Two Nvidia consumer GPUs, one Ada Lovelace and one Blackwell: 16 GB GDDR6X against 32 GB GDDR7 and 2.5x the memory bandwidth on the RTX 5090.
At a glance
Memory
- RTX 4080
- 16 GB
- RTX 5090
- 32 GB
2x more on RTX 5090
Memory bandwidth
- RTX 4080
- 717 GB/s
- RTX 5090
- 1,792 GB/s
2.5x higher on RTX 5090
Cheapest on-demand
- RTX 4080
- $0.25 /GPU/hr
- RTX 5090
- $0.35 /GPU/hr
29% cheaper on RTX 4080
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | RTX 4080 | RTX 5090 | Difference |
|---|---|---|---|
| On-demand | $0.27 /GPU/hr from $0.25 | $0.72 /GPU/hr from $0.35 | 2.7x more expensive on RTX 5090 |
| Reserved | No reserved listings | $0.60 /GPU/hr from $0.21 (1 mo) | |
| Spot | -- /GPU/hr from $0.11 | -- /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 spot on the RTX 4080 and spot on the RTX 5090: 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 4080 vs RTX 5090 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 4080 | RTX 5090 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 16 GB GDDR6X | 32 GB GDDR7 | 2x more on RTX 5090 |
| Bandwidth | 717 GB/s | 1,792 GB/s | 2.5x higher on RTX 5090 |
| Bus width | 256-bit | 512-bit | 2x wider on RTX 5090 |
| Compute · dense | |||
| FP4 | Not supported | 1,676 TFLOPS | RTX 5090 only |
| FP8 | 194.9 TFLOPS | 419 TFLOPS | 2.1x higher on RTX 5090 |
| FP16 / BF16 | 97.5 TFLOPS | 209.5 TFLOPS | 2.1x higher on RTX 5090 |
| INT8 | 389.9 TOPS | 838 TOPS | 2.1x higher on RTX 5090 |
| FP32 | 48.7 TFLOPS | 104.8 TFLOPS | 2.2x higher on RTX 5090 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 320 W | 575 W | 1.8x higher on RTX 5090 |
| Released | Q4 2022 | Q1 2025 | 2 years newer on RTX 5090 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32INT4INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 4080 datasheet and RTX 5090 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | RTX 4080 /GPU/hr | RTX 5090 /GPU/hr |
|---|---|---|
GPUhub
|
On-Demand $0.25 | On-Demand $0.46 |
Theta EdgeCloud
|
On-Demand $0.26 | On-Demand $0.82 |
Salad
|
On-Demand $0.28 | On-Demand $0.45 |
SwissGPU
|
On-Demand $0.25 | On-Demand $0.61 |
Runpod
|
On-Demand $0.27 | -- |
HyperAI
|
-- | On-Demand $0.35 |
Vast.ai
|
-- | On-Demand $13.60 |
GPU.ai
|
-- | On-Demand $53.34 |
GPUaaS
|
-- | Custom on request |
5 providers list the RTX 4080 and 18 the RTX 5090. Every listing is on the RTX 4080 page and the RTX 5090 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia RTX 4080 was last updated on Sept. 7, 2026, and for Nvidia RTX 5090 on Sept. 7, 2026.