A workstation GPU against a data center card on the Ampere architecture: 16 GB GDDR6 against 40 GB HBM2, 3.5x the memory bandwidth on the A800, and NVLink 400 GB/s on the A800 where the A4000 is PCIe only.
At a glance
Memory
- A4000
- 16 GB
- A800
- 40 GB
2.5x more on A800
Memory bandwidth
- A4000
- 448 GB/s
- A800
- 1,555 GB/s
3.5x higher on A800
Cheapest on-demand
- A4000
- $0.15 /GPU/hr
- A800
- $1.18 /GPU/hr
7.9x more expensive on A800
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | A4000 | A800 | Difference |
|---|---|---|---|
| On-demand | $0.21 /GPU/hr from $0.15 | -- /GPU/hr from $1.18 | |
| Reserved | $0.25 /GPU/hr from $0.11 | No reserved listings | |
| Spot | -- /GPU/hr from $0.10 | No spot listings |
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 spot on the A4000 and on-demand on the A800: 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
A4000 vs A800 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A4000 | A800 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 16 GB GDDR6 | 40 GB HBM2 | 2.5x more on A800 |
| Bandwidth | 448 GB/s | 1,555 GB/s | 3.5x higher on A800 |
| Bus width | 256-bit | 5,120-bit | 20x wider on A800 |
| Compute · dense | |||
| FP16 / BF16 | 76.7 TFLOPS | 312 TFLOPS | 4.1x higher on A800 |
| INT8 | 623.5 TOPS | ||
| FP32 | 19.2 TFLOPS | 19.5 TFLOPS | Within 5% |
| FP64 | Not supported | 9.7 TFLOPS | A800 only |
| Platform | |||
| Interconnect | PCIe only | NVLink 400 GB/s | NVLink on A800 only |
| Board power | 140 W | 240 W | 1.7x higher on A800 |
| Released | Q2 2021 | Q4 2022 | 2 years newer on A800 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32FP64INT4INT8 | |
Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A4000 datasheet and A800 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A4000 /GPU/hr | A800 /GPU/hr |
|---|---|---|
Runcrate
|
On-Demand $0.17 | -- |
Sesterce
|
On-Demand $0.88 | -- |
Vast.ai
|
Spot $0.10 | -- |
Hyperstack
|
On-Demand $0.15 | -- |
TensorDock
|
On-Demand $0.21 | -- |
Paperspace
|
On-Demand $0.76 | -- |
Database Mart
|
Reserved (24mo) $0.17 | -- |
Cirrascale
|
Reserved (12mo) $0.28 | -- |
GPUhub
|
-- | On-Demand $1.18 |
12 providers list the A4000 and 1 the A800. Every listing is on the A4000 page and the A800 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia A4000 was last updated on Sept. 7, 2026, and for Nvidia A800 on Sept. 7, 2026.