The NVIDIA L40S is an enterprise-class GPU built on the Ada Lovelace architecture, featuring 48 GB of memory with 864 GB/s of memory bandwidth and a 350 W TDP. It delivers 91.6 TFLOPS of FP32 performance and 362 TFLOPS of FP16 tensor performance. Currently available from 8 providers starting at $0.40/GPU/hour, with a market median of $1.21/GPU/hour across 36 configurations.
Cheapest
$0.400/hr
Vast.ai
Median
$1.205/hr
across 36 offerings
Most expensive
$1.870/hr
Sesterce Cloud
90-day trend
-14.4%
median rate
Hardware specifications
Same across all providers
Memory
48 GB
FP16 performance
362 TFLOPS
peak
TDP
350 W
Memory bandwidth
864 GB/s
Architecture
Ada Lovelace
FP32 performance
91.6 TFLOPS
Brand
NVIDIA
Series
L40
Market price history
Median price across all providers
10 offerings from 8 providers
Sorted by price ascending. Compare configurations side-by-side.
| Provider | Count | vCPU | RAM | Region | Per GPU hour | Total/hr | Action |
|---|---|---|---|---|---|---|---|
| ×2 | 96 | 503 GB |    +3 | From$0.400 +19.4% 30d | From$0.80 | Launch | |
 Novita AIReferral link | ×1 | 22 | 125 GB |    +7 | $0.550 | $0.55 | |
| ×1 | 12 | 72 GB |   +1 | $0.970 0.0% 30d | From$0.97 | Launch | |
 RunpodReferral link | ×1 | 16 | 188 GB | -- | $0.990 +15.1% 30d | $0.99 | Launch |
 Thunder ComputeReferral link | ×1 | 4 | 32 GB | -- | From$0.990 | From$0.99 | Launch |
| ×1 | 20 | 60 GB | -- | $1.370 | From$1.37 | Launch | |
| ×1 | 16 | 180 GB | -- | From$1.560 | From$1.56 | Launch | |
| ×1 | 8 | 64 GB | -- | $1.570 | $1.57 | ||
| ×1 | 8 | 96 GB |  | $1.679 -1.6% 30d | From$1.68 | Launch | |
 Sesterce CloudReferral link | ×1 | 8 | 96 GB | | From$1.870 0.0% 30d | From$1.87 | Launch |
Similar GPUs
Want more from this page?
See incorrect data?
Frequently asked questions
How much does the L40S cost per hour?
The L40S starts at $0.40 per GPU-hour, with a median of $1.21 across 36 available configurations. Actual pricing depends on the provider, region, and whether you rent on-demand, spot or reserved.
Which cloud providers offer the L40S?
11 providers currently list the L40S: Verda, Runcrate, Novita AI, Runpod, Sesterce Cloud, Scaleway and 5 more. Compare their hourly pricing, regions and machine specs in the table above.
How much VRAM does the L40S have?
The L40S comes with 48 GB of VRAM, which determines the largest models and batch sizes it can hold in memory for training and inference.
Where is the L40S cheapest?
The lowest on-demand price we currently track for the L40S is $0.40 per GPU-hour at Vast.ai. Because providers update pricing and availability frequently, check the live table above before you launch.