AWS Amazon EC2

g6e.48xlarge

⚡ 8x GPU • 384 GB

Part of the G6E Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.

Name:gGPU Graphics & ML Inference66th GenerationeExtra Capacity / Expansion48xlarge48 Extra Large

On-demand, Linux

$30.1312/ hr

~$21,995.76/mo · lowest in US East (N. Virginia), us-east-1

Savings Plan, Linux

$15.6200/ hr

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

Per unit

$0.1569/ vCPU-hr

$0.0196/GB-hr

Compute

192vCPUs

96 cores • x86_64

Memory

1,536GB

8.0 GB / vCPU

Network

400 Gbps

Baseline: 400 Gbps

Storage & EBS

7,500MB/s

Max: 240,000 IOPS

On Google Cloud

g6e.48xlarge 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

ProcessorAMD EPYC 7R13 Processor

Architecturex86_64 (x86_64)

vCPUs / Threads192 vCPUs

Physical Cores96

Clock Speed2.6 GHz

Intel AVX / AVX2No

Intel Turbo BoostNo

Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score3,635,503

CoreMark / vCPU18,934.9 / core

FFmpeg Transcoding142 FPS

CoreMark per Dollar120,656 score / $

Cost per vCPU-hr$0.1569

Cost per GB-hr$0.0196

Containers & Kubernetes

EKS Max Pods492 pods

Max ENIs40

IPv4/v6 per ENI50

Total Addressable IPs2000 total IPs

CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network Performance400 Gbps

Baseline Bandwidth400 Gbps

Burst / Peak Bandwidth400 Gbps

Enhanced NetworkingYes

EFA SupportNo

IPv6 SupportNative Dual-Stack Supported

Storage & EBS Optimization

Storage ConfigurationNVMe SSD (7600 GB)

Local NVMe Disk7,600 GB NVMe SSD

EBS Baseline Throughput7500 MB/s

EBS Max Throughput7500 MB/s

EBS Baseline IOPS240,000

EBS Max / Burst IOPS240,000

Dedicated EBS Throughput60000 Mbps

GPU & Hardware Acceleration

GPU ModelNVIDIA L40S

Accelerator Count8x GPU(s)

Total GPU Memory384.0 GB

Memory Architecture—

Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)

Nitro EnclavesSupported (Isolated security enclaves)

Normalization Factor16.19085487

Generation StatusCurrent Generation (Recommended)

Rates by region

Live on-demand rates across all global AWS cloud regions

Lowest Price Region

$30.1312/ hr

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

Highest Price Region

$43.6991/ hr

Asia Pacific (Tokyo) (ap-northeast-1)

Global Fleet Average

$35.4634/ hr

Across all active regions

Regional Coverage

12regions

Available worldwide

Linux — Available in all 12 regions
Region	On-Demand Price	% vs On-Demand Rate
ap-south-2 Asia Pacific (Hyderabad)	$36.1866
ap-south-1 Asia Pacific (Mumbai)	$36.1866
ap-northeast-3 Asia Pacific (Osaka) Highest	$43.6991
ap-northeast-2 Asia Pacific (Seoul)	$37.0447
ap-northeast-1 Asia Pacific (Tokyo) Highest	$43.6991
eu-central-1 EU (Frankfurt)	$37.6761
eu-north-1 EU (Stockholm)	$31.9608
eu-south-2 Europe (Spain)	$31.7503
me-central-1 Middle East (UAE)	$36.9637
us-east-1 US East (N. Virginia) Lowest	$30.1312
us-east-2 US East (Ohio) Lowest	$30.1312
us-west-2 US West (Oregon) Lowest	$30.1312

Windows — Available in all 12 regions
Region	On-Demand Price	% vs On-Demand Rate
ap-south-2 Asia Pacific (Hyderabad)	$45.0186
ap-south-1 Asia Pacific (Mumbai)	$45.0186
ap-northeast-3 Asia Pacific (Osaka) Highest	$52.5311
ap-northeast-2 Asia Pacific (Seoul)	$45.8767
ap-northeast-1 Asia Pacific (Tokyo) Highest	$52.5311
eu-central-1 EU (Frankfurt)	$46.5081
eu-north-1 EU (Stockholm)	$40.7927
eu-south-2 Europe (Spain)	$40.5823
me-central-1 Middle East (UAE)	$45.7957
us-east-1 US East (N. Virginia) Lowest	$38.9632
us-east-2 US East (Ohio) Lowest	$38.9632
us-west-2 US West (Oregon) Lowest	$38.9632

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 g6e.48xlarge 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.