Skip to content
GPUVerse Blog

NVIDIA · Ampere · A100

A100 40GB

40 GB of HBM2e at 1.6 TB/s, built for training, inference, fine-tuning.

From $1.29/hrindicative · across 8 providers

A100 40GB render

Specifications

Memory

40 GB

HBM2e

Bandwidth

1.55 TB/s

FP16 tensor

312 TFLOPS

FP8 tensor

-

no FP8 path

Interconnect

PCIe/NVLink

TDP

400 W

CUDA cores

6,912

Tensor cores

432

Where to rent it

ProviderTypeSpotDetails
RunPod logoRunPodneocloudYesView
Vast.ai logoVast.aimarketplaceYesView
Lambda logoLambdaneocloud-View
Crusoe logoCrusoeneocloud-View
Amazon AWS logoAmazon AWShyperscalerYesView
Google Cloud logoGoogle CloudhyperscalerYesView
Microsoft Azure logoMicrosoft AzurehyperscalerYesView
Oracle Cloud logoOracle CloudhyperscalerYesView

Compare with

A100 80GB render
NVIDIA · Ampere

A100 80GB

Memory
80 GB HBM2e
Bandwidth
2.0 TB/s
midfrom$1.19/hr
RTX 4090 render
NVIDIA · Ada Lovelace

RTX 4090

Memory
24 GB GDDR6X
Bandwidth
1.0 TB/s
budgetfrom$0.59/hr
L40S render
NVIDIA · Ada Lovelace

L40S

Memory
48 GB GDDR6
Bandwidth
0.9 TB/s
midfrom$1.19/hr
A10 render
NVIDIA · Ampere

A10

Memory
24 GB GDDR6
Bandwidth
0.6 TB/s
budgetfrom$0.75/hr