A data center GPU against a workstation card on the Ampere architecture: 48 GB GDDR6 against 24 GB GDDR6, 10% more memory bandwidth on the A5000, and NVLink 112 GB/s on the A5000 where the A40 is PCIe only.
At a glance
Memory
- A40
- 48 GB
- A5000
- 24 GB
2x more on A40
Memory bandwidth
- A40
- 696 GB/s
- A5000
- 768 GB/s
1.1x higher on A5000
Cheapest on-demand
- A40
- $0.49 /GPU/hr
- A5000
- $0.25 /GPU/hr
49% cheaper on A5000
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | A40 | A5000 | Difference |
|---|---|---|---|
| On-demand | $0.49 /GPU/hr from $0.49 | $1.38 /GPU/hr from $0.25 | 2.8x more expensive on A5000 |
| Reserved | -- /GPU/hr from $0.61 (24 mo) | $0.35 /GPU/hr from $0.25 (24 mo) | |
| Spot | No spot listings | -- /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 on the A40 and spot on the A5000: 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
A40 vs A5000 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A40 | A5000 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 48 GB GDDR6 | 24 GB GDDR6 | 2x more on A40 |
| Bandwidth | 696 GB/s | 768 GB/s | 1.1x higher on A5000 |
| Bus width | 384-bit | 384-bit | Identical |
| Compute · dense | |||
| FP16 / BF16 | 149.7 TFLOPS | 111.1 TFLOPS | 1.3x higher on A40 |
| INT8 | 299.3 TOPS | ||
| FP32 | 37.4 TFLOPS | 27.8 TFLOPS | 1.3x higher on A40 |
| Platform | |||
| Interconnect | PCIe only | NVLink 112 GB/s | NVLink on A5000 only |
| Board power | 300 W | 230 W | 1.3x higher on A40 |
| Released | Q4 2020 | Q2 2021 | 6 months newer on A5000 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A40 datasheet and A5000 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A40 /GPU/hr | A5000 /GPU/hr |
|---|---|---|
Exoscale
|
On-Demand $1.05 | On-Demand $1.14 |
Sesterce
|
On-Demand $1.21 | On-Demand $1.56 |
Database Mart
|
Reserved (24mo) $0.61 | Reserved (24mo) $0.25 |
Utho
|
Reserved (36mo) $0.65 | Reserved (36mo) $0.29 |
GPU.ai
|
On-Demand $0.49 | -- |
Runpod
|
On-Demand $0.49 | -- |
EmpirioLabs AI
|
On-Demand $1.49 | -- |
Salad
|
-- | On-Demand $0.25 |
Runcrate
|
-- | On-Demand $1.56 |
Paperspace
|
-- | On-Demand $1.38 |
Cirrascale
|
-- | Reserved (12mo) $0.42 |
7 providers list the A40 and 11 the A5000. Every listing is on the A40 page and the A5000 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia RTX 6000 Ada
From $0.55 /GPU/hr Compare -
Nvidia L40
From $0.33 /GPU/hr Compare -
Nvidia A4000
From $0.15 /GPU/hr Compare
Our data for Nvidia A40 was last updated on Sept. 7, 2026, and for Nvidia A5000 on Sept. 7, 2026.