L40S vs H100 rental price per hour

The H100 costs 135% more per hour than the L40S at the median and has 67% more memory (80 GB against 48 GB). Per GB of memory, the H100 is 41% dearer. Based on 12 L40S and 28 H100 public on-demand listings, last checked .

L40S median
$1.70
10 providers
H100 median
$3.99
17 providers
L40S memory
48 GB
Per GPU
H100 memory
80 GB
Per GPU

Side by side

L40S and H100 public on-demand prices

US dollars per GPU-hour from public pages. Every row behind these numbers is on the L40S price page and the H100 price page, each with its source link.

L40S and H100 price and memory side by side
MeasureL40SH100
Memory per GPU48 GB80 GB
Cheapest on-demand$0.97 (Massed Compute)$1.68 (Latitude.sh)
Median on-demand$1.70$3.99
Priciest on-demand$3.77$12.29
Providers with a price1017
Median price per GB of memory, per hour$0.035$0.050
What this page does not compare. Speed. Throughput depends on the model, precision and software, and we publish no benchmarks. This page compares public rental price and memory only.

Same provider

Providers that list both

Lowest public on-demand price per GPU-hour at each provider that lists both chips.

Providers that list both L40S and H100 on demand
ProviderL40S lowestH100 lowestH100 vs L40S
Amazon Web Services$1.86$6.88+270%
Civo$1.29$2.99+132%
CoreWeave$2.25$6.16+174%
Crusoe$1.50$3.90+160%
DigitalOcean$1.57$4.41+181%
Massed Compute$0.97$2.73+181%
Nebius$1.55$4.50+190%
Oracle Cloud (OCI)$3.50$10.00+186%
RunPod$1.09$2.89+165%
Verda$1.96$4.99+155%

FAQ

L40S vs H100 questions

Is the H100 more expensive than the L40S to rent?

At the median public on-demand price, yes by 135%: $3.99 against $1.70 per GPU-hour.

Which has more memory, the L40S or the H100?

The L40S has 48 GB and the H100 has 80 GB per GPU, as published by the vendors and shown in provider listings.

Where is the cheapest L40S and H100?

The cheapest public on-demand L40S we track is $0.97 at Massed Compute. For the H100 it is $1.68 at Latitude.sh. Last checked Nov 9, 2026.

Need L40S or H100 capacity?

Tell us the GPU, count, term, region and budget. We reply with the public prices that match and the checks to run before you pay. Free, and we do not sell compute.