NVIDIA A100 Full Specs - HMC

6,912 Shaders

1.41GHz

80GB HBM2e1.94TB/s

··

19.49 TFLOPS

Form Factor

PCIe Card

TDP

250W

Power Connectors

1x 8-Pin EPS

A100

A1002.5 POPSINT4 Tensor Sparse

x1

3.35 PFLOPS FP4 Tensor SparseGeForce RTX 5090

x1.34

Clock Speed

···

Peak OPS

2.5 POPSINT4 Tensor Sparse

Tensor FP16-16

311.9 TFLOPS

FP16-16 Tensor Sparse

623.7 TFLOPS

Tensor FP16-32

311.9 TFLOPS

FP16-32 Tensor Sparse

623.7 TFLOPS

BF16

38.98 TFLOPS

Tensor BF16

311.9 TFLOPS

BF16 Tensor Sparse

623.7 TFLOPS

Tensor TF32

155.9 TFLOPS

FP32

19.49 TFLOPS

FP64

9.75 TFLOPS

Tensor FP64

19.49 TFLOPS

Tensor INT4

1.25 POPS

Tensor INT8

623.7 TOPS

Pixel Rate

225.6 GPixel/s

Texture Rate

609.1 GTexel/s

Shaders

6,912 Shaders

TMUs

432 TMUs

ROPs

160 ROPs

Tensor Cores

432 T-Cores

SMs

108 SMs

Base Clock

765MHz

Boost Clock

1.41GHz

L1

64KB/SM Tex

L2 Cache

40MB shared

80GB HBM2e

Memory Bus

5120-bit

Memory Speed

3GT/s

Memory Bandwidth

1.94TB/s

ECC

No

TDP

250W

Adaptive Sync

G-Sync

FreeSync

DSC

Not Supported

OpenCL

3

Vulkan

1.2

CUDA

8

PureVideo HD

VP10

VDPAU

Feature Set J

Decoder

5x NVDEC 4

Codec

MPEG-1

MPEG-2

MPEG-4

VC-1

VP8

VP9

AVC (H.264)

HEVC (H.265)

Form Factor

PCIe Card

PCIe

PCIe 4.0 x16

2-Slots

Height

111mm (4.37")

Width

267mm (10.51")

Depth

40mm (1.57")

Cooling

Passive

Power Connectors

1x 8-Pin EPS

Multi-GPU

Multi-GPU Support

Supported

Multi-GPU Type

NVLink

Manufacturer

Chip Designer

Architecture

Family

Branding

Codename

NV170

Market Segment

Server

Release Date

Jun 28, 2021

Foundry

TSMC

Fabrication Node

7N

Die Size

826mm²

Transistor Count

54.2B

Transistor Density

65.62 MTr/mm²