inf1 Instance Family

The INF1 instance family in Amazon EC2 is designed for high-performance, low-cost machine learning inference. These instances, generally available, are powered by AWS Inferentia chips, custom-built by AWS for ML inference, along with 2nd Generation Intel Xeon Scalable processors. INF1 instances offer up to 100 Gbps of networking bandwidth. They do not have instance storage, relying on EBS for storage needs.

Source

vCPUs
Up to 96
Memory
Up to 192 GiB
Architecture
x86_64
Processor
Intel
Ranked by CPU performance (Passmark)
🇿🇦 af-south-1
🇭🇰 ap-east-1
🇯🇵 ap-northeast-1
🇰🇷 ap-northeast-2
🇯🇵 ap-northeast-3
🇮🇳 ap-south-1
🇸🇬 ap-southeast-1
🇦🇺 ap-southeast-2
🇨🇦 ca-central-1
🇩🇪 eu-central-1
🇸🇪 eu-north-1
🇮🇹 eu-south-1
🇮🇪 eu-west-1
🇬🇧 eu-west-2
🇫🇷 eu-west-3
🇧🇭 me-south-1
🇧🇷 sa-east-1
🇺🇸 us-east-1
🇺🇸 us-east-2
🇺🇸 us-west-1
🇺🇸 us-west-2
inf1.xlarge 4 vCPU, 8 GiB

On-Demand Price

Spot Price

inf1.2xlarge 8 vCPU, 16 GiB

On-Demand Price

Spot Price

inf1.6xlarge 24 vCPU, 48 GiB

On-Demand Price

Spot Price

inf1.24xlarge 96 vCPU, 192 GiB

On-Demand Price

Spot Price