The RTX 5090 has 33% more memory and 4.1x the bandwidth, and is cheaper right now.
At a glance
Memory
33% more on RTX 5090
- RTX 5090
- 32 GB
- RTX PRO 4000 SFF
- 24 GB
Memory bandwidth
4.1x higher on RTX 5090
- RTX 5090
- 1,792 GB/s
- RTX PRO 4000 SFF
- 432 GB/s
Cheapest on-demand
8% cheaper on RTX 5090
- RTX 5090
- $0.35 /GPU/hr
- RTX PRO 4000 SFF
- $0.38 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | RTX 5090 | RTX PRO 4000 SFF | Difference |
|---|---|---|---|
| On-demand | $0.72 /GPU/hr from $0.35 | -- /GPU/hr from $0.38 | |
| Reserved | $0.62 /GPU/hr from $0.21 (1 mo) | No reserved listings | |
| Spot | $0.62 /GPU/hr from $0.25 | 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 on-demand on the RTX PRO 4000 SFF: 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
RTX 5090 vs RTX PRO 4000 SFF price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 5090 | RTX PRO 4000 SFF | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 32 GB GDDR7 | 24 GB GDDR7 | 33% more on RTX 5090 |
| Bandwidth | 1,792 GB/s | 432 GB/s | 4.1x higher on RTX 5090 |
| Bus width | 512-bit | 192-bit | 2.7x wider on RTX 5090 |
| Compute · dense | |||
| FP4 | 1,676 TFLOPS | 385 TFLOPS | 4.4x higher on RTX 5090 |
| FP8 | 419 TFLOPS | 192.5 TFLOPS | 2.2x higher on RTX 5090 |
| FP16 / BF16 | 209.5 TFLOPS | 96.3 TFLOPS | 2.2x higher on RTX 5090 |
| INT8 | 838 TOPS | 192.5 TOPS | 4.4x higher on RTX 5090 |
| FP32 | 104.8 TFLOPS | 24 TFLOPS | 4.4x higher on RTX 5090 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 575 W | 70 W | 8.2x higher on RTX 5090 |
| Released | Q1 2025 | Q3 2025 | 6 months newer on RTX PRO 4000 SFF |
| Precision | |||
| Supported formats | FP4FP8FP16BF16TF32FP32INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 5090 datasheet and RTX PRO 4000 SFF datasheet.
Where to rent them
Providers that rent these GPUs, with each one's cheapest rate. No provider we track offers both right now.
| Provider | RTX 5090 | RTX PRO 4000 SFF |
|---|---|---|
HyperAI
Hong Kong
|
$0.35 /GPU/hr On-demand | -- |
Lium
Distributed
|
$0.50 /GPU/hr On-demand | -- |
GPU.ai
UAE
|
$0.52 /GPU/hr On-demand | -- |
Salad
Distributed
|
$0.60 /GPU/hr On-demand | -- |
Theta EdgeCloud
Distributed
|
$0.68 /GPU/hr On-demand | -- |
Sesterce
France
|
$0.71 /GPU/hr On-demand | -- |
Nova Cloud
Canada
|
$0.72 /GPU/hr On-demand PCIe | -- |
Runcrate
USA
|
$0.72 /GPU/hr On-demand PCIe | -- |
Hetzner
Germany
|
-- | $0.38 /GPU/hr On-demand |
22 providers list the RTX 5090 and 1 the RTX PRO 4000 SFF. Every listing is on the RTX 5090 page and the RTX PRO 4000 SFF page.
Similar GPUs
Other comparisons in the same class.
-
-
Nvidia RTX 4090
24 GB GDDR6X · 44% less bandwidth than the RTX 5090
On-demand from $0.26 /GPU/hr -
-
Nvidia RTX PRO 4000
24 GB GDDR7 · 56% more bandwidth than the RTX PRO 4000 SFF
On-demand from $0.27 /GPU/hr
Our data for Nvidia RTX 5090 was last updated on Sept. 14, 2026, and for Nvidia RTX PRO 4000 SFF on Nov. 9, 2026.