Two Nvidia data center GPUs, one Ampere and one Ada Lovelace: 24 GB GDDR6 against 48 GB GDDR6 and 44% more memory bandwidth on the L40.
At a glance
Updated 6 minutes ago- Memory
- 24 GB
- GDDR6
- Memory bandwidth
- 600 GB/s
- From
- $1.00 /GPU/hr
- 2 providers
- Memory
- 48 GB
- GDDR6
- Memory bandwidth
- 864 GB/s
- From
- $0.33 /GPU/hr
- 14 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | A10G | L40 | Difference |
|---|---|---|---|
| On-demand | $1.00 | $0.33 | 3x more expensive on A10G |
| Spot | No spot listings | $0.40 | |
| Reserved | $0.43 | $0.70 | 38% cheaper on A10G |
| Median | $1.60 | $1.00 | 38% cheaper on L40 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
A10G vs L40 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A10G | L40 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR6 | 48 GB GDDR6 | 2x more on L40 |
| Bandwidth | 600 GB/s | 864 GB/s | 1.4x higher on L40 |
| Bus width | 384-bit | ||
| Compute · dense | |||
| FP8 | Not supported | 362 TFLOPS | L40 only |
| FP16 / BF16 | 70 TFLOPS | 181.1 TFLOPS | 2.6x higher on L40 |
| INT8 | 140 TOPS | 362 TOPS | 2.6x higher on L40 |
| FP32 | 35 TFLOPS | 90.5 TFLOPS | 2.6x higher on L40 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 300 W | 300 W | Identical |
| Released | Q2 2021 | Q4 2022 | 2 years newer on L40 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A10G datasheet and L40 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A10G /GPU/hr | L40 /GPU/hr |
|---|---|---|
Hugging Face
|
On-Demand $1.00 | -- |
AWS
|
On-Demand $1.01 | -- |
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 |
2 providers list the A10G and 14 the L40. Every listing is on the A10G page and the L40 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia A10G was last updated on Sept. 7, 2026, and for Nvidia L40 on Sept. 7, 2026.