AWS Amazon EC2

g7e.8xlarge

⚡ 1x GPU • 96 GB

Part of the G7E Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.

Name:gGPU Graphics & ML Inference77th GenerationeExtra Capacity / Expansion8xlarge8 Extra Large

On-demand, Linux

$5.2682/ hr

~$3,845.82/mo · lowest in US East (N. Virginia), us-east-1

Savings Plan, Linux

$3.07244/ hr

From, with a 3-year Compute Savings Plan, No Upfront: −42% vs on-demand rate (us-east-1)

Per unit

$0.1646/ vCPU-hr

$0.0206/GB-hr

Compute

32vCPUs

16 cores • x86_64

Memory

256GB

8.0 GB / vCPU

Network

100 Gbps

Baseline: 100 Gbps

Storage & EBS

2,000MB/s

Max: 64,000 IOPS

On Google Cloud

g7e.8xlarge 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 Scalable (Emerald Rapids)

Architecturex86_64 (x86_64)

vCPUs / Threads32 vCPUs

Physical Cores16

Clock Speed3.2 GHz

Intel AVX / AVX2No

Intel Turbo BoostNo

Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score852,916

CoreMark / vCPU26,653.6 / core

FFmpeg Transcoding153 FPS

CoreMark per Dollar161,898 score / $

Cost per vCPU-hr$0.1646

Cost per GB-hr$0.0206

Containers & Kubernetes

EKS Max Pods506 pods

Max ENIs8

IPv4/v6 per ENI64

Total Addressable IPs512 total IPs

CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network Performance100 Gbps

Baseline Bandwidth100 Gbps

Burst / Peak Bandwidth100 Gbps

Enhanced NetworkingYes

EFA SupportNo

IPv6 SupportNative Dual-Stack Supported

Storage & EBS Optimization

Storage ConfigurationNVMe SSD (1900 GB)

Local NVMe Disk1,900 GB NVMe SSD

EBS Baseline Throughput2000 MB/s

EBS Max Throughput2000 MB/s

EBS Baseline IOPS64,000

EBS Max / Burst IOPS64,000

Dedicated EBS Throughput16000 Mbps

GPU & Hardware Acceleration

GPU ModelNVIDIA RTX PRO Server 6000

Accelerator Count1x GPU(s)

Total GPU Memory96.0 GB

Memory Architecture—

Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)

Nitro EnclavesSupported (Isolated security enclaves)

Normalization Factor1.566473988

Generation StatusCurrent Generation (Recommended)

Rates by region

Live on-demand rates across all global AWS cloud regions

Lowest Price Region

$5.2682/ hr

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

Highest Price Region

$10.0292/ hr

Asia Pacific (Jakarta) (ap-southeast-3)

Global Fleet Average

$7.4953/ hr

Across all active regions

Regional Coverage

14regions

Available worldwide

Linux — Available in all 14 regions
Region	On-Demand Price	% vs On-Demand Rate
ap-southeast-3 Asia Pacific (Jakarta) Highest	$10.0292
ap-southeast-5 Asia Pacific (Malaysia)	$9.0254
ap-south-1 Asia Pacific (Mumbai)	$8.6047
ap-northeast-2 Asia Pacific (Seoul)	$6.4770
ap-southeast-7 Asia Pacific (Thailand)	$9.0254
ap-northeast-1 Asia Pacific (Tokyo)	$7.6405
eu-central-1 EU (Frankfurt)	$8.9589
eu-west-2 EU (London)	$9.1565
eu-north-1 EU (Stockholm)	$7.5999
eu-south-2 Europe (Spain)	$5.5513
us-east-1 US East (N. Virginia) Lowest	$5.2682
us-east-2 US East (Ohio) Lowest	$5.2682
us-west-2-lax-1 US West (Los Angeles)	$7.0601
us-west-2 US West (Oregon) Lowest	$5.2682

Windows — Available in all 14 regions
Region	On-Demand Price	% vs On-Demand Rate
ap-southeast-3 Asia Pacific (Jakarta) Highest	$11.5012
ap-southeast-5 Asia Pacific (Malaysia)	$10.4974
ap-south-1 Asia Pacific (Mumbai)	$10.0767
ap-northeast-2 Asia Pacific (Seoul)	$7.9490
ap-southeast-7 Asia Pacific (Thailand)	$10.4974
ap-northeast-1 Asia Pacific (Tokyo)	$9.1125
eu-central-1 EU (Frankfurt)	$10.4309
eu-west-2 EU (London)	$10.6285
eu-north-1 EU (Stockholm)	$9.0719
eu-south-2 Europe (Spain)	$7.0233
us-east-1 US East (N. Virginia) Lowest	$6.7402
us-east-2 US East (Ohio) Lowest	$6.7402
us-west-2-lax-1 US West (Los Angeles)	$8.5321
us-west-2 US West (Oregon) Lowest	$6.7402

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 g7e.8xlarge on Linux is 58% of the on-demand rate, so the instance has to run at least 58% 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.