NVIDIA® A2 GPU Computing Accelerator - 16GB GDDR6 - PCIe 4.0 x8 - Passive Cooler (w/o CEC)

NVIDIA part #: 900-2G179-0020-100

Versatile Entry-Level Inference
The NVIDIA A2 Tensor Core GPU provides entry-level inference with low power, a small footprint, and high performance for NVIDIA AI at the edge. Featuring a low-profile PCIe Gen4 card and a low 40-60 watt (W) configurable thermal design power (TDP) capability, the A2 brings adaptable inference acceleration to any server.

Price: $1,399.00

Add To Order
Main Specifications
Product SeriesNvidia A2
Core Clock Speed1440 MHz (1770 MHz Boost Clock)
Host InterfacePCI Express 4.0 x8
GPU ArchitectureAmpere
Detailed Specifications
Streaming Processor Cores1280 CUDA Cores
NVIDIA Tensor Cores40 | Gen 3
NVIDIA RT Cores10 | Gen 2
Memory Clock Speed6251 MHz
Memory Interface128-bit
Max Memory Size16 GB GDDR6 ECC
Max Memory Bandwidth200 GB/s
Peak FP324.5 TFLOPS
Peak TF32 Tensor Core9 TFLOPS | 18 TFLOPS Sparsity
Peak FP16 Tensor Core18 TFLOPS | 36 TFLOPS Sparsity
NVIDIA CUDA™ Technology11.1 or later
Peak INT4 Performance72 TOPS | 144 TOPS Sparsity
Peak INT8 Performance36 TOPS | 72 TOPS Sparsity
ECC ProtectionOn by Default
Dual SlotSingle-slot
Dimensions6.61” L x 2.71” H
Form FactorLow-Profile PCIe
Max Graphics Card Power (W)40-60 W | Configurable