p5.4xlarge
Amazon EC2 · P5 family
The p5.4xlarge is a gpu & accelerated Amazon EC2 instance type with 16 vCPUs and 256 GiB of memory, 1 × NVIDIA H100, running on AMD EPYC 7R13 Processor with 100 Gigabit networking, starting at $6.88 per hour on demand.
- vCPUs
- 16
- memory
- 256 GiB
- gpu
- 1× H100
- on-demand
- $6.88
- spot
- –
- regions
- 12
dedicated
16 GiB per vCPU
80 GiB total
$5,022/mo · us-east-1
not offered
across 5 areas
p5.4xlarge price by region
12 regions · hourly list prices in USD · On-demand · Spot · 1-yr Reserved · 3-yr Reserved
| US East (N. Virginia)us-east-1cheapestreferenceNorth America | $6.88 | $2.972 |
| US East (Ohio)us-east-2cheapestNorth America | $6.88 | $2.972 |
| US West (Oregon)us-west-2cheapestNorth America | $6.88 | $2.972 |
| EU (Stockholm)eu-north-1Europe | $7.362 | – |
| Asia Pacific (Mumbai)ap-south-1Asia | $8.256 | $3.567 |
| AWS GovCloud (US)us-gov-west-1North America | $8.256 | – |
| Asia Pacific (Tokyo)ap-northeast-1Asia | $8.60 | $3.715 |
| EU (Frankfurt)eu-central-1Europe | $8.60 | – |
| Asia Pacific (Sydney)ap-southeast-2Oceania | $8.944 | $3.864 |
| EU (London)eu-west-2Europe | $8.944 | $3.864 |
| Asia Pacific (Jakarta)ap-southeast-3Asia | $9.494 | $4.102 |
| South America (Sao Paulo)sa-east-1South America | $11.56 | $4.993 |
Specifications
compute
- vCPUs
- 16
- Memory
- 256 GiB
- Memory per vCPU
- 16 GiB
- Processor
- AMD EPYC 7R13 Processor
- CPU vendor
- AMD
- Architecture
- x86-64
- Clock speed
- 2.95 GHz
- Burstable CPU
- No
storage & network
- Local storage
- 3.8 TB SSD
- Storage detail
- 1 x 3840 GB SSD
- Network
- 100 Gigabit
- Peak bandwidth
- 100 Gbps
accelerators
- Model
- NVIDIA H100
- Count
- 1
- Memory (total)
- 80 GiB
- Architecture
- Hopper
- Vendor
- NVIDIA
aws details
- API name
p5.4xlarge- Family
- P5
- Category
- GPU & accelerated
- Generation
- Current
- Regions
- 12 (Asia, Europe, North America, Oceania, South America)
- Cheapest region
- US East (N. Virginia) · $6.88/hr
- Dedicated EBS throughput
- 10 Gbps
- Enhanced networking
- Yes
- Normalization size factor
- 48
- Processor features
- AMD Turbo; AVX; AVX2
- Spot interruption rate (us-east-1)
- <5%
- Spot savings vs on-demand (us-east-1)
- 9%
p5.4xlarge equivalents on other clouds
closest match by vCPU, memory, CPU architecture, accelerator and class · Linux on-demand in each provider's reference region
| Provider | Instance | vCPU | Memory | Processor | On-demand | vs p5.4xlarge |
|---|---|---|---|---|---|---|
| AWS | p5.4xlarge | 16 | 256 GiB | 1 × NVIDIA H100 | $6.88 | – |
| Google Cloud | a3-highgpu-1g26 vCPU (1.63×) · 234 GiB (1.1× less) | 26 | 234 GiB | 1 × NVIDIA H100 | $11.06 | +61% |
P5 family sizes
| Size | vCPU | Memory | Accelerator | Local storage | On-demand | Monthly | Spot |
|---|---|---|---|---|---|---|---|
| p5.4xlarge | 16 | 256 GiB | 1 × NVIDIA H100 | 3.8 TB SSD | $6.88 | $5,022 | – |
| p5.48xlarge | 192 | 2 TiB | 8 × NVIDIA H100 | 30.7 TB SSD (8 disks) | $55.04 | $40,179 | $57.76 |
p5.4xlarge FAQ
How much does p5.4xlarge cost?
p5.4xlarge costs $6.88 per hour on demand with Linux in US East (N. Virginia) (us-east-1), about $5,022 per month. Across 12 regions the Linux on-demand price ranges from $6.88 in US East (N. Virginia) to $11.56 in South America (Sao Paulo). With Windows it costs $7.616 per hour.
How many vCPUs and how much memory does p5.4xlarge have?
p5.4xlarge has 16 vCPUs and 256 GiB of memory (16 GiB per vCPU), running on AMD EPYC 7R13 Processor. It is a gpu & accelerated instance type in the P5 family.
What GPU does p5.4xlarge have?
p5.4xlarge has 1 × NVIDIA H100 with 80 GiB of accelerator memory in total.
What are the equivalents of p5.4xlarge on other clouds?
The closest matches by vCPU, memory, CPU architecture, accelerator and instance class are on Google Cloud, a3-highgpu-1g (26 vCPU, 234 GiB, $11.06/hour in Iowa).