A data center GPU against a consumer card, one Ada Lovelace and one Ampere: 48 GB GDDR6 against 8 GB GDDR6 and 3.9x the memory bandwidth on the L40.
At a glance
Updated 3 minutes ago- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 864 GB/s
- From
- $0.33 /GPU/hr
- 14 providers
- Memory
- 8 GB
- GDDR6
- Memory bandwidth
- 224 GB/s
- From
- $0.07 /GPU/hr
- 1 provider
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | L40 | RTX 3050 | Difference |
|---|---|---|---|
| On-demand | $0.33 | $0.07 | 4.7x more expensive on L40 |
| Spot | $0.40 | $0.03 | 13x more expensive on L40 |
| Reserved | $0.70 | No reserved listings | |
| Median | $1.00 | $0.07 | 14x more expensive on L40 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
L40 vs RTX 3050 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | L40 | RTX 3050 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 48 GB GDDR6 | 8 GB GDDR6 | 6x more on L40 |
| Bandwidth | 864 GB/s | 224 GB/s | 3.9x higher on L40 |
| Bus width | 384-bit | 128-bit | 3x wider on L40 |
| Compute · dense | |||
| FP8 | 362 TFLOPS | Not supported | L40 only |
| FP16 / BF16 | 181.1 TFLOPS | 18.2 TFLOPS | 9.9x higher on L40 |
| INT8 | 362 TOPS | 72.8 TOPS | 5x higher on L40 |
| FP32 | 90.5 TFLOPS | 9.1 TFLOPS | 9.9x higher on L40 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 300 W | 130 W | 2.3x higher on L40 |
| Released | Q4 2022 | Q1 2022 | 9 months newer on L40 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: L40 datasheet and RTX 3050 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | L40 /GPU/hr | RTX 3050 /GPU/hr |
|---|---|---|
Lium
|
On-Demand $0.33 | -- |
Vast.ai
|
On-Demand $0.46 | -- |
EmpirioLabs AI
|
On-Demand $0.70 | -- |
GPU.ai
|
On-Demand $0.78 | -- |
Thunder Compute
|
On-Demand $0.79 | -- |
Massed Compute
|
On-Demand $0.86 | -- |
Sesterce
|
On-Demand $0.97 | -- |
Runcrate
|
On-Demand $0.97 | -- |
Salad
|
-- | On-Demand $0.07 |
14 providers list the L40 and 1 the RTX 3050. Every listing is on the L40 page and the RTX 3050 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia L40 was last updated on Sept. 7, 2026, and for Nvidia RTX 3050 on Sept. 7, 2026.