NVIDIA · Ampere Architecture

Rent NVIDIA A30 in the Cloud

Compact Ampere GPU designed for mainstream AI inference and MIG partitioning.

VRAM 24 GB HBM2e
Bandwidth 933 GB/s
FP16 165.0 TFLOPS
FP32 10.3 TFLOPS
TDP 165W
Architecture Ampere
Cheapest On-Demand $0.25/hr
Average On-Demand $0.26/hr
Providers 2

Compare NVIDIA A30 Cloud Pricing — 2 Providers

On-Demand

NVIDIA A30 — On-Demand pricing
Provider Price / GPU / hr Availability Notes Action
Massed Compute United States $0.25/hr CHEAPEST Available — Visit Provider
RunPod United States $0.26/hr Available Secure Cloud Visit Provider

Prices last verified: April 13, 2026

NVIDIA A30 Technical Specifications

Manufacturer NVIDIA
Architecture Ampere
VRAM 24 GB HBM2e
Memory Bandwidth 933 GB/s
FP16 (Tensor) 165.0 TFLOPS
FP32 10.3 TFLOPS
TDP 165W
Release Year 2021
Segment Data center
Memory Type HBM2e

Best For

Inference multi-instance GPU workloads

Compare with Other GPUs

See how NVIDIA A30 stacks up against other popular cloud GPUs in specs, pricing, and availability.

NVIDIA A30 vs NVIDIA H100 SXM NVIDIA A30 vs NVIDIA H200 SXM NVIDIA A30 vs NVIDIA A100 SXM (80GB) NVIDIA A30 vs NVIDIA L40S