p6-b300.48xlarge by Amazon Web Services
(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | aws |
Name | p6-b300.48xlarge |
Description | GPU accelerated Gen6 48xlarge |
Family | p6-b300 |
Hw Virt | - |
Status | active |
Observed At | 2026-08-05T20:53:57.583840 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Northern Virgina (US) / us-east-1 | 44.707 USD/h | 142.416 USD/h |
Oregon (US) / us-west-2 | 51.506 USD/h | 142.416 USD/h |
Processor
vCPUs | 192 |
Hypervisor | nitro |
CPU Allocation | Dedicated |
CPU Cores | 96 |
CPU Speed | 2.4 GHz |
CPU Architecture | x86_64 |
CPU Manufacturer | Intel |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 4 TB |
| GPU | |
|---|---|
GPU Count | 8 |
GPU Memory Min | 269 GiB |
GPU Memory Total | 2 TB |
GPU Manufacturer | NVIDIA |
GPU Model | B300 |
GPUs |
|
| STORAGE | |
|---|---|
Storage Size | 30400 GB |
Storage Type | nvme ssd |
Storages |
|
| NETWORK | |
|---|---|
Network Speed Baseline | 350 Gbps |
Network Speed Max | 350 Gbps |
Network Storage Speed Baseline | 100 Gbps |
Network Storage Speed Max | 100 Gbps |
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
An ultra-dense GPU instance featuring eight accelerators, massive system memory, and high-capacity local NVMe storage for complex parallel computing workloads.
Amazon Web Services p6-b300.48xlarge is a high-performance, GPU-accelerated instance designed for intensive computational workloads. Built on the AWS Nitro hypervisor, it features 192 vCPUs across 96 physical Intel cores running at 2.4 GHz, paired with 4096.0 GB of system memory. Acceleration is driven by eight NVIDIA B300 GPUs providing a total of 2149 GB of VRAM. For high-speed data access, the instance includes 30400 GB of local NVMe SSD storage and a baseline network bandwidth of 350 Gbps. This dense resource allocation makes the instance highly efficient for large-scale parallel processing without the latency of multi-node clustering. It is optimized for demanding workloads such as large language model training, deep learning, high-performance computing, and massive database management.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | GPUs |
|---|
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | GPUs |
|---|