The RTX 5090 is newer, with 60% more memory and 2.8x the bandwidth.
At a glance
Memory
60% more on RTX 5090
- A4500
- 20 GB
- RTX 5090
- 32 GB
Memory bandwidth
2.8x higher on RTX 5090
- A4500
- 640 GB/s
- RTX 5090
- 1,792 GB/s
Cheapest on-demand
- A4500
- --
- RTX 5090
- $0.35 /GPU/hr
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | A4500 | RTX 5090 | Difference |
|---|---|---|---|
| On-demand | Sold out | $0.65 /GPU/hr from $0.35 | |
| Reserved | On request | $0.59 /GPU/hr from $0.21 (1 mo) | |
| Spot | No spot listings | $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. Verify before provisioning. More on how we price.
Price history
A4500 vs RTX 5090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | A4500 | RTX 5090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 20 GB GDDR6 | 32 GB GDDR7 | 60% more on RTX 5090 |
| Bandwidth | 640 GB/s | 1,792 GB/s | 2.8x higher on RTX 5090 |
| Bus width | 320-bit | 512-bit | 60% 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 | 94.6 TFLOPS | 209.5 TFLOPS | 2.2x higher on RTX 5090 |
| INT8 | 838 TOPS | ||
| FP32 | 23.7 TFLOPS | 104.8 TFLOPS | 4.4x higher on RTX 5090 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 200 W | 575 W | 2.9x higher on RTX 5090 |
| Released | Q4 2021 | Q1 2025 | 3 years newer on RTX 5090 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: A4500 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 | A4500 | RTX 5090 |
|---|---|---|
Runpod
USA
|
On request Reserved | $0.69 /GPU/hr On-demand |
GPU.ai
UAE
|
$0.19 /GPU/hr On-demand | $0.41 /GPU/hr On-demand |
HyperAI
Hong Kong
|
-- | $0.35 /GPU/hr On-demand |
Vast.ai
USA
|
-- | $0.40 /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 |
Lium
Distributed
|
-- | $0.75 /GPU/hr On-demand |
2 providers list the A4500 and 21 the RTX 5090. Every listing is on the A4500 page and the RTX 5090 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia L4
24 GB GDDR6 · 53% less bandwidth than the A4500
From $0.32 /GPU/hr Compare -
Nvidia A10
24 GB GDDR6 · data center card
From $1.00 /GPU/hr Compare
-
Nvidia A5000
24 GB GDDR6 · 20% more bandwidth than the A4500
From $0.21 /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 A4500 was last updated on Sept. 14, 2026, and for Nvidia RTX 5090 on Sept. 14, 2026.