Two Nvidia consumer GPUs, one Ampere and one Blackwell: 24 GB GDDR6X against 16 GB GDDR7, memory bandwidth within 5%, and NVLink 112 GB/s on the RTX 3090 where the RTX 5080 is PCIe only.
At a glance
Memory
- RTX 3090
- 24 GB
- RTX 5080
- 16 GB
1.5x more on RTX 3090
Memory bandwidth
- RTX 3090
- 936 GB/s
- RTX 5080
- 960 GB/s
Within 5%
Cheapest on-demand
- RTX 3090
- $0.10 /GPU/hr
- RTX 5080
- $0.18 /GPU/hr
44% cheaper on RTX 3090
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | RTX 3090 | RTX 5080 | Difference |
|---|---|---|---|
| On-demand | $0.24 /GPU/hr from $0.10 | $0.31 /GPU/hr from $0.18 | 23% cheaper on RTX 3090 |
| Reserved | -- /GPU/hr from $0.06 (1 mo) | No reserved listings | |
| Spot | $0.11 /GPU/hr from $0.06 | -- /GPU/hr from $0.14 |
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 reserved on the RTX 3090 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 3090 vs RTX 5080 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 3090 | RTX 5080 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR6X | 16 GB GDDR7 | 1.5x more on RTX 3090 |
| Bandwidth | 936 GB/s | 960 GB/s | Within 5% |
| Bus width | 384-bit | 256-bit | 1.5x wider on RTX 3090 |
| Compute · dense | |||
| FP4 | Not supported | 900.4 TFLOPS | RTX 5080 only |
| FP8 | Not supported | 225.1 TFLOPS | RTX 5080 only |
| FP16 / BF16 | 71.2 TFLOPS | 112.6 TFLOPS | 1.6x higher on RTX 5080 |
| INT8 | 284.7 TOPS | 450.2 TOPS | 1.6x higher on RTX 5080 |
| FP32 | 35.6 TFLOPS | 56.3 TFLOPS | 1.6x higher on RTX 5080 |
| Platform | |||
| Interconnect | NVLink 112 GB/s | PCIe only | NVLink on RTX 3090 only |
| Board power | 350 W | 360 W | Within 5% |
| Released | Q3 2020 | Q1 2025 | 4 years newer on RTX 5080 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP4FP8FP16BF16TF32FP32INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 3090 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 3090 /GPU/hr | RTX 5080 /GPU/hr |
|---|---|---|
Vast.ai
|
On-Demand $0.11 | On-Demand $0.18 |
Salad
|
On-Demand $0.17 | On-Demand $0.26 |
Theta EdgeCloud
|
On-Demand $0.13 | On-Demand $0.36 |
Runpod
|
On-Demand $0.22 | On-Demand $0.39 |
SwissGPU
|
On-Demand $0.31 | On-Demand $0.31 |
HyperAI
|
On-Demand $0.10 | -- |
GPU.ai
|
On-Demand $0.15 | -- |
Lium
|
On-Demand $0.16 | -- |
10 providers list the RTX 3090 and 5 the RTX 5080. Every listing is on the RTX 3090 page and the RTX 5080 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia RTX 5090
32 GB GDDR7 · 91% more bandwidth than the RTX 3090
From $0.35 /GPU/hr Compare -
Nvidia RTX 4090
24 GB GDDR6X · PCIe only
From $0.23 /GPU/hr Compare
-
Nvidia RTX 3090 Ti
24 GB GDDR6X · PCIe only
From $0.19 /GPU/hr Compare
-
Nvidia RTX 5070 Ti
16 GB GDDR7 · released Q1 2025
From $0.14 /GPU/hr Compare
Our data for Nvidia RTX 3090 was last updated on Sept. 14, 2026, and for Nvidia RTX 5080 on Sept. 14, 2026.