gpu.gl

NVIDIA L40S cloud prices

L40S on-demand cloud prices run from $1.55 to $3.50 per GPU-hour across 6 providers gpu.gl tracks (median $1.83). The lowest is Nebius.

On-demand$1.55low · median $1.83 · high $3.50 · 6 providers
Spot / interruptible$0.970low · median $0.970 · high $0.970 · 1 provider
Reserved 1 yr$1.17low · median $1.17 · high $1.17 · 1 provider
Reserved 3 yr$0.804low · median $0.804 · high $0.804 · 1 provider

Lowest price by provider

L40S prices by provider
ConfigurationWhereFetchedSource
NebiusOn-demand $1.55 L40S with Intel CPU (lowest configuration) · 1 GPU Not stated per region Nebius ↗
DigitalOceanOn-demand $1.57 NVIDIA L40S GPU Droplet · 1 GPU Not stated per region DigitalOcean ↗
OVHcloudOn-demand $1.80 l40s-180 · 1, 2, 4 GPUs US catalog (OVHcloud US regions) OVHcloud ↗
AWSOn-demand $1.86 g6e.xlarge · 1, 4, 8 GPUs 4 regions AWS ↗
DataCrunchOn-demand $1.94 8L40S.160V · 1, 2, 4, 8 GPUs Not stated per region DataCrunch ↗
OCIOn-demand $3.50 B109479 (Compute - GPU - L40S) · 1 GPU All commercial regions OCI ↗
DataCrunchSpot / interruptible $0.970 1L40S.20V · 1, 2, 4, 8 GPUs Not stated per region DataCrunch ↗
AWSReserved 1 yr $1.17 g6e.xlarge · 1, 4, 8 GPUs 4 regions AWS ↗
AWSReserved 3 yr $0.804 g6e.xlarge · 1, 4, 8 GPUs 4 regions AWS ↗

Price history

History starts 2026-11-08. gpu.gl stores one snapshot a day from launch and never back-fills, so a chart appears here once there are two daily snapshots with on-demand prices for this GPU (1 so far).

L40S specifications

The L40S has 48 GB of GDDR6 and 0.864 TB/s of memory bandwidth. That is about 419 dense FP16 TFLOPS per TB/s of bandwidth; workloads that stream large weights, like LLM inference at small batch sizes, tend to be limited by the bandwidth figure, while large-batch training leans on the compute figure.

ArchitectureAda Lovelace
Memory48 GB GDDR6
Memory bandwidth0.864 TB/s
FP16 / BF16 tensor (dense)362.05 TFLOPS
FP8 tensor (dense)733 TFLOPS
Max power350 W
Form factorPCIe Gen4, dual slot

Source: NVIDIA spec sheet. Tensor figures are dense (without sparsity).

Every configuration

Each instance size or listing, at its cheapest region.

ProviderPrice type$/GPU-hrInstanceGPUsInstance priceCheapest regionFetchedSource
NebiusOn-demand$1.55L40S with Intel CPU (lowest configuration)1$1.55 per GPU-hourNot stated per regionNebius ↗
DigitalOceanOn-demand$1.57NVIDIA L40S GPU Droplet1$1.57 per GPU-hourNot stated per regionDigitalOcean ↗
OVHcloudOn-demand$1.80l40s-1802$3.60/hrUS catalog (OVHcloud US regions)OVHcloud ↗
OVHcloudOn-demand$1.80l40s-3604$7.20/hrUS catalog (OVHcloud US regions)OVHcloud ↗
OVHcloudOn-demand$1.80l40s-901$1.80/hrUS catalog (OVHcloud US regions)OVHcloud ↗
NebiusOn-demand$1.82L40S with AMD CPU (lowest configuration)1$1.82 per GPU-hourNot stated per regionNebius ↗
AWSOn-demand$1.86g6e.xlarge1$1.86/hrUS East (N. Virginia) +3 moreAWS ↗
DataCrunchOn-demand$1.948L40S.160V8$15.51/hrNot stated per regionDataCrunch ↗
DataCrunchOn-demand$1.941L40S.20V1$1.94/hrNot stated per regionDataCrunch ↗
DataCrunchOn-demand$1.942L40S.40V2$3.88/hrNot stated per regionDataCrunch ↗
DataCrunchOn-demand$1.944L40S.80V4$7.76/hrNot stated per regionDataCrunch ↗
AWSOn-demand$2.24g6e.2xlarge1$2.24/hrUS East (N. Virginia) +3 moreAWS ↗
AWSOn-demand$2.62g6e.12xlarge4$10.49/hrUS East (N. Virginia) +3 moreAWS ↗
AWSOn-demand$3.00g6e.4xlarge1$3.00/hrUS East (N. Virginia) +3 moreAWS ↗
OCIOn-demand$3.50B109479 (Compute - GPU - L40S)1$3.50 per GPU-hourAll commercial regionsOCI ↗
AWSOn-demand$3.77g6e.24xlarge4$15.07/hrUS East (N. Virginia) +3 moreAWS ↗
AWSOn-demand$3.77g6e.48xlarge8$30.13/hrUS East (N. Virginia) +3 moreAWS ↗
AWSOn-demand$4.53g6e.8xlarge1$4.53/hrUS East (N. Virginia) +3 moreAWS ↗
AWSOn-demand$7.58g6e.16xlarge1$7.58/hrUS East (N. Virginia) +3 moreAWS ↗
DataCrunchSpot / interruptible$0.9701L40S.20V1$0.9695/hrNot stated per regionDataCrunch ↗
DataCrunchSpot / interruptible$0.9702L40S.40V2$1.94/hrNot stated per regionDataCrunch ↗
DataCrunchSpot / interruptible$0.9704L40S.80V4$3.88/hrNot stated per regionDataCrunch ↗
DataCrunchSpot / interruptible$0.9708L40S.160V8$7.76/hrNot stated per regionDataCrunch ↗
AWSReserved 1 yr$1.17g6e.xlarge1$1.17/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$1.41g6e.2xlarge1$1.41/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$1.65g6e.12xlarge4$6.61/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$1.89g6e.4xlarge1$1.89/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$2.37g6e.24xlarge4$9.49/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$2.37g6e.48xlarge8$18.98/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$2.85g6e.8xlarge1$2.85/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 1 yr$4.77g6e.16xlarge1$4.77/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$0.804g6e.xlarge1$0.8040/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$0.969g6e.2xlarge1$0.9686/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$1.13g6e.12xlarge4$4.53/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$1.30g6e.4xlarge1$1.30/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$1.63g6e.48xlarge8$13.02/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$1.63g6e.24xlarge4$6.51/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$1.96g6e.8xlarge1$1.96/hrUS East (N. Virginia) +3 moreAWS ↗
AWSReserved 3 yr$3.27g6e.16xlarge1$3.27/hrUS East (N. Virginia) +3 moreAWS ↗

Compare