ml.p3.8xlarge
$14.6880/hr
~$10722.24/month on-demand in N. Virginia (us-east-1)
The ml.p3.8xlarge is a GPU Accelerated SageMaker ML instance with 32 vCPUs and 244 GiB of memory with 4 GPU, starting at $14.688 per hour.
Last updated: Feb 10, 2026, 05:00 AM UTC
All pricing cards update to show this payment type
Pricing Options
On-Demand
Pay as you go - no commitment
Hourly Rate
$14.6880/hr
~$10722.24/mo
Reserved 1yr
One year commitment
Not available
Reserved 3yr
Three year commitment
Not available
P3 Family
Switch between sizes in N. Virginia (us-east-1)
| Size | vCPU | Mem |
|---|---|---|
| ml.p3.16xlarge | 64 | 488 |
| ml.p3.2xlarge | 8 | 61 |
| ml.p3.8xlarge | 32 | 244 |
Instance Specifications
Detailed hardware and capabilities for ml.p3.8xlarge
Processor
vCPUs32
Memory244 GiB
Instance FamilyP3
Architecture64-bit
Clock Speed2.3 GHz
Current Gen
GPU & Accelerator
Has GPU
GPU Count4
GPU NameV100
GPU ManufacturerNVIDIA
GPU Memory64 GiB
Networking
Network Performance10 Gigabit
Network Bandwidth5 Gbps
Max Network Interfaces8
IPv6 Support
ENA Support
EBS (Elastic Block Store)
EBS Optimizeddefault
Baseline IOPS40,000 IOPS
Max IOPS40,000 IOPS
Baseline Bandwidth7,000 Mbps
Max Bandwidth7,000 Mbps
Instance Storage
Instance Storage
Regional Pricing Comparison
Compare ml.p3.8xlarge pricing across all regions - sorted by cost
Current region (us-east-1)Other regions
Click any bar to open that region in a new tab