NVIDIA A100 PCIe 80 GB vs NVIDIA Tesla K80

Comparison of NVIDIA A100 PCIe 80 GB with 80 GB HBM2e and 6,912 cores vs NVIDIA Tesla K80 with 12 GB GDDR5 and 2,496 cores.

Loading...

Performance Rating

A100 A100
H200 H200
MI325X MI325X

NVIDIA A100 PCIe 80 GB

25.6

NVIDIA A100 PCIe 80 GB

25.6
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA Tesla K80

NVIDIA Tesla K80

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Additional

Memory

Memory Size

🔥 +567% 80 ГБ
12 GB ×2 (24 ГБ)

Memory Type

HBM2e GDDR5

Memory Bandwidth

🔥 1.94 TB/s
240.6 GB/s ×2 (481.2 GB/s)

Memory Bus Width

5,120 бит 384 бит ×2 (768 бит)

ML Performance

FP16 (Half Precision)

🔥 77.97 TFLOPS
No

BF16 (Brain Float)

🔥 311.84 TFLOPS
No

TF32 (TensorFloat)

🔥 155.92
No

Compute Power

FP32 (Single Precision)

🔥 +374% 19.49 TFLOPS
4.113 TFLOPS

FP64 (Double Precision)

🔥 +611% 9.746 TFLOPS
1.3711 TFLOPS

CUDA Cores

🔥 +177% 6,912
2,496 ×2 (4992)

RT Cores

No No

Architecture & Compatibility

GPU Architecture

Ampere Kepler 2.0

SM (Streaming Multiprocessor)

🔥 108
No

PCIe Version

PCIe 4.0 x16 PCIe 3.0 x16

ML Software Support

CUDA Version

🔥 8.0
3.7

Clocks & Performance

Base Clock

🔥 +90% 1,065
562

Boost Clock

🔥 +71% 1,410
824

Memory Clock

🔥 +21% 1,512
1,253

Power Consumption

TDP/TGP

300 W 300 W

Recommended PSU

700 W 700 W

Power Connector

8-pin EPS 1x 8-pin

Rendering

Texture Units (TMU)

🔥 +108% 432
208 ×2 (416)

ROP

No No

L2 Cache

🔥 80 MB
1536 KB

Additional

Slots

Dual-slot Dual-slot

Release Date

June 28, 2021 Nov. 17, 2014

Display Outputs

No outputs
No outputs

Renting is cheaper than buying