NVIDIA A10

The NVIDIA A10 is a data center graphics card based on the Ampere architecture. Released on 2021-04-12 with a launch price of $3,000.00, it features 9216 CUDA cores and 24 GB of GDDR6 memory.

A10 GPU

General info about A10

Architecture Ampere (GA102)
Market Segment Data Center
Release Date 2021-04-12
Launch Price $3,000.00
Manufacturing Process 8nm

Technical specs of A10

CUDA Cores 9216
Base Clock Speed 885 MHz
Boost Clock Speed 1695 MHz
Transistor Count 28.3B
VRAM Capacity 24 GB GDDR6
Memory Bus Width 384 bits
Memory Bandwidth 600.0 GB/s
TDP 150 W

Key technical parameters of the NVIDIA A10 include its 9216 CUDA cores, 885 MHz base clock, and 1695 MHz boost clock, delivering high performance for data center applications.

Benchmark performance of A10

FP32 (float) 31.2 TFLOPS
FP16 (half) 62.4 TFLOPS

The NVIDIA A10 achieves 31.2 TFLOPS in FP32 performance, making it a top contender in data center workloads.

Gaming performance of A10

The NVIDIA A10 is not designed for gaming.

Features and connectivity of A10

Tensor_cores Gen 3
Ports PCIe Gen4

Physical dimensions of A10

Length 9.5in
Width 4.4in
Height 1-slot

Rankings and efficiency of A10

Ranking Position #20
Popularity Ranking #20
Cost Effectiveness 2.5/5
Power Efficiency 1.6/5

Top GPUs for Training and Inference

Category Rank 1 Rank 2 Rank 3
Best for Training NVIDIA H200 NVIDIA H100 NVIDIA B200
Best for Inference NVIDIA A40 NVIDIA A100 NVIDIA A10
0 of 44 row(s) selected.
Page 1 of 0