Two Nvidia data center GPUs, one Ada Lovelace and one Volta: 48 GB GDDR6 against 32 GB HBM2, 31% more memory bandwidth on the V100S, and NVLink 300 GB/s on the V100S where the L40 is PCIe only.
At a glance
Updated 7 minutes ago- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 864 GB/s
- From
- $0.78 /GPU/hr
- 12 providers
- Memory
- 32 GB
- HBM2
- Memory bandwidth
- 1.13 TB/s
- From
- $0.88 /GPU/hr
- 1 provider
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | L40 | V100S | Difference |
|---|---|---|---|
| On-demand | $0.78 | $0.88 | 11% cheaper on L40 |
| Spot | $0.78 | No spot listings | |
| Reserved | $0.70 | No reserved listings | |
| Median | $1.05 | $0.88 | 16% cheaper on V100S |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
L40 vs V100S price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | L40 | V100S | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 48 GB GDDR6 | 32 GB HBM2 | 1.5x more on L40 |
| Bandwidth | 864 GB/s | 1,134 GB/s | 1.3x higher on V100S |
| Bus width | 384-bit | 4,096-bit | 11x wider on V100S |
| Compute · dense | |||
| FP8 | 362 TFLOPS | Not supported | L40 only |
| FP16 / BF16 | 181.1 TFLOPS | 130 TFLOPS | 1.4x higher on L40 |
| INT8 | 362 TOPS | ||
| FP32 | 90.5 TFLOPS | 16.4 TFLOPS | 5.5x higher on L40 |
| FP64 | Not supported | 8.2 TFLOPS | V100S only |
| Platform | |||
| Interconnect | PCIe only | NVLink 300 GB/s | NVLink on V100S only |
| Board power | 300 W | 250 W | 1.2x higher on L40 |
| Released | Q4 2022 | Q4 2019 | 3 years newer on L40 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32INT4INT8 | FP16FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: L40 datasheet and V100S datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | L40 /GPU/hr | V100S /GPU/hr |
|---|---|---|
GPU.ai
|
On-Demand $0.78 | -- |
Thunder Compute
|
On-Demand $0.79 | -- |
Massed Compute
|
On-Demand $0.86 | -- |
Runcrate
|
On-Demand $0.97 | -- |
Oblivus
|
On-Demand $1.05 | -- |
Spheron
|
On-Demand $1.06 | -- |
Sesterce
|
On-Demand $1.10 | -- |
EmpirioLabs AI
|
On-Demand $1.50 | -- |
OVH
|
-- | On-Demand $0.88 |
12 providers list the L40 and 1 the V100S. Every listing is on the L40 page and the V100S page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia L40 was last updated on Sept. 7, 2026, and for Nvidia V100S on Sept. 7, 2026.