Two Nvidia consumer GPUs, one Ampere and one Ada Lovelace: 24 GB GDDR6X against 16 GB GDDR6X, 31% more memory bandwidth on the RTX 3090, and NVLink 112 GB/s on the RTX 3090 where the RTX 4080 is PCIe only.
At a glance
Updated 6 minutes ago- Memory
- 24 GB
- GDDR6X
- Memory bandwidth
- 936 GB/s
- From
- $0.10 /GPU/hr
- 10 providers
- Memory
- 16 GB
- GDDR6X
- Memory bandwidth
- 717 GB/s
- From
- $0.14 /GPU/hr
- 6 providers
Price comparison
Cheapest listed price per GPU per hour by billing type, and the median across providers.
| Billing | RTX 3090 | RTX 4080 | Difference |
|---|---|---|---|
| On-demand | $0.10 | $0.14 | 27% cheaper on RTX 3090 |
| Spot | $0.07 | $0.11 | 38% cheaper on RTX 3090 |
| Reserved | $0.06 | No reserved listings | |
| Median | $0.25 | $0.25 | Identical |
Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.
Price history
RTX 3090 vs RTX 4080 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | RTX 3090 | RTX 4080 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 24 GB GDDR6X | 16 GB GDDR6X | 1.5x more on RTX 3090 |
| Bandwidth | 936 GB/s | 717 GB/s | 1.3x higher on RTX 3090 |
| Bus width | 384-bit | 256-bit | 1.5x wider on RTX 3090 |
| Compute · dense | |||
| FP8 | Not supported | 194.9 TFLOPS | RTX 4080 only |
| FP16 / BF16 | 71.2 TFLOPS | 97.5 TFLOPS | 1.4x higher on RTX 4080 |
| INT8 | 284.7 TOPS | 389.9 TOPS | 1.4x higher on RTX 4080 |
| FP32 | 35.6 TFLOPS | 48.7 TFLOPS | 1.4x higher on RTX 4080 |
| Platform | |||
| Interconnect | NVLink 112 GB/s | PCIe only | NVLink on RTX 3090 only |
| Board power | 350 W | 320 W | 1.1x higher on RTX 3090 |
| Released | Q3 2020 | Q4 2022 | 2 years newer on RTX 4080 |
| Precision | |||
| Supported formats | FP16BF16TF32FP32INT4INT8 | FP8FP16BF16TF32FP32INT4INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: RTX 3090 datasheet and RTX 4080 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 4080 /GPU/hr |
|---|---|---|
Vast.ai
|
On-Demand $0.12 | On-Demand $0.14 |
Theta EdgeCloud
|
On-Demand $0.13 | On-Demand $0.26 |
Salad
|
On-Demand $0.25 | On-Demand $0.28 |
SwissGPU
|
On-Demand $0.31 | On-Demand $0.25 |
HyperAI
|
On-Demand $0.10 | -- |
Lium
|
On-Demand $0.13 | -- |
GPU.ai
|
On-Demand $0.21 | -- |
EmpirioLabs AI
|
On-Demand $0.45 | -- |
GPUhub
|
-- | On-Demand $0.25 |
Runpod
|
-- | On-Demand $0.27 |
10 providers list the RTX 3090 and 6 the RTX 4080. Every listing is on the RTX 3090 page and the RTX 4080 page.
Similar GPUs
Other comparisons in the same class.
Our data for Nvidia RTX 3090 was last updated on Sept. 7, 2026, and for Nvidia RTX 4080 on Sept. 7, 2026.