Two Nvidia data center GPUs, one Kepler and one Ada Lovelace: 12 GB GDDR5 against 48 GB GDDR6 and 3.6x the memory bandwidth on the L40.
At a glance
Memory
- K80
- 12 GB
- L40
- 48 GB
4x more on L40
Memory bandwidth
- K80
- 240 GB/s
- L40
- 864 GB/s
3.6x higher on L40
Cheapest on-demand
- K80
- $0.90 /GPU/hr
- L40
- $0.78 /GPU/hr
13% cheaper on L40
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | K80 | L40 | Difference |
|---|---|---|---|
| On-demand | -- /GPU/hr from $0.90 | $1.05 /GPU/hr from $0.78 | |
| Reserved | -- /GPU/hr from $0.04 (24 mo) | -- /GPU/hr from $0.70 | |
| Spot | No spot listings | $0.79 /GPU/hr from $0.77 |
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 on-demand and reserved on the K80 and reserved on the L40: 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
K80 vs L40 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | K80 | L40 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 12 GB GDDR5 | 48 GB GDDR6 | 4x more on L40 |
| Bandwidth | 240 GB/s | 864 GB/s | 3.6x higher on L40 |
| Bus width | 384-bit | 384-bit | Identical |
| Compute · dense | |||
| FP8 | Not supported | 362 TFLOPS | L40 only |
| FP16 / BF16 | Not supported | 181.1 TFLOPS | L40 only |
| INT8 | Not supported | 362 TOPS | L40 only |
| FP32 | 4.4 TFLOPS | 90.5 TFLOPS | 21x higher on L40 |
| FP64 | 1.5 TFLOPS | Not supported | K80 only |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 150 W | 300 W | 2x higher on L40 |
| Released | Q4 2014 | Q4 2022 | 8 years newer on L40 |
| Precision | |||
| Supported formats | FP32FP64 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: K80 datasheet and L40 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | K80 /GPU/hr | L40 /GPU/hr |
|---|---|---|
AWS
|
On-Demand $0.90 | -- |
Database Mart
|
Reserved (24mo) $0.0447 | -- |
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 |
Oblivus
|
-- | On-Demand $1.05 |
Spheron
|
-- | On-Demand $1.06 |
EmpirioLabs AI
|
-- | On-Demand $1.50 |
2 providers list the K80 and 12 the L40. Every listing is on the K80 page and the L40 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia K80 was last updated on Sept. 7, 2026, and for Nvidia L40 on Sept. 7, 2026.