A30 GPU
The NVIDIA A30 is a versatile data center GPU for mainstream AI inference, training, and HPC workloads with Multi-Instance GPU support.
Starting Price
$0.11/hr
Available on 2 cloud providers
Key Specifications
💾Memory
24GB VRAM
🏗️Architecture
Ampere
⚙️Compute Units
N/A
🧮Tensor Cores
224
Technical Specifications
Hardware Details
ManufacturerNVIDIA
ArchitectureAmpere
CUDA Cores3584
Tensor Cores224
RT CoresN/A
Compute UnitsN/A
GenerationN/A
Memory & Performance
VRAM24GB
Memory InterfaceN/A
Memory Bandwidth933 GB/s
FP32 Performance10.3 TFLOPS
FP16 Performance165 TFLOPS
INT8 PerformanceN/A
Performance
Computing Power
CUDA Cores3,584
Tensor Cores224
Computational Performance
FP32 (TFLOPS)10.3
FP16 (TFLOPS)165
Common Use Cases
AI inference, training, HPC, MIG workloads
Machine Learning & AI
- Training large language models and transformers
- Computer vision and image processing
- Deep learning model development
- High-performance inference workloads
Graphics & Compute
- 3D rendering and visualization
- Scientific simulations
- Data center graphics virtualization
- High-performance computing (HPC)
Market Context
The A30 sits within NVIDIA’s Ampere architecture lineup, positioned in the high performance tier. It’s designed specifically for data center and enterprise use.
Cloud Availability
Available across 2 cloud providers with prices ranging from $0.11/hr. Pricing and availability may vary by region and provider.
Market Position
Released in 2021, this GPU is positioned for enterprise and data center workloads.