu-1g
12 vCPU
a2-ultragpu-1g
12 vCPU
a2-highgpu-2g
24 vCPU
a2-ultragpu-2g
24 vCPU
a2-highgpu-4g
48 vCPU
a2-ultragpu-4g
48 vCPU
a2-highgpu-8g
96 vCPU
a2-ultragpu-8g
96 vCPU
a2-megagpu-16g
96 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 bandwidth24 Gbps
Tier 1 bandwidthN/A
Enhanced networking iYes (gvnic/virtio)
IPv6 supportYes (Dual-stack VPC)

Storage

Max persistent disks128
Max disk size257 TB
Local SSDsSupported (Scratch Disk)

Compare with another instance

vs.
Speca2-ultragpu-1ga2-highgpu-1gΔ
vCPUs12120%
Memory170 GB85 GB-50%
Hourly Price$5.0688$3.6734-28%
Monthly Price$3700.22$2681.57-28%
Cost optimization

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

Connect your GCP account. We'll identify every underutilized a2.* instance and show exactly where downsizing, cleaning up orphaned disks, or shifting to CUDs saves money — with zero code changes.