A data center GPU against a workstation card on the Pascal architecture: 24 GB GDDR5 against 8 GB GDDR5 and 42% more memory bandwidth on the P40.
At a glance
Updated 6 minutes ago- Memory
- 24 GB
- GDDR5
- Memory bandwidth
- 346 GB/s
- From
- $0.11 /GPU/hr
- 1 provider
- Memory
- 8 GB
- GDDR5
- Memory bandwidth
- 243 GB/s
- From
- $0.07 /GPU/hr
- 2 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | P40 | P4000 | Difference |
|---|---|---|---|
| On-demand | $0.11 | $0.07 | 37% cheaper on P4000 |
| Spot | $0.07 | $0.04 | 39% cheaper on P4000 |
| Reserved | No reserved listings | $0.06 | |
| Median | $0.11 | $0.29 | 2.7x more expensive on P4000 |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
P40 vs P4000 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | P40 | P4000 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR5 | 8 GB GDDR5 | 3x more on P40 |
| Bandwidth | 346 GB/s | 243 GB/s | 1.4x higher on P40 |
| Bus width | 384-bit | 256-bit | 1.5x wider on P40 |
| Compute · dense | |||
| INT8 | 47 TOPS | Not supported | P40 only |
| FP32 | 12 TFLOPS | 5.3 TFLOPS | 2.3x higher on P40 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 250 W | 105 W | 2.4x higher on P40 |
| Released | Q3 2016 | Q1 2017 | |
| Precision | |||
| Supported formats | FP16FP32INT8 | FP16FP32 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: P40 datasheet and P4000 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | P40 /GPU/hr | P4000 /GPU/hr |
|---|---|---|
Vast.ai
|
On-Demand $0.11 | On-Demand $0.07 |
Paperspace
|
-- | On-Demand $0.51 |
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia P40 was last updated on Sept. 7, 2026, and for Nvidia P4000 on Sept. 7, 2026.