SERVCHIP
Interactive view
ARCHITECTURE
Ampere
SERIES
A30
Memory
24GB HBM2e
Bandwidth
933 GB/s
TDP
300W
In StockA30·Best for: Mainstream Enterprise AI
NVIDIA A30 GPU
Ampere
|Data Center GPUsOverview
The NVIDIA A30 GPU is a versatile mainstream accelerator built on the Ampere architecture, featuring 24GB of HBM2e memory with 933 GB/s bandwidth. Designed for balanced AI training and inference workloads, it supports MIG for multi-tenant deployments and delivers strong performance per dollar for enterprise customers. The A30 is ideal for organizations looking to modernize their data center infrastructure with a cost-effective GPU that handles both training and inference.
Key Features
- 24GB HBM2e memory with 933 GB/s bandwidth
- Ampere architecture with Tensor Cores
- Multi-Instance GPU (MIG) support
- TF32 and BF16 precision
- PCIe 4.0 x16 form factor
Ideal Use Cases
ai trainingai inferencehpc
Full Specifications
Complete technical specifications for the A30.
Memory
- Memory
- 24GB HBM2e
- Memory Bandwidth
- 933 GB/s
Compute
- Tensor Cores
- 224 (3rd-gen)
- CUDA Cores
- 3584
- FP16 TFLOPS
- 165
- TF32 TFLOPS
- 82.5
- FP64 TFLOPS
- 5.2
Interconnect
- Interconnect
- PCIe 4.0 x16
Physical
- TDP
- 300W
- Form Factor
- PCIe Gen4
- Cooling
- Passive
Manufacturing
- Launch Date
- Q2 2021
- Manufacturing Process
- 7nm
