Compare instance

inf1.6xlarge

⚡ 4x 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.6xlarge6 Extra Large6xlarge = 6 Extra Large24 vCPUs • 48 GB RAM • EBS only • 25 Gbps Network • 4x GPU
Price (Linux)
$1.1800/ hr
~$861.40/mo
Compute
24vCPUs
12 cores • x86_64
Memory
48GB
2.0 GB / vCPU
Value
$0.0492/ vCPU-hr
$0.0246/GB-hr
Network
25 Gbps
Baseline: 25 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 / Threads24 vCPUs
Physical Cores12
Clock Speed—
Intel AVX / AVX2No
Intel Turbo BoostNo
Bare MetalNo (Virtualized)

Performance & Benchmarks

CoreMark Score382,638
CoreMark / vCPU15,943.2 / core
FFmpeg Transcoding137 FPS
CoreMark per Dollar324,269 score / $
Cost per vCPU-hr$0.0492
Cost per GB-hr$0.0246

Containers & Kubernetes

EKS Max Pods234 pods
Max ECS Tasks60 tasks
Max ENIs8
IPv4/v6 per ENI30
Total Addressable IPs240 total IPs
CNI Prefix DelegationSupported (High Density)

Networking & Bandwidth

Network Performance25 Gbps
Baseline Bandwidth25 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 Throughput593.75 MB/s
EBS Max Throughput593.75 MB/s
EBS Baseline IOPS20,000
EBS Max / Burst IOPS20,000
Dedicated EBS Throughput3500 Mbps

GPU & Hardware Acceleration

GPU ModelNVIDIA Accelerator
Accelerator Count4x 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
$1.1800/ hr
US East (N. Virginia) (us-east-1)
Highest Price Region
$1.9500/ hr
South America (Sao Paulo) (sa-east-1)
Global Fleet Average
$1.4290/ hr
Across all active regions
Regional Coverage
21regions
Available worldwide
Linux — Available in all 21 regions
RegionOn-Demand Price
ap-east-1
Asia Pacific (Hong Kong)
$1.8190
ap-south-1
Asia Pacific (Mumbai)
$1.2410
ap-northeast-2
Asia Pacific (Seoul)
$1.4530
ap-southeast-1
Asia Pacific (Singapore)
$1.5940
ap-southeast-2
Asia Pacific (Sydney)
$1.4750
ap-northeast-1
Asia Pacific (Tokyo)
$1.5940
us-gov-east-1
AWS GovCloud (US-East)
$1.4880
us-gov-west-1
AWS GovCloud (US-West)
$1.4880
ca-central-1
Canada (Central)
$1.3150
eu-central-1
EU (Frankfurt)
$1.4750
eu-west-1
EU (Ireland)
$1.3150
eu-west-2
EU (London)
$1.3820
eu-south-1
EU (Milan)
$1.3820
eu-west-3
EU (Paris)
$1.3790
eu-north-1
EU (Stockholm)
$1.2540
me-south-1
Middle East (Bahrain)
$1.4470
sa-east-1
South America (Sao Paulo)
Highest
$1.9500
us-east-1
US East (N. Virginia)
Lowest
$1.1800
us-east-2
US East (Ohio)
Lowest
$1.1800
us-west-1
US West (N. California)
$1.4180
us-west-2
US West (Oregon)
Lowest
$1.1800