Tesla A100 and Tesla T40 24 GB - video card comparison

Tesla A100

Tesla T40 24 GB

Architecture

Ampere (2020−2022)

Turing (2018−2021)

GPU code name

GA100

TU102

Market segment

Workstation

Workstation

Release date

-

-

Interface

PCIe 4.0 x16

PCIe 3.0 x16

Supplementary power connectors

None

1x 6-pin + 1x 8-pin

Boost clock speed

-

1560 MHz

Core clock speed

1110 MHz

-

Manufacturing process technology

7 nm

12 nm

Number of transistors

38,000 million

18,600 million

Pipelines / CUDA cores

7936

4608

Power consumption (TDP)

260 Watt

260 Watt

Texture fill rate

550.6

449.3

Memory type

HBM2e

GDDR6

Maximum RAM amount

32 GB

24 GB

Memory bus width

4096 Bit

384 Bit

Memory clock speed

2400 MHz

13 GB/s

Memory bandwidth

1,229 GB/s

624.0 GB/s

Display Connectors

No outputs

No outputs

DirectX

12 (12_1)

12 Ultimate (12_2)

CUDA

-

7.5

OpenCL

-

3.0

OpenGL

4.6

4.6

Shader Model

-

6.7

Vulkan

-

1.3