AWS Amazon EC2

g4dn.4xlarge

⚡ 1x GPU • 16 GB

Part of the G4DN Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.

Name:gGPU Graphics & ML Inference44th GenerationdDirect-Attached NVMe SSDnEnhanced Networking4xlarge4 Extra Large

On-demand, Linux

$1.2040/ hr

~$878.92/mo · lowest in US East (N. Virginia), us-east-1

Savings Plan, Linux

$0.6240/ hr

From, with a 3-year Compute Savings Plan, No Upfront: −48% vs on-demand rate (us-east-1)

Per unit

$0.0752/ vCPU-hr

$0.0188/GB-hr

Compute

16vCPUs

8 cores • x86_64

Memory

64GB

4.0 GB / vCPU

Network

Up to 25 Gbps

Baseline: 20 Gbps

Storage & EBS

593.75MB/s

Max: 20,000 IOPS

Alternatives to g4dn.4xlarge

Option	Price vs g4dn.4xlarge	Why look at it
g5.4xlarge Compare →	+35%+$0.42/hr Linux on-demand, us-east-1	AMD (x86_64), newer generation, same vCPU and RAM
g6.4xlarge Compare →	+10%+$0.119/hr Linux on-demand, us-east-1	AMD (x86_64), newer generation, same vCPU and RAM

On Google Cloud

g4dn.4xlarge is an accelerated machine. Those are compared by the accelerator, not by vCPU and memory, so this page doesn't pick a match. Browse Google Cloud machines.

Technical Specifications & Limits

Hardware specs, container thresholds, and system limits

Compute & Processor

ProcessorIntel Xeon Family

Architecturex86_64 (x86_64)

vCPUs / Threads16 vCPUs

Physical Cores8

Clock Speed2.5 GHz

Intel AVX / AVX2AVX & AVX2 Supported

Intel Turbo BoostYes

Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score230,991

CoreMark / vCPU14,437 / core

FFmpeg Transcoding106 FPS

CoreMark per Dollar191,853 score / $

Cost per vCPU-hr$0.0752

Cost per GB-hr$0.0188

Containers & Kubernetes

EKS Max Pods29 pods

Max ENIs3

IPv4/v6 per ENI10

Total Addressable IPs30 total IPs

CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network PerformanceUp to 25 Gbps

Baseline Bandwidth20 Gbps

Burst / Peak Bandwidth25 Gbps

Enhanced NetworkingYes

EFA SupportNo

IPv6 SupportNative Dual-Stack Supported

Storage & EBS Optimization

Storage ConfigurationNVMe SSD (225 GB)

Local NVMe Disk225 GB NVMe SSD

EBS Baseline Throughput593.75 MB/s

EBS Max Throughput593.75 MB/s

EBS Baseline IOPS20,000

EBS Max / Burst IOPS20,000

Dedicated EBS Throughput3500 Mbps

GPU & Hardware Acceleration

GPU ModelNVIDIA T4

Accelerator Count1x GPU(s)

Total GPU Memory16.0 GB

Memory Architecture—

Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)

Nitro EnclavesSupported (Isolated security enclaves)

Normalization Factor2.288973384

Generation StatusCurrent Generation (Recommended)

Rates by region

Live on-demand rates across all global AWS cloud regions

Lowest Price Region

$1.2040/ hr

US East (N. Virginia) (us-east-1)

Highest Price Region

$2.0460/ hr

South America (Sao Paulo) (sa-east-1)

Global Fleet Average

$1.5365/ hr

Across all active regions

Regional Coverage

34regions

Available worldwide

Linux — Available in all 34 regions
Region	On-Demand Price	% vs On-Demand Rate
af-south-1 Africa (Cape Town)	$1.5970
ap-east-1 Asia Pacific (Hong Kong)	$1.8540
ap-south-1 Asia Pacific (Mumbai)	$1.3250
ap-northeast-3 Asia Pacific (Osaka)	$1.6250
ap-northeast-2 Asia Pacific (Seoul)	$1.4810
ap-southeast-1 Asia Pacific (Singapore)	$1.6850
ap-southeast-2 Asia Pacific (Sydney)	$1.5660
ap-northeast-1 Asia Pacific (Tokyo)	$1.6250
us-gov-east-1 AWS GovCloud (US-East)	$1.5180
us-gov-west-1 AWS GovCloud (US-West)	$1.5180
ca-central-1 Canada (Central)	$1.3370
eu-central-1 EU (Frankfurt)	$1.5050
eu-west-1 EU (Ireland)	$1.3420
eu-west-2 EU (London)	$1.4090
eu-south-1 EU (Milan)	$1.4100
eu-west-3 EU (Paris)	$1.4080
eu-north-1 EU (Stockholm)	$1.2770
us-east-1-qro-1 Mexico (Queretaro)	$1.7460
me-south-1 Middle East (Bahrain)	$1.4770
sa-east-1 South America (Sao Paulo) Highest	$2.0460
us-east-1-bos-1 US East (Boston)	$1.6250
us-east-1-chi-1 US East (Chicago)	$1.6250
us-east-1-dfw-1 US East (Dallas)	$1.6250
us-east-1-iah-1 US East (Houston)	$1.6250
us-east-1 US East (N. Virginia) Lowest	$1.2040
us-east-1-nyc-1 US East (New York City)	$1.6250
us-east-2 US East (Ohio) Lowest	$1.2040
us-east-1-phl-1 US East (Philadelphia)	$1.6250
us-west-2-den-1 US West (Denver)	$1.6250
us-west-2-hnl-1 US West (Honolulu)	$1.9870
us-west-2-lax-1 US West (Los Angeles)	$1.4450
us-west-1 US West (N. California)	$1.4450
us-west-2 US West (Oregon) Lowest	$1.2040
us-west-2-phx-1 US West (Phoenix)	$1.6250

Windows — Available in all 34 regions
Region	On-Demand Price	% vs On-Demand Rate
af-south-1 Africa (Cape Town)	$2.3330
ap-east-1 Asia Pacific (Hong Kong)	$2.5900
ap-south-1 Asia Pacific (Mumbai)	$2.0610
ap-northeast-3 Asia Pacific (Osaka)	$2.3610
ap-northeast-2 Asia Pacific (Seoul)	$2.2170
ap-southeast-1 Asia Pacific (Singapore)	$2.4210
ap-southeast-2 Asia Pacific (Sydney)	$2.3020
ap-northeast-1 Asia Pacific (Tokyo)	$2.3610
us-gov-east-1 AWS GovCloud (US-East)	$2.2540
us-gov-west-1 AWS GovCloud (US-West)	$2.2540
ca-central-1 Canada (Central)	$2.0730
eu-central-1 EU (Frankfurt)	$2.2410
eu-west-1 EU (Ireland)	$2.0780
eu-west-2 EU (London)	$2.1450
eu-south-1 EU (Milan)	$2.1460
eu-west-3 EU (Paris)	$2.1440
eu-north-1 EU (Stockholm)	$2.0130
us-east-1-qro-1 Mexico (Queretaro)	$2.4820
me-south-1 Middle East (Bahrain)	$2.2130
sa-east-1 South America (Sao Paulo) Highest	$2.7820
us-east-1-bos-1 US East (Boston)	$2.3610
us-east-1-chi-1 US East (Chicago)	$2.3610
us-east-1-dfw-1 US East (Dallas)	$2.3610
us-east-1-iah-1 US East (Houston)	$2.3610
us-east-1 US East (N. Virginia) Lowest	$1.9400
us-east-1-nyc-1 US East (New York City)	$2.3610
us-east-2 US East (Ohio) Lowest	$1.9400
us-east-1-phl-1 US East (Philadelphia)	$2.3610
us-west-2-den-1 US West (Denver)	$2.3610
us-west-2-hnl-1 US West (Honolulu)	$2.7230
us-west-2-lax-1 US West (Los Angeles)	$2.1810
us-west-1 US West (N. California)	$2.1810
us-west-2 US West (Oregon) Lowest	$1.9400
us-west-2-phx-1 US West (Phoenix)	$2.3610

Savings Plans: what to know

A Savings Plan commits you to a dollar amount per hour of spend for one or three years, not to an instance: usage up to that amount is billed at the plan's rates and the rest at on-demand, and unused commitment is still billed.
A Compute Savings Plan follows your usage across instance families, sizes, regions and operating systems; an EC2 Instance Savings Plan is locked to one family in one region in exchange for a lower rate (what that flexibility costs).
In US East (N. Virginia) (us-east-1), the 3-year Compute Savings Plan rate (No Upfront) for g4dn.4xlarge on Linux is 52% of the on-demand rate, so the instance has to run at least 52% of the hours for the commitment to cost less than on-demand.
On Windows a plan discounts only the compute part of the rate: the Windows license part costs the same as on demand, so the percentage off is smaller than on Linux.
Rates are AWS's effective hourly rate for each payment option; All Upfront pays the whole term's commitment at purchase, which is why its hourly rate is lower; Partial Upfront (at least half up front) isn't shown on this page. The Savings Plans calculator has all three and sizes a commitment for this instance or a mix of them. How AWS applies a plan each hour: how Compute Savings Plans decide what they cover. Sizing one for elastic workloads: managing Compute Savings Plans.