g7e.48xlarge
⚡ 8x GPU • 768.0 GBPart of the G7E Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.
Name:
gGPU Graphics & ML Inferenceg = GPU Graphics & ML InferencePowered by GPUs (NVIDIA/AMD) for graphics-intensive applications, video encoding, remote workstations, and machine learning inference.77th Generation7 = 7th GenerationAWS 7th generation hardware platform with updated CPU/system architecture.eExtra Capacity / Expansione = Extra Capacity / ExpansionExpanded configuration offering a higher memory-to-vCPU ratio or increased local NVMe storage density per core.48xlarge48 Extra Large48xlarge = 48 Extra Large192 vCPUs • 2048.0 GiB RAM • 15200 GB NVMe SSD • 1600 Gigabit Network • 8x NVIDIA RTX PRO 6000 Blackwell (768.0 GB VRAM)Price (Linux)
$33.1443/ hr
~$24,195.35/mo
Compute
192vCPUs
96 cores • x86_64
Memory
2048.0GiB
10.7 GiB / vCPU
Value
$0.1726/ vCPU-hr
$0.0162/GiB-hr
Network
400Gbps
Baseline: 400 Gbps
Storage & EBS
12,500MB/s
Max: 400,000 IOPS
Technical Specifications & Limits
Hardware specs, container thresholds, and system limitsCompute & Processor
ProcessorIntel Xeon Scalable (Emerald Rapids)
Architecturex86_64 (x86_64)
vCPUs / Threads192 vCPUs
Physical Cores96
Clock Speed3.2 GHz
Intel AVX / AVX2No
Intel Turbo BoostNo
Bare MetalNo (Virtualized)
Performance & Benchmarks
CoreMark Score4,097,020
CoreMark / vCPU21,338.6 / core
FFmpeg Transcoding122 FPS
CoreMark per Dollar123,612 score / $
Cost per vCPU-hr$0.1726
Cost per GiB-hr$0.0162
Containers & Kubernetes
EKS Max Pods632 pods
Max ECS Tasks158 tasks
Max ENIs40
IPv4/v6 per ENI64
Total Addressable IPs2560 total IPs
CNI Prefix DelegationSupported (High Density)
Networking & Bandwidth
Network Performance1600 Gigabit
Baseline Bandwidth400 Gbps
Burst / Peak Bandwidth400 Gbps
Enhanced NetworkingYes
EFA SupportNo
IPv6 SupportNative Dual-Stack Supported
Storage & EBS Optimization
Storage ConfigurationNVMe SSD (15200 GB)
Local NVMe Disk15200 GB
EBS Baseline Throughput12500 MB/s
EBS Max Throughput12500 MB/s
EBS Baseline IOPS400,000
EBS Max / Burst IOPS400,000
Dedicated EBS Throughput100000 Mbps
GPU & Hardware Acceleration
GPU ModelNVIDIA RTX PRO 6000 Blackwell
Accelerator Count8x GPU(s)
Total GPU Memory768.0 GB
Memory ArchitectureGDDR / HBM
CUDA Compute Capabilityv12.0
Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX
Platform & Architecture Details
Hypervisor / SystemAWS Nitro System (PCIe Offload)
Nitro EnclavesSupported (Isolated security enclaves)
Normalization Factor9.855229668
Generation StatusCurrent Generation (Recommended)
Date Introduced2026-01-20
Regional Pricing Matrix
Live on-demand rates across all global AWS cloud regionsLowest Price Region
$33.1443/ hr
US East (N. Virginia) (us-east-1)Highest Price Region
$57.6069/ hr
EU (London) (eu-west-2)Global Fleet Average
$43.9095/ hr
Across all active regionsRegional Coverage
10regions
Available worldwide| Region Name | Region Code | On-Demand Price |
|---|---|---|
Asia Pacific (Mumbai) | ap-south-1 | $54.1351 |
Asia Pacific (Seoul) | ap-northeast-2 | $40.7492 |
Asia Pacific (Tokyo) | ap-northeast-1 | $48.0691 |
EU (Frankfurt) | eu-central-1 | $56.3635 |
EU (London)Highest | eu-west-2 | $57.6069 |
EU (Stockholm) | eu-north-1 | $47.8133 |
Europe (Spain) | eu-south-2 | $34.9254 |
US East (N. Virginia)Lowest | us-east-1 | $33.1443 |
US East (Ohio)Lowest | us-east-2 | $33.1443 |
US West (Oregon)Lowest | us-west-2 | $33.1443 |
| Region Name | Region Code | On-Demand Price |
|---|---|---|
Asia Pacific (Mumbai) | ap-south-1 | $62.9671 |
Asia Pacific (Seoul) | ap-northeast-2 | $49.5812 |
Asia Pacific (Tokyo) | ap-northeast-1 | $56.9011 |
EU (Frankfurt) | eu-central-1 | $65.1955 |
EU (London)Highest | eu-west-2 | $66.4389 |
EU (Stockholm) | eu-north-1 | $56.6453 |
Europe (Spain) | eu-south-2 | $43.7574 |
US East (N. Virginia)Lowest | us-east-1 | $41.9763 |
US East (Ohio)Lowest | us-east-2 | $41.9763 |
US West (Oregon)Lowest | us-west-2 | $41.9763 |
No regions matched your search filter.