A data center GPU against a consumer card, one Ada Lovelace and one Ampere: 48 GB GDDR6 against 24 GB GDDR6X, 8% more memory bandwidth on the RTX 3090, and NVLink 112 GB/s on the RTX 3090 where the L40S is PCIe only.
At a glance
Memory
- L40S
- 48 GB
- RTX 3090
- 24 GB
2x more on L40S
Memory bandwidth
- L40S
- 864 GB/s
- RTX 3090
- 936 GB/s
1.1x higher on RTX 3090
Cheapest on-demand
- L40S
- $0.38 /GPU/hr
- RTX 3090
- $0.10 /GPU/hr
3.8x more expensive on L40S
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | L40S | RTX 3090 | Difference |
|---|---|---|---|
| On-demand | $1.29 /GPU/hr from $0.38 | $0.22 /GPU/hr from $0.10 | 5.9x more expensive on L40S |
| Reserved | $1.11 /GPU/hr from $0.60 (12 mo) | -- /GPU/hr from $0.06 (1 mo) | |
| Spot | $1.11 /GPU/hr from $0.74 | -- /GPU/hr from $0.09 |
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. No median for reserved and spot on the RTX 3090: we only show one when at least 3 providers list the GPU on that billing type. Verify before provisioning. More on how we price.
Price history
L40S vs RTX 3090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | L40S | RTX 3090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 48 GB GDDR6 | 24 GB GDDR6X | 2x more on L40S |
| Bandwidth | 864 GB/s | 936 GB/s | 1.1x higher on RTX 3090 |
| Bus width | 384-bit | 384-bit | Identical |
| Compute · dense | |||
| FP8 | 733 TFLOPS | Not supported | L40S only |
| FP16 / BF16 | 362.1 TFLOPS | 71.2 TFLOPS | 5.1x higher on L40S |
| INT8 | 733 TOPS | 284.7 TOPS | 2.6x higher on L40S |
| FP32 | 91.6 TFLOPS | 35.6 TFLOPS | 2.6x higher on L40S |
| Platform | |||
| Interconnect | PCIe only | NVLink 112 GB/s | NVLink on RTX 3090 only |
| Board power | 350 W | 350 W | Identical |
| Released | Q3 2023 | Q3 2020 | 3 years newer on L40S |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: L40S datasheet and RTX 3090 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | L40S /GPU/hr | RTX 3090 /GPU/hr |
|---|---|---|
EmpirioLabs AI
|
On-Demand $0.80 | On-Demand $0.40 |
GPU.ai
|
On-Demand $0.80 | On-Demand $17.34 |
Lium
|
On-Demand $0.38 | -- |
Novita
|
On-Demand $0.55 | -- |
Packet·ai
|
On-Demand $0.92 | -- |
Sesterce
|
On-Demand $0.97 | -- |
Massed Compute
|
On-Demand $0.97 | -- |
Runcrate
|
On-Demand $0.97 | -- |
HyperAI
|
-- | On-Demand $0.10 |
Vast.ai
|
-- | On-Demand $0.14 |
Salad
|
-- | On-Demand $0.17 |
Runpod
|
-- | On-Demand $0.22 |
SwissGPU
|
-- | On-Demand $0.31 |
TensorDock
|
-- | On-Demand $0.31 |
35 providers list the L40S and 8 the RTX 3090. Every listing is on the L40S page and the RTX 3090 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
From $1.10 /GPU/hr Compare -
Nvidia A100
From $0.80 /GPU/hr Compare -
Nvidia L40
From $0.33 /GPU/hr Compare -
Nvidia RTX 5090
From $0.35 /GPU/hr Compare -
Nvidia RTX 4090
From $0.33 /GPU/hr Compare
Our data for Nvidia L40S was last updated on Sept. 7, 2026, and for Nvidia RTX 3090 on Sept. 7, 2026.