inf2.48xlarge
Amazon EC2 · Inf2 family
The inf2.48xlarge is a gpu & accelerated Amazon EC2 instance type with 192 vCPUs and 768 GiB of memory, 12 × AWS Inferentia2, running on AMD EPYC 7R13 Processor with 100 Gigabit networking, starting at $12.98 per hour on demand.
- vCPUs
- 192
- memory
- 768 GiB
- accelerator
- 12× Inferentia2
- on-demand
- $12.98
- spot
- $5.048
- 1-yr reserved
- $8.178
dedicated
4 GiB per vCPU
384 GiB total
$9,476/mo · us-east-1
−61% vs on-demand
−37% vs on-demand
inf2.48xlarge price by region
13 regions · hourly list prices in USD · On-demand · Spot · 1-yr Reserved · 3-yr Reserved
| US East (N. Virginia)us-east-1cheapestreferenceNorth America | $12.98 | $5.048 | $8.178 | $5.608 |
| US East (Ohio)us-east-2cheapestNorth America | $12.98 | $2.045 | $8.178 | $5.608 |
| US West (Oregon)us-west-2cheapestNorth America | $12.98 | $1.814 | $8.178 | $5.608 |
| EU (Stockholm)eu-north-1Europe | $14.28 | $1.444 | $8.996 | $6.169 |
| EU (Ireland)eu-west-1Europe | $16.23 | $5.26 | $10.22 | $7.01 |
| Asia Pacific (Mumbai)ap-south-1Asia | $16.88 | $5.074 | $10.63 | $7.851 |
| Asia Pacific (Sydney)ap-southeast-2Oceania | $16.88 | $1.718 | $10.63 | $7.29 |
| Asia Pacific (Singapore)ap-southeast-1Asia | $18.17 | $1.817 | $11.45 | $7.851 |
| EU (Paris)eu-west-3Europe | $18.17 | $1.875 | $11.45 | $8.412 |
| Asia Pacific (Tokyo)ap-northeast-1Asia | $19.47 | $3.273 | $12.27 | $8.973 |
| EU (Frankfurt)eu-central-1Europe | $19.47 | $2.073 | $12.27 | $8.973 |
| EU (London)eu-west-2Europe | $19.47 | $2.272 | $12.27 | $8.973 |
| South America (Sao Paulo)sa-east-1South America | $22.07 | $2.207 | $13.90 | $9.533 |
Specifications
compute
- vCPUs
- 192
- Memory
- 768 GiB
- Memory per vCPU
- 4 GiB
- Processor
- AMD EPYC 7R13 Processor
- CPU vendor
- AMD
- Architecture
- x86-64
- Clock speed
- 2.95 GHz
- Burstable CPU
- No
storage & network
- Local storage
- Network storage only
- Storage detail
- EBS only
- Network
- 100 Gigabit
- Peak bandwidth
- 100 Gbps
accelerators
- Model
- AWS Inferentia2
- Count
- 12
- Memory (total)
- 384 GiB
- Vendor
- AWS
aws details
- API name
inf2.48xlarge- Family
- Inf2
- Category
- GPU & accelerated
- Generation
- Current
- Regions
- 13 (Asia, Europe, North America, Oceania, South America)
- Cheapest region
- US East (N. Virginia) · $12.98/hr
- Dedicated EBS throughput
- 60 Gbps
- Enhanced networking
- Yes
- Normalization size factor
- 17.12
- Processor features
- AMD Turbo; AVX; AVX2
- Spot interruption rate (us-east-1)
- >20%
- Spot savings vs on-demand (us-east-1)
- 65%
Inf2 family sizes
| Size | vCPU | Memory | Accelerator | Local storage | On-demand | Monthly | Spot | 1-yr |
|---|---|---|---|---|---|---|---|---|
| inf2.xlarge | 4 | 16 GiB | 1 × AWS Inferentia2 | – | $0.7582 | $553 | $0.3178 | $0.4777 |
| inf2.8xlarge | 32 | 128 GiB | 1 × AWS Inferentia2 | – | $1.968 | $1,437 | $0.7988 | $1.24 |
| inf2.24xlarge | 96 | 384 GiB | 6 × AWS Inferentia2 | – | $6.491 | $4,738 | $2.445 | $4.089 |
| inf2.48xlarge | 192 | 768 GiB | 12 × AWS Inferentia2 | – | $12.98 | $9,476 | $5.048 | $8.178 |
inf2.48xlarge FAQ
How much does inf2.48xlarge cost?
inf2.48xlarge costs $12.98 per hour on demand with Linux in US East (N. Virginia) (us-east-1), about $9,476 per month. Across 13 regions the Linux on-demand price ranges from $12.98 in US East (N. Virginia) to $22.07 in South America (Sao Paulo). Spot capacity costs $5.048 per hour (61% less), but can be reclaimed. A 1-yr Reserved commitment brings it to $8.178 per hour.
How many vCPUs and how much memory does inf2.48xlarge have?
inf2.48xlarge has 192 vCPUs and 768 GiB of memory (4 GiB per vCPU), running on AMD EPYC 7R13 Processor. It is a gpu & accelerated instance type in the Inf2 family.
What accelerator does inf2.48xlarge have?
inf2.48xlarge has 12 × AWS Inferentia2 with 384 GiB of accelerator memory in total.