a2-ultragpu-4g

The a2-ultragpu-4g instance is in the Accelerator Optimized family with 48 vCPUs and 680 GiB of memory starting at $4.3985 per hour on-demand or $2.1991 per hour with spot instances.

Pricing

$4.953

On Demand

N/A

Spot

Request a demo

Family Sizes

SizevCPUsMemory (GiB)
a2-highgpu-1g1285
a2-ultragpu-1g12170
a2-highgpu-2g24170
a2-ultragpu-2g24340
a2-highgpu-4g48340
a2-ultragpu-4g48680
a2-highgpu-8g96680
a2-megagpu-16g961360
a2-ultragpu-8g961360

Compare a2-ultragpu-4g to other instances

Instance Variants

Variant
a2
a3

Instance Details

ComputeValue
vCPUs48
Memory (GiB)680
Memory per vCPU (GiB)14.17
Shared CPU
false
GPU
4
GPU Modelnvidia-a100-80gb
GPU Memory (GiB)0
Accelerator TypeN/A
StorageValue
Local SSD
false
Local SSD Size (GB)0
NetworkingValue
Network PerformanceVariable
GCPValue
Instance Typea2-ultragpu-4g
Pretty NameA2 Ultragpu 4g
FamilyAccelerator optimized
Generationcurrent
Compute Optimized
false
Memory Optimized
false
The Vantage Remote MCP Server: Use AI to Analyze Your Cost and Usage Data

Ask questions about your organization's previous and current cloud cost spend, cost tagging, provider integrations, and more.

Vantage Supports GCP Active Resources and Cost Recommendations

Save on GCP with GCP active resources and with cost recommendations for GCP compute resources.

Vantage Launches Support for Anthropic Costs

Track Claude model usage and costs alongside your other providers.

DocsBy Vantage