← All instances

trn1.2xlarge

⚡ 1x GPU • 32.0 GB

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

Name:trnAWS Trainium Acceleratorstrn = AWS Trainium AcceleratorsPurpose-built AWS Trainium silicon engineered for high-performance, cost-effective deep learning model training.11st Generation1 = 1st GenerationAWS 1st generation hardware platform with updated CPU/system architecture.2xlarge2 Extra Large2xlarge = 2 Extra Large8 vCPUs • 32.0 GiB RAM • 475 GB NVMe SSD • 12500 Megabit Network • 1x GPU (32.0 GB VRAM)
Price (Linux)
$1.3438/ hr
~$980.94/mo
Compute
8vCPUs
4 cores • x86_64
Memory
32.0GiB
4.0 GiB / vCPU
Value
$0.1680/ vCPU-hr
$0.0420/GiB-hr
Network
12.5Gbps
Baseline: 3.125 Gbps
Storage & EBS
2,500MB/s
Max: 65,000 IOPS

Technical Specifications & Limits

Hardware specs, container thresholds, and system limits

Compute & Processor

ProcessorIntel Xeon Scalable (Icelake)
Architecturex86_64 (x86_64)
vCPUs / Threads8 vCPUs
Physical Cores4
Clock SpeedUp to 3.5 GHz
Intel AVX / AVX2AVX & AVX2 Supported
Intel Turbo BoostYes
Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score143,698
CoreMark / vCPU17,962.2 / core
FFmpeg Transcoding56 FPS
CoreMark per Dollar106,938 score / $
Cost per vCPU-hr$0.1680
Cost per GiB-hr$0.0420

Containers & Kubernetes

EKS Max Pods58 pods
Max ECS Tasks19 tasks
Max ENIs4
IPv4/v6 per ENI15
Total Addressable IPs60 total IPs
CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network Performance12500 Megabit
Baseline Bandwidth3.125 Gbps
Burst / Peak Bandwidth12.5 Gbps
Enhanced NetworkingYes
EFA SupportNo
IPv6 SupportNative Dual-Stack Supported

Storage & EBS Optimization

Storage ConfigurationNVMe SSD (475 GB)
Local NVMe Disk475 GB
EBS Baseline Throughput625 MB/s
EBS Max Throughput2500 MB/s
EBS Baseline IOPS16,250
EBS Max / Burst IOPS65,000
Dedicated EBS ThroughputUp to 20 Gbps

GPU & Hardware Acceleration

GPU ModelNVIDIA Accelerator
Accelerator Count1x GPU(s)
Total GPU Memory32.0 GB
Memory ArchitectureGDDR / HBM
CUDA Compute Capability—
Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)
Nitro EnclavesNot Supported
Normalization Factor16
Generation StatusCurrent Generation (Recommended)
Date Introduced2022-10-10

Regional Pricing Matrix

Live on-demand rates across all global AWS cloud regions
Lowest Price Region
$1.3438/ hr
US East (N. Virginia) (us-east-1)
Highest Price Region
$1.3438/ hr
US East (N. Virginia) (us-east-1)
Global Fleet Average
$1.3438/ hr
Across all active regions
Regional Coverage
3regions
Available worldwide
Linux — Available in all 3 regions
Region NameRegion CodeOn-Demand Price
US East (N. Virginia)Lowest
us-east-1$1.3438
US East (Ohio)Lowest
us-east-2$1.3438
US West (Oregon)Lowest
us-west-2$1.3438