The RTX 5090 is newer, with 33% more memory and 5.2x the bandwidth. The P40 is cheaper right now.
At a glance
Memory
33% more on RTX 5090
- P40
- 24 GB
- RTX 5090
- 32 GB
Memory bandwidth
5.2x higher on RTX 5090
- P40
- 346 GB/s
- RTX 5090
- 1,792 GB/s
Cheapest on-demand
70% cheaper on P40
- P40
- $0.11 /GPU/hr
- RTX 5090
- $0.35 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | P40 | RTX 5090 | Difference |
|---|---|---|---|
| On-demand | -- /GPU/hr from $0.11 | $0.65 /GPU/hr from $0.35 | |
| Reserved | No reserved listings | $0.59 /GPU/hr from $0.21 (1 mo) | |
| Spot | -- /GPU/hr from $0.07 | $0.43 /GPU/hr from $0.25 |
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 on-demand and spot on the P40: 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
P40 vs RTX 5090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | P40 | RTX 5090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR5 | 32 GB GDDR7 | 33% more on RTX 5090 |
| Bandwidth | 346 GB/s | 1,792 GB/s | 5.2x higher on RTX 5090 |
| Bus width | 384-bit | 512-bit | 33% wider on RTX 5090 |
| Compute · dense | |||
| FP4 | Not supported | 1,676 TFLOPS | RTX 5090 only |
| FP8 | Not supported | 419 TFLOPS | RTX 5090 only |
| FP16 / BF16 | 209.5 TFLOPS | ||
| INT8 | 47 TOPS | 838 TOPS | 18x higher on RTX 5090 |
| FP32 | 12 TFLOPS | 104.8 TFLOPS | 8.7x higher on RTX 5090 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 250 W | 575 W | 2.3x higher on RTX 5090 |
| Released | Q3 2016 | Q1 2025 | 8 years newer on RTX 5090 |
| Precision | |||
| Supported formats | FP16FP32INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: P40 datasheet and RTX 5090 datasheet.
Where to rent them
Providers that rent these GPUs, with each one's cheapest rate. Providers that rent both are listed first.
| Provider | P40 | RTX 5090 |
|---|---|---|
Vast.ai
USA
|
$0.11 /GPU/hr On-demand | $0.40 /GPU/hr On-demand |
HyperAI
Hong Kong
|
-- | $0.35 /GPU/hr On-demand |
GPU.ai
UAE
|
-- | $0.41 /GPU/hr On-demand |
GPUhub
Singapore
|
-- | $0.46 /GPU/hr On-demand |
Salad
Distributed
|
-- | $0.50 /GPU/hr On-demand |
Nova Cloud
Canada
|
-- | $0.66 /GPU/hr On-demand PCIe |
Runpod
USA
|
-- | $0.69 /GPU/hr On-demand |
Lium
Distributed
|
-- | $0.75 /GPU/hr On-demand |
1 provider list the P40 and 21 the RTX 5090. Every listing is on the P40 page and the RTX 5090 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia L4
24 GB GDDR6 · 13% less bandwidth than the P40
From $0.32 /GPU/hr Compare -
Nvidia T4
16 GB GDDR6 · 13% less bandwidth than the P40
From $0.15 /GPU/hr Compare
-
Nvidia A10
24 GB GDDR6 · 73% more bandwidth than the P40
From $1.00 /GPU/hr Compare
-
Nvidia RTX PRO 6000
96 GB GDDR7 · workstation card
From $0.65 /GPU/hr Compare
-
Nvidia RTX 4090
24 GB GDDR6X · 44% less bandwidth than the RTX 5090
From $0.33 /GPU/hr Compare
Our data for Nvidia P40 was last updated on Sept. 14, 2026, and for Nvidia RTX 5090 on Sept. 14, 2026.