← All instances

inf1.2xlarge

⚡ 1x GPU

Part of the INF1 Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.

Name:infAWS Inferentia Acceleratorsinf = AWS Inferentia AcceleratorsPurpose-built AWS Inferentia silicon designed for ultra-low latency, high-throughput machine learning inference.11st Generation1 = 1st GenerationAWS 1st generation hardware platform with updated CPU/system architecture.2xlarge2 Extra Large2xlarge = 2 Extra Large8 vCPUs • 16.0 GiB RAM • EBS only • Up to 25 Gigabit Network • 1x GPU
Price (Linux)
$0.3620/ hr
~$264.26/mo
Compute
8vCPUs
4 cores • x86_64
Memory
16.0GiB
2.0 GiB / vCPU
Value
$0.0452/ vCPU-hr
$0.0226/GiB-hr
Network
25Gbps
Baseline: 5 Gbps
Storage & EBS
593.75MB/s
Max: 20,000 IOPS

Technical Specifications & Limits

Hardware specs, container thresholds, and system limits

Compute & Processor

ProcessorIntel Xeon Platinum 8275CL (Cascade Lake)
Architecturex86_64 (x86_64)
vCPUs / Threads8 vCPUs
Physical Cores4
Clock Speed—
Intel AVX / AVX2No
Intel Turbo BoostNo
Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score131,628
CoreMark / vCPU16,453.5 / core
FFmpeg Transcoding47 FPS
CoreMark per Dollar363,613 score / $
Cost per vCPU-hr$0.0452
Cost per GiB-hr$0.0226

Containers & Kubernetes

EKS Max Pods38 pods
Max ECS Tasks40 tasks
Max ENIs4
IPv4/v6 per ENI10
Total Addressable IPs40 total IPs
CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network PerformanceUp to 25 Gigabit
Baseline Bandwidth5 Gbps
Burst / Peak Bandwidth25 Gbps
Enhanced NetworkingNo
EFA SupportNo
IPv6 SupportNative Dual-Stack Supported

Storage & EBS Optimization

Storage ConfigurationEBS only
Local NVMe DiskNone (EBS only)
EBS Baseline Throughput148.75 MB/s
EBS Max Throughput593.75 MB/s
EBS Baseline IOPS6,000
EBS Max / Burst IOPS20,000
Dedicated EBS Throughput875 Mbps

GPU & Hardware Acceleration

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

Platform & Architecture Details

Hypervisor / SystemAWS Nitro System (PCIe Offload)
Nitro EnclavesSupported (Isolated security enclaves)
Normalization FactorNA
Generation StatusCurrent Generation (Recommended)
Date Introduced2019-12-03

Regional Pricing Matrix

Live on-demand rates across all global AWS cloud regions
Lowest Price Region
$0.3620/ hr
US East (N. Virginia) (us-east-1)
Highest Price Region
$0.5980/ hr
South America (Sao Paulo) (sa-east-1)
Global Fleet Average
$0.4384/ hr
Across all active regions
Regional Coverage
21regions
Available worldwide
Linux — Available in all 21 regions
Region NameRegion CodeOn-Demand Price
Asia Pacific (Hong Kong)
ap-east-1$0.5580
Asia Pacific (Mumbai)
ap-south-1$0.3810
Asia Pacific (Seoul)
ap-northeast-2$0.4460
Asia Pacific (Singapore)
ap-southeast-1$0.4890
Asia Pacific (Sydney)
ap-southeast-2$0.4530
Asia Pacific (Tokyo)
ap-northeast-1$0.4890
AWS GovCloud (US-East)
us-gov-east-1$0.4560
AWS GovCloud (US-West)
us-gov-west-1$0.4560
Canada (Central)
ca-central-1$0.4030
EU (Frankfurt)
eu-central-1$0.4530
EU (Ireland)
eu-west-1$0.4030
EU (London)
eu-west-2$0.4240
EU (Milan)
eu-south-1$0.4240
EU (Paris)
eu-west-3$0.4230
EU (Stockholm)
eu-north-1$0.3850
Middle East (Bahrain)
me-south-1$0.4440
South America (Sao Paulo)Highest
sa-east-1$0.5980
US East (N. Virginia)Lowest
us-east-1$0.3620
US East (Ohio)Lowest
us-east-2$0.3620
US West (N. California)
us-west-1$0.4350
US West (Oregon)Lowest
us-west-2$0.3620