(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | aws |
Name | g7.8xlarge |
Description | Graphics intensive Gen7 8xlarge |
Family | g7 |
Hw Virt | - |
Status | active |
Observed At | 2026-11-02T00:23:49.883332 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Oregon (US) / us-west-2 | 0.843825 USD/h | 4.0862 USD/h |
Northern Virgina (US) / us-east-1 | 0.792425 USD/h | 4.0862 USD/h |
Ohio (US) / us-east-2 | 1.841 USD/h | 4.0862 USD/h |
Aragón (ES) / eu-south-2 | 0.4291 USD/h | 4.2905 USD/h |
Stockholm (SE) / eu-north-1 | 0.4334 USD/h | 4.334 USD/h |
Frankfurt (DE) / eu-central-1 | 1.35705 USD/h | 5.1099 USD/h |
Processor
vCPUs | 32 |
Hypervisor | nitro |
CPU Allocation | Dedicated |
CPU Cores | 32 |
CPU Speed | 3.9 GHz |
CPU Architecture | x86_64 |
CPU Manufacturer | Intel |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 128 GiB |
| ACCELERATORS | |
|---|---|
Accelerator Count | 1 |
Accelerator Memory Min | 32 GiB |
Accelerator Memory Total | 32 GiB |
Accelerator Type | GPU |
Accelerator Manufacturer | NVIDIA |
Accelerator Family | Blackwell |
Accelerator Model | RTX PRO 4500 |
Accelerators |
|
| STORAGE | |
|---|---|
Storage Size | 930 GB |
Storage Type | nvme ssd |
Storages |
|
| NETWORK | |
|---|---|
Network Speed Baseline | 100 Gbps |
Network Speed Max | 100 Gbps |
Network Storage Speed Baseline | 16 Gbps |
Network Storage Speed Max | 16 Gbps |
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
High-performance GPU instance featuring NVIDIA Blackwell graphics acceleration, dedicated 32-core Intel CPU, 128 GB RAM, and high-speed 100 Gbps networking.
Amazon Web Services g7.8xlarge is a graphics-intensive GPU server engineered on the Nitro hypervisor. It features a 32-core Intel x86_64 CPU running at 3.9 GHz with dedicated 1:1 physical core mapping and 128 GB of system RAM. Hardware acceleration is delivered by one NVIDIA Blackwell RTX PRO 4500 GPU with 32 GB of VRAM. Storage includes a 930 GB local NVMe SSD, complemented by 100 Gbps baseline network throughput. The balance of 4 GB RAM per dedicated vCPU without hyperthreading overhead ensures predictable compute execution. This instance is suited for 3D graphics rendering, CAD modeling, real-time video processing, and AI model inference.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | Accelerators |
|---|---|---|---|
| g7.2xlarge | 8 | 32 GiB | 1 |
| g7.4xlarge | 16 | 64 GiB | 1 |
| g7.12xlarge | 48 | 192 GiB | 2 |
| g7.24xlarge | 96 | 384 GiB | 4 |
| g7.48xlarge | 192 | 768 GiB | 8 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | Accelerators |
|---|