Two Nvidia consumer GPUs, one Ada Lovelace and one Blackwell: 24 GB GDDR6X against 16 GB GDDR7 and 5% more memory bandwidth on the RTX 4090.
At a glance
Memory
- RTX 4090
- 24 GB
- RTX 5080
- 16 GB
1.5x more on RTX 4090
Memory bandwidth
- RTX 4090
- 1,010 GB/s
- RTX 5080
- 960 GB/s
1.1x higher on RTX 4090
Cheapest on-demand
- RTX 4090
- $0.23 /GPU/hr
- RTX 5080
- $0.26 /GPU/hr
10% cheaper on RTX 4090
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | RTX 4090 | RTX 5080 | Difference |
|---|---|---|---|
| On-demand | $0.44 /GPU/hr from $0.23 | $1.05 /GPU/hr from $0.26 | 2.4x more expensive on RTX 5080 |
| Reserved | $0.52 /GPU/hr from $0.30 (24 mo) | -- /GPU/hr from $1.60 (1 mo) | |
| Spot | -- /GPU/hr from $0.16 | -- /GPU/hr from $0.15 |
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 4090 and reserved and spot on the RTX 5080: 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 4090 vs RTX 5080 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 4090 | RTX 5080 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR6X | 16 GB GDDR7 | 1.5x more on RTX 4090 |
| Bandwidth | 1,010 GB/s | 960 GB/s | 1.1x higher on RTX 4090 |
| Bus width | 384-bit | 256-bit | 1.5x wider on RTX 4090 |
| Compute · dense | |||
| FP4 | Not supported | 900.4 TFLOPS | RTX 5080 only |
| FP8 | 330.3 TFLOPS | 225.1 TFLOPS | 1.5x higher on RTX 4090 |
| FP16 / BF16 | 165.2 TFLOPS | 112.6 TFLOPS | 1.5x higher on RTX 4090 |
| INT8 | 660.6 TOPS | 450.2 TOPS | 1.5x higher on RTX 4090 |
| FP32 | 82.6 TFLOPS | 56.3 TFLOPS | 1.5x higher on RTX 4090 |
| Platform | |||
| Interconnect | PCIe only | PCIe only | |
| Board power | 450 W | 360 W | 1.2x higher on RTX 4090 |
| Released | Q4 2022 | Q1 2025 | 2 years newer on RTX 5080 |
| Precision | |||
| Supported formats | FP8FP16BF16TF32FP32INT4INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 4090 datasheet and RTX 5080 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | RTX 4090 /GPU/hr | RTX 5080 /GPU/hr |
|---|---|---|
Salad
|
On-Demand $0.33 | On-Demand $0.26 |
Vast.ai
|
On-Demand $2.00 | On-Demand $0.27 |
GPU.ai
|
On-Demand $0.34 | On-Demand $2.01 |
Runpod
|
On-Demand $0.34 | On-Demand $0.39 |
Theta EdgeCloud
|
On-Demand $0.23 | -- |
Packet·ai
|
On-Demand $0.39 | -- |
BreezeHost
|
On-Demand $0.35 | -- |
GPUhub
|
On-Demand $0.44 | -- |
SwissGPU
|
-- | On-Demand $0.31 |
17 providers list the RTX 4090 and 5 the RTX 5080. Every listing is on the RTX 4090 page and the RTX 5080 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia RTX 5090
From $0.35 /GPU/hr Compare -
Nvidia L40S
From $0.38 /GPU/hr Compare -
Nvidia RTX 3090
From $0.10 /GPU/hr Compare -
Nvidia RTX 5070 Ti
From $0.23 /GPU/hr Compare
Our data for Nvidia RTX 4090 was last updated on Sept. 7, 2026, and for Nvidia RTX 5080 on Sept. 7, 2026.