G4 family · General Purposex86_64 · Intel/AMD

g4-standard-384

The g4-standard-384 machine type has 384 vCPUs and 1440 GB of memory. Pricing for this instance starts at $36.00 per hour and $26279.59 monthly in the us-central1 region.

Updated June 18, 2026
vCPUs
384
Memory
1440 GB
Network
Up to 400 Gbps
Storage
Local SSD

Save up to $20887/mo on every g4-standard-384

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
$35.9994$26279.59 Baseline
1-Year Committed Use Discount (CUD)
$24.8400$18133.20 -31%
3-Year Committed Use Discount (CUD)
$15.8356$11559.99 -56%
Preemptible / Spot
$7.3869$5392.42 -79%
You could save $20887.17/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
$26279.59
730 hrs × $35.9994/hr × 1

Compute

x86_64 · Intel/AMD
vCPUs i384
Memory i1440 GB
Physical processorAMD Turin
Nested VirtualizationNot supported
Sole TenantNot supported
GPU8 GPU
Within G4 family
g4-standard-6
6 vCPU
g4-standard-12
12 vCPU
g4-standard-24
24 vCPU
g4-standard-48
48 vCPU
g4-standard-96
96 vCPU
g4-standard-192
192 vCPU
g4-standard-384
384 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 bandwidth400 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.
Specg4-standard-384g4-standard-6Δ
vCPUs3846-98%
Memory1440 GB22 GB-98%
Hourly Price$35.9994$0.5596-98%
Monthly Price$26279.59$408.48-98%
Cost optimization

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

Connect your GCP account. We'll identify every underutilized g4.* instance and show exactly where downsizing, cleaning up orphaned disks, or shifting to CUDs saves money — with zero code changes.