Two Nvidia workstation GPUs on the Ampere architecture: 16 GB GDDR6 against 48 GB GDDR6, 71% more memory bandwidth on the RTX A6000, and NVLink 112 GB/s on the RTX A6000 where the A4000 is PCIe only.
At a glance
Memory
- A4000
- 16 GB
- RTX A6000
- 48 GB
3x more on RTX A6000
Memory bandwidth
- A4000
- 448 GB/s
- RTX A6000
- 768 GB/s
1.7x higher on RTX A6000
Cheapest on-demand
- A4000
- $0.15 /GPU/hr
- RTX A6000
- $0.35 /GPU/hr
2.3x more expensive on RTX A6000
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | A4000 | RTX A6000 | Difference |
|---|---|---|---|
| On-demand | $0.21 /GPU/hr from $0.15 | $0.56 /GPU/hr from $0.35 | 2.7x more expensive on RTX A6000 |
| Reserved | $0.25 /GPU/hr from $0.11 | $0.54 /GPU/hr from $0.35 | 2.2x more expensive on RTX A6000 |
| Spot | -- /GPU/hr from $0.10 | -- /GPU/hr from $0.40 |
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 spot on the RTX A6000: 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 RTX A6000 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A4000 | RTX A6000 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 16 GB GDDR6 | 48 GB GDDR6 | 3x more on RTX A6000 |
| Bandwidth | 448 GB/s | 768 GB/s | 1.7x higher on RTX A6000 |
| Bus width | 256-bit | 384-bit | 1.5x wider on RTX A6000 |
| Compute · dense | |||
| FP16 / BF16 | 76.7 TFLOPS | 154.8 TFLOPS | 2x higher on RTX A6000 |
| INT8 | 309.7 TOPS | ||
| FP32 | 19.2 TFLOPS | 38.7 TFLOPS | 2x higher on RTX A6000 |
| Platform | |||
| Interconnect | PCIe only | NVLink 112 GB/s | NVLink on RTX A6000 only |
| Board power | 140 W | 300 W | 2.1x higher on RTX A6000 |
| Released | Q2 2021 | Q4 2020 | 6 months newer on A4000 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A4000 datasheet and RTX A6000 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | A4000 /GPU/hr | RTX A6000 /GPU/hr |
|---|---|---|
Runcrate
|
On-Demand $0.17 | On-Demand $0.55 |
Sesterce
|
On-Demand $0.88 | On-Demand $0.55 |
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 | -- |
Thunder Compute
|
-- | On-Demand $0.35 |
GPU.ai
|
-- | On-Demand $0.45 |
Massed Compute
|
-- | On-Demand $0.55 |
Oblivus
|
-- | On-Demand $0.55 |
Koyeb
|
-- | On-Demand $0.75 |
EmpirioLabs AI
|
-- | On-Demand $0.90 |
12 providers list the A4000 and 18 the RTX A6000. Every listing is on the A4000 page and the RTX A6000 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia A4000 was last updated on Sept. 7, 2026, and for Nvidia RTX A6000 on Sept. 7, 2026.