NVIDIA L40S pricing.
prices as of · 5 providers · 13 configurations
48 GB of GDDR6 at a fraction of an H100's price. Good for inference and rendering, slow for training.
The cheapest L40S today is L40S · Community Cloud at RunPod, $0.790/GPU-hr, read 17 Sept 2026; the cheapest interruptible rate is $0.720/GPU-hr at Verda (DataCrunch).
Cheapest on demand
$0.790
per GPU-hour at RunPod
Median of the spread
$1.47
across 12 configurations
Cheapest interruptible
$0.720
5 of 12 publish one
Per GB of memory
$0.0165/GB-hr
market median $0.0300/GB-hr
What the board is
published by NVIDIA, not by a provider- Memory
- 48 GB
- Memory bandwidth
- 864 GB/s
- Architecture
- Ada Lovelace
- Launched
- 2023
These are the chip's own figures, kept in this site's registry and the same whoever rents it out. A provider's feed is trusted for its prices, its GPU counts and the memory it states per card, never for what the board is. There is no speed score on this page: bandwidth is a published number, and turning it into a benchmark would be inventing one.
Every provider renting the L40S
Cheapest per GPU-hour first. The rate is for one GPU for one hour, so a node of eight sits beside a single card at a comparable number, with the whole machine's hourly price under it. Where a provider publishes an interruptible rate it is in the same cell.
RunPodL40S · Community Cloud
1x L40S · 48 GB each · 16 vCPU · 94 GB RAM
disk none, billed separately · regions not stated · billed per second · marketplace
Community Cloud: vetted third-party hosts, cheaper and less consistent · storage billed separately · billed per second
$0.790per GPU-hour$0.79/hr all inno spot rate
- RunPod
L40S · Secure Cloud
1x L40S · 48 GB each · 16 vCPU · 94 GB RAM
disk none, billed separately · regions not stated · billed per second
Secure Cloud: RunPod's own datacentres · storage billed separately · billed per second
$1.09per GPU-hour$1.09/hr all inno spot rate
Verda (DataCrunch)8L40S.160V · 8x L40S 48GB
8x L40S · 48 GB each · 160 vCPU · 480 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 50GB/s between GPUs
$1.44per GPU-hour$11.51/hr all in$0.720 spot
- Verda (DataCrunch)
1L40S.20V · 1x L40S 48GB
1x L40S · 48 GB each · 20 vCPU · 60 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately
$1.44per GPU-hour$1.44/hr all in$0.720 spot
- Verda (DataCrunch)
2L40S.40V · 2x L40S 48GB
2x L40S · 48 GB each · 40 vCPU · 120 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 50GB/s between GPUs
$1.44per GPU-hour$2.88/hr all in$0.720 spot
- Verda (DataCrunch)
4L40S.80V · 4x L40S 48GB
4x L40S · 48 GB each · 80 vCPU · 240 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 50GB/s between GPUs
$1.44per GPU-hour$5.76/hr all in$0.720 spot
CrusoeNVIDIA L40S
1x L40S · 48 GB each · host size not stated
disk none, billed separately · regions not stated · billed per hour
Priced per GPU-hour · vCPU, RAM and disks billed separately · egress free · no minimum commitment
$1.50per GPU-hour$1.50/hr all inno spot rate
Vultrvcg-l40s-16c-180g-48vram
1x L40S · 48 GB each · 16 vCPU · 180 GB RAM
disk 1.2 TB · regions not stated · billed per hour
Cloud GPU · $1122.91/mo if kept a whole month
$1.67per GPU-hour$1.67/hr all inno spot rate
- Vultr
vcg-l40s-32c-375g-96vram
2x L40S · 48 GB each · 32 vCPU · 375 GB RAM
disk 2.2 TB · regions not stated · billed per hour
Cloud GPU · $2245.82/mo if kept a whole month
$1.67per GPU-hour$3.34/hr all inno spot rate
- Vultr
vcg-l40s-64c-750g-192vram
4x L40S · 48 GB each · 64 vCPU · 750 GB RAM
disk 2.6 TB · regions not stated · billed per hour
Cloud GPU · $4491.65/mo if kept a whole month
$1.67per GPU-hour$6.68/hr all inno spot rate
- Vultr
vbm-64c-2048gb-8-l40-gpu
8x L40S · 48 GB each · 64 vCPU · 2048 GB RAM
disk 480 GB · regions not stated · billed per hour
Bare metal · $12000/mo if kept a whole month
$1.67per GPU-hour$13.37/hr all inno spot rate
CoreWeaveNVIDIA L40S
8x L40S · 48 GB each · 128 vCPU · 1024 GB RAM · NVLink
disk 7.7 TB · 2 regions · billed per hour
$18 an hour for the whole node · egress and VPC free · IPv4 and storage billed separately
$2.25per GPU-hour$18/hr all in$0.985 spot
Billed as a function
1 rate · never rankedThese rates buy a container that runs when it is called, not a machine you hold. CPU and memory are billed on top, a chosen region or a guarantee against preemption costs more, and the figure is a floor rather than a price. They are listed for what they are and ranked nowhere on this site.
ModalNvidia L40S
1x L40S · 48 GB each · host size not stated
disk none, billed separately · regions not stated · billed per second · serverless
$0.000542 a GPU-second, the preemptible region-agnostic rate · a chosen region costs 1.15x to 1.75x and non-preemptible 3x · CPU, memory and volumes billed on top · memory is the board's published spec
$1.95per GPU-hour$1.95/hr all inno spot rate
Frequently asked
- Who rents the L40S cheapest?
- RunPod, at $0.790/GPU-hr on L40S · Community Cloud, read on 17 Sept 2026. That is the on-demand rate for one GPU for one hour; the whole machine bills $0.79/hr.
- What is the NVIDIA L40S?
- 48 GB of GDDR6 at a fraction of an H100's price. Good for inference and rendering, slow for training. It is a Ada Lovelace board from 2023 with 48 GB of memory at 864 GB/s, in the PCIe card package. Those figures come from the vendor, not from any provider's price page.
- Can I rent a L40S on spot?
- Yes. 5 of the 12 configurations here publish an interruptible rate, the cheapest being $0.720/GPU-hr at Verda (DataCrunch), 50% under its own on-demand rate. The machine can be taken back mid-run, so it suits work that checkpoints.
- Why do L40S prices differ so much?
- The spread here runs from $0.790/GPU-hr to $2.25/GPU-hr for the same silicon, because what else comes with it differs: the vCPU and RAM around the GPU, whether a disk is included, how many regions it runs in, and whether the machine belongs to the provider or to a third party on a marketplace. Each row says which.
- How current are these prices?
- They are read every morning from each provider's own public page or API, and this page recomputes with them. Last refresh: 17 Sept 2026. Reserved and committed rates are not indexed. Prices exclude tax.