AWS Amazon EC2

trn1.32xlarge

⚡ 16x GPU • 512 GB

Part of the TRN1 Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.

Name:trnAWS Trainium Accelerators11st Generation32xlarge32 Extra Large

On-demand, Linux

$21.5000/ hr

~$15,695.00/mo · lowest in US East (N. Virginia), us-east-1

Savings Plan, Linux

$9.84528/ hr

From, with a 3-year Compute Savings Plan, No Upfront: −54% vs on-demand rate (us-east-1)

Per unit

$0.1680/ vCPU-hr

$0.0420/GB-hr

Compute

128vCPUs

64 cores • x86_64

Memory

512GB

4.0 GB / vCPU

Network

800 Gbps

Baseline: 800 Gbps

Storage & EBS

10,000MB/s

Max: 260,000 IOPS

On Google Cloud

trn1.32xlarge 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 (Icelake)

Architecturex86_64 (x86_64)

vCPUs / Threads128 vCPUs

Physical Cores64

Clock SpeedUp to 3.5 GHz

Intel AVX / AVX2AVX & AVX2 Supported

Intel Turbo BoostYes

Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score2,177,889

CoreMark / vCPU17,014.8 / core

FFmpeg Transcoding221 FPS

CoreMark per Dollar101,297 score / $

Cost per vCPU-hr$0.1680

Cost per GB-hr$0.0420

Containers & Kubernetes

EKS Max Pods247 pods

Max ENIs40

IPv4/v6 per ENI50

Total Addressable IPs2000 total IPs

CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network Performance800 Gbps

Baseline Bandwidth800 Gbps

Burst / Peak Bandwidth800 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 Throughput10000 MB/s

EBS Max Throughput10000 MB/s

EBS Baseline IOPS260,000

EBS Max / Burst IOPS260,000

Dedicated EBS ThroughputUp to 80 Gbps

GPU & Hardware Acceleration

GPU Model—

Accelerator Count16x GPU(s)

Total GPU Memory512.0 GB

Memory Architecture—

Acceleration Runtimes—

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)

Nitro EnclavesNot Supported

Normalization Factor256

Generation StatusCurrent Generation (Recommended)

Rates by region

Live on-demand rates across all global AWS cloud regions

Lowest Price Region

$21.5000/ hr

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

Highest Price Region

$21.5000/ hr

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

Global Fleet Average

$21.5000/ hr

Across all active regions

Regional Coverage

3regions

Available worldwide

Linux — Available in all 3 regions
Region	On-Demand Price	% vs On-Demand Rate
us-east-1 US East (N. Virginia) Lowest	$21.5000
us-east-2 US East (Ohio) Lowest	$21.5000
us-west-2 US West (Oregon) Lowest	$21.5000

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 trn1.32xlarge on Linux is 46% of the on-demand rate, so the instance has to run at least 46% of the hours for the commitment to cost less than on-demand.
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.