Skip to content
InfratoApp
InfratoApp / Compare / A100 vs RTX-4090-24GB

A100 vs RTX-4090-24GB

RTX-4090-24GB is the cheaper of the two: $0.16/hr at Salad, against $1.09/hr for A100 at Civo — 6.8× the price.

Cheapest on each

A100
12 SELLERSView A100 →
$1.09/hr
cheapest, at Civo
RTX-4090-24GB
13 SELLERSView RTX-4090-24GB →
$0.16/hr
cheapest, at Salad

Aligned facts

Same sheet, same order, both cards — including the gaps.

AttributeA100RTX-4090-24GB
VRAM○ no answer24 GB
Bandwidth○ no answer1.008 TB/s
FP16○ no answer165 TFLOPS
FP8○ no answer330 TFLOPS
TDP○ no answer450 W
NVLink○ no answerPCIe 4 (no NVLink)
ArchitectureAmpereAda Lovelace
Released○ no answer2022

A blank is not a zero. Where one card states a figure and the other states none, neither wins the row — the second number is unknown, and ranking an unknown as nothing is how a comparison starts to lie.

Every seller, both cards

One row per seller, cheapest tier, on-demand list prices only. Sellers of both cards lead, cheapest combined rate first.

SellerA100RTX-4090-24GB
Spheron
Runcrate
Civo○ no answer
Hyperstack○ no answer
Massed Compute○ no answer
Sesterce○ no answer
Koyeb○ no answer
Cerebrium○ no answer
Modal○ no answer
Lyceum○ no answer
CoreWeave○ no answer
Replicate○ no answer
Salad○ no answer
Theta EdgeCloud○ no answer
TensorDock○ no answer
Packet·ai○ no answer
Shadeform○ no answer
Beam○ no answer
GPUhub○ no answer
Vast.ai○ no answer
Oblivus○ no answer
Lium○ no answer
RunPod○ no answer

What this page will not tell you

On-demand list rates only · spot and committed terms priced separately and not mixed in here · no performance-per-dollar figure appears on this page, because it needs a throughput benchmark we do not hold and dividing a price by a vendor’s own TFLOPS claim produces a number that looks authoritative and means nothing