NVIDIA A10 Tensor Core GPU «NEW»

NEW

NVIDIA® A10 Tensor Core GPU delivers the performance that designers, engineers, artists, and scientists need to meet today’s challenges. A compact, single-slot, 150W GPU, when combined with NVIDIA virtual GPU (vGPU) software, can accelerate multiple data centre workloads—from graphics-rich virtual desktop infrastructure (VDI) to AI—in an easily managed, secure, and flexible infrastructure that can scale to accommodate every need.

Get a Quote

* With sparsity

XENON NVIDIA A10
Specifications
NVIDIA A10 Tensor Core GPU
Specifications

FP32

NVIDIA A10 Tensor Core GPU

31.2 teraFLOPS

Specifications

TF32 Tensor Core

NVIDIA A10 Tensor Core GPU

62.5 teraFLOPS | 125 teraFLOPS*

Specifications

BFLOAT16 Tensor Core

NVIDIA A10 Tensor Core GPU

125 teraFLOPS | 250 teraFLOPS*

Specifications

FP16 Tensor Core

NVIDIA A10 Tensor Core GPU

125 teraFLOPS | 250 teraFLOPS*

Specifications

INT8 Tensor Core

NVIDIA A10 Tensor Core GPU

250 TOPS | 500 TOPS*

Specifications

INT4 Tensor Core

NVIDIA A10 Tensor Core GPU

500 TOPS | 1,000 TOPS*

Specifications

RT Core

NVIDIA A10 Tensor Core GPU

72 RT Cores

Specifications

Encode/decode

NVIDIA A10 Tensor Core GPU

1 encoder 2 decoder (+AV1 decode)

Specifications

GPU memory

NVIDIA A10 Tensor Core GPU

24GB GDDR6

Specifications

GPU memory bandwidth

NVIDIA A10 Tensor Core GPU

600GB/s

Specifications

Interconnect

NVIDIA A10 Tensor Core GPU

PCIe Gen4 64GB/s

Specifications

Form factors

NVIDIA A10 Tensor Core GPU

Single-slot, full-height, full-length (FHFL)

Specifications

Max thermal design power (TDP)

NVIDIA A10 Tensor Core GPU

150W

Specifications

vGPU software support

NVIDIA A10 Tensor Core GPU

  • NVIDIA Virtual PC
  • NVIDIA Virtual Applications
  • NVIDIA RTX Virtual Workstation
  • NVIDIA Virtual Compute Server

Get a Quote Talk to a Solutions Architect