NVIDIA A100 PCIe 40GB

Ampere PCIe 2020 7nm Spec confidence: Official
FP32
20
TFLOPS
VRAM
40 GB
Bandwidth
1.6 TB/s
memory
TDP
250 W
78.0 TFLOPS/kW
Download datasheet

Performance

Peak theoretical throughput by precision type

INT8
8-bit integer
624 TOPS
1,248 with sparsity
INT4
4-bit integer
1,248 TOPS
2,496 with sparsity
FP16
16-bit floating point
312 TFLOPS
624 with sparsity
BF16
Brain Float 16
312 TFLOPS
624 with sparsity
TF32
TensorFloat-32
156 TFLOPS
312 with sparsity
FP32
32-bit floating point
20 TFLOPS
FP64
64-bit floating point
9.7 TFLOPS
Dense Added with sparsity
1.248 TFLOPS/W FP16 at 250 W
0.078 TFLOPS/W FP32 at 250 W

The A100 PCIe 40GB in the GPU landscape

Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper

06001,2001,8002,4003,0000 W250 W500 W750 W1000 W1250 W1500 WInstinct MI355XA100 PCIe 40GB

Higher and further left is better: more half-precision throughput for less power.

Specifications

Architecture

Ampere

Form Factor

PCIe

Launch Year

2020

Process Node

7nm

Memory

40 GB

Bandwidth

1,555 GB/s

TDP

250 W

Max Power

~288 W

Spec Confidence

Official

Full Specifications

Memory
VRAM 40 GB
Bandwidth 1.6 TB/s
Power & Thermal
TDP 250 W
Enterprise Features
Sparsity Yes
General
Form Factor PCIe
Architecture Ampere
Launch Year 2020

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2020-05-14

Data Quality

Spec confidence
Official
Clock basis
Boost
Precisions tracked
7
Normalization
All values in TFLOPS

Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.

Browse GPU Cloud Providers

Similar GPUs

NVIDIA RTX A6000 48GB

PCIe · 2020
FP32: 39 TFLOPS
Compare vs RTX A6000

NVIDIA A40 48GB

PCIe · 2021
FP32: 37 TFLOPS
Compare vs A40

NVIDIA A10G 24GB

PCIe · 2022
FP32: 35 TFLOPS
Compare vs A10G

Systems Using This GPU

Pre-configured systems featuring the NVIDIA A100

SystemGPU CountPeak PerformanceTotal Power
NVIDIA DGX A100 320GB
DGX · rack · 2020
8x A100 4.99 PFLOPS 6.0 kW View System

Frequently Asked Questions

How many TFLOPS does the NVIDIA A100 have?

The NVIDIA A100 delivers 19.5 TFLOPS for FP32 operations, 312 TFLOPS for FP16, and N/A TFLOPS for FP8 precision.

What is the power consumption of the NVIDIA A100?

The NVIDIA A100 has a TDP (Thermal Design Power) rating of 250 watts.

How much memory does the NVIDIA A100 have?

The NVIDIA A100 is equipped with 40 GB of VRAM with 1555 GB/s memory bandwidth.

What architecture is the NVIDIA A100 based on?

The NVIDIA A100 is based on the Ampere architecture, launched in 2020.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs